1FFMPEG-ALL(1) FFMPEG-ALL(1)
2
6 ffmpeg - ffmpeg video converter
7
9 ffmpeg [global_options] {[input_file_options] -i input_url} ...
10 {[output_file_options] output_url} ...
11
13 ffmpeg is a very fast video and audio converter that can also grab from
14 a live audio/video source. It can also convert between arbitrary sample
15 rates and resize video on the fly with a high quality polyphase filter.
16 ffmpeg reads from an arbitrary number of input "files" (which can be
18 regular files, pipes, network streams, grabbing devices, etc.),
19 specified by the "-i" option, and writes to an arbitrary number of
20 output "files", which are specified by a plain output url. Anything
21 found on the command line which cannot be interpreted as an option is
22 considered to be an output url.
23 Each input or output url can, in principle, contain any number of
25 streams of different types (video/audio/subtitle/attachment/data). The
26 allowed number and/or types of streams may be limited by the container
27 format. Selecting which streams from which inputs will go into which
28 output is either done automatically or with the "-map" option (see the
29 Stream selection chapter).
30 To refer to input files in options, you must use their indices
32 (0-based). E.g. the first input file is 0, the second is 1, etc.
33 Similarly, streams within a file are referred to by their indices. E.g.
34 "2:3" refers to the fourth stream in the third input file. Also see the
35 Stream specifiers chapter.
36 As a general rule, options are applied to the next specified file.
38 Therefore, order is important, and you can have the same option on the
39 command line multiple times. Each occurrence is then applied to the
40 next input or output file. Exceptions from this rule are the global
41 options (e.g. verbosity level), which should be specified first.
42 Do not mix input and output files -- first specify all input files,
44 then all output files. Also do not mix options which belong to
45 different files. All options apply ONLY to the next input or output
46 file and are reset between files.
47 • To set the video bitrate of the output file to 64 kbit/s:
49 ffmpeg -i input.avi -b:v 64k -bufsize 64k output.avi
51 • To force the frame rate of the output file to 24 fps:
53 ffmpeg -i input.avi -r 24 output.avi
55 • To force the frame rate of the input file (valid for raw formats
57 only) to 1 fps and the frame rate of the output file to 24 fps:
58 ffmpeg -r 1 -i input.m2v -r 24 output.avi
60 The format option may be needed for raw input files.
62
64 The transcoding process in ffmpeg for each output can be described by
65 the following diagram:
66 _______ ______________
68 | | | |
69 | input | demuxer | encoded data | decoder
70 | file | ---------> | packets | -----+
71 |_______| |______________| |
72 v
73 _________
74 | |
75 | decoded |
76 | frames |
77 |_________|
78 ________ ______________ |
79 | | | | |
80 | output | <-------- | encoded data | <----+
81 | file | muxer | packets | encoder
82 |________| |______________|
83 ffmpeg calls the libavformat library (containing demuxers) to read
85 input files and get packets containing encoded data from them. When
86 there are multiple input files, ffmpeg tries to keep them synchronized
87 by tracking lowest timestamp on any active input stream.
88 Encoded packets are then passed to the decoder (unless streamcopy is
90 selected for the stream, see further for a description). The decoder
91 produces uncompressed frames (raw video/PCM audio/...) which can be
92 processed further by filtering (see next section). After filtering, the
93 frames are passed to the encoder, which encodes them and outputs
94 encoded packets. Finally those are passed to the muxer, which writes
95 the encoded packets to the output file.
96 Filtering
98 Before encoding, ffmpeg can process raw audio and video frames using
99 filters from the libavfilter library. Several chained filters form a
100 filter graph. ffmpeg distinguishes between two types of filtergraphs:
101 simple and complex.
102 Simple filtergraphs
104 Simple filtergraphs are those that have exactly one input and output,
106 both of the same type. In the above diagram they can be represented by
107 simply inserting an additional step between decoding and encoding:
108 _________ ______________
110 | | | |
111 | decoded | | encoded data |
112 | frames |\ _ | packets |
113 |_________| \ /||______________|
114 \ __________ /
115 simple _\|| | / encoder
116 filtergraph | filtered |/
117 | frames |
118 |__________|
119 Simple filtergraphs are configured with the per-stream -filter option
121 (with -vf and -af aliases for video and audio respectively). A simple
122 filtergraph for video can look for example like this:
123 _______ _____________ _______ ________
125 | | | | | | | |
126 | input | ---> | deinterlace | ---> | scale | ---> | output |
127 |_______| |_____________| |_______| |________|
128 Note that some filters change frame properties but not frame contents.
130 E.g. the "fps" filter in the example above changes number of frames,
131 but does not touch the frame contents. Another example is the "setpts"
132 filter, which only sets timestamps and otherwise passes the frames
133 unchanged.
134 Complex filtergraphs
136 Complex filtergraphs are those which cannot be described as simply a
138 linear processing chain applied to one stream. This is the case, for
139 example, when the graph has more than one input and/or output, or when
140 output stream type is different from input. They can be represented
141 with the following diagram:
142 _________
144 | |
145 | input 0 |\ __________
146 |_________| \ | |
147 \ _________ /| output 0 |
148 \ | | / |__________|
149 _________ \| complex | /
150 | | | |/
151 | input 1 |---->| filter |\
152 |_________| | | \ __________
153 /| graph | \ | |
154 / | | \| output 1 |
155 _________ / |_________| |__________|
156 | | /
157 | input 2 |/
158 |_________|
159 Complex filtergraphs are configured with the -filter_complex option.
161 Note that this option is global, since a complex filtergraph, by its
162 nature, cannot be unambiguously associated with a single stream or
163 file.
164 The -lavfi option is equivalent to -filter_complex.
166 A trivial example of a complex filtergraph is the "overlay" filter,
168 which has two video inputs and one video output, containing one video
169 overlaid on top of the other. Its audio counterpart is the "amix"
170 filter.
171 Stream copy
173 Stream copy is a mode selected by supplying the "copy" parameter to the
174 -codec option. It makes ffmpeg omit the decoding and encoding step for
175 the specified stream, so it does only demuxing and muxing. It is useful
176 for changing the container format or modifying container-level
177 metadata. The diagram above will, in this case, simplify to this:
178 _______ ______________ ________
180 | | | | | |
181 | input | demuxer | encoded data | muxer | output |
182 | file | ---------> | packets | -------> | file |
183 |_______| |______________| |________|
184 Since there is no decoding or encoding, it is very fast and there is no
186 quality loss. However, it might not work in some cases because of many
187 factors. Applying filters is obviously also impossible, since filters
188 work on uncompressed data.
189
191 ffmpeg provides the "-map" option for manual control of stream
192 selection in each output file. Users can skip "-map" and let ffmpeg
193 perform automatic stream selection as described below. The "-vn / -an /
194 -sn / -dn" options can be used to skip inclusion of video, audio,
195 subtitle and data streams respectively, whether manually mapped or
196 automatically selected, except for those streams which are outputs of
197 complex filtergraphs.
198 Description
200 The sub-sections that follow describe the various rules that are
201 involved in stream selection. The examples that follow next show how
202 these rules are applied in practice.
203 While every effort is made to accurately reflect the behavior of the
205 program, FFmpeg is under continuous development and the code may have
206 changed since the time of this writing.
207 Automatic stream selection
209 In the absence of any map options for a particular output file, ffmpeg
211 inspects the output format to check which type of streams can be
212 included in it, viz. video, audio and/or subtitles. For each acceptable
213 stream type, ffmpeg will pick one stream, when available, from among
214 all the inputs.
215 It will select that stream based upon the following criteria:
217 • for video, it is the stream with the highest resolution,
219 • for audio, it is the stream with the most channels,
221 • for subtitles, it is the first subtitle stream found but there's a
223 caveat. The output format's default subtitle encoder can be either
224 text-based or image-based, and only a subtitle stream of the same
225 type will be chosen.
226 In the case where several streams of the same type rate equally, the
228 stream with the lowest index is chosen.
229 Data or attachment streams are not automatically selected and can only
231 be included using "-map".
232 Manual stream selection
234 When "-map" is used, only user-mapped streams are included in that
236 output file, with one possible exception for filtergraph outputs
237 described below.
238 Complex filtergraphs
240 If there are any complex filtergraph output streams with unlabeled
242 pads, they will be added to the first output file. This will lead to a
243 fatal error if the stream type is not supported by the output format.
244 In the absence of the map option, the inclusion of these streams leads
245 to the automatic stream selection of their types being skipped. If map
246 options are present, these filtergraph streams are included in addition
247 to the mapped streams.
248 Complex filtergraph output streams with labeled pads must be mapped
250 once and exactly once.
251 Stream handling
253 Stream handling is independent of stream selection, with an exception
255 for subtitles described below. Stream handling is set via the "-codec"
256 option addressed to streams within a specific output file. In
257 particular, codec options are applied by ffmpeg after the stream
258 selection process and thus do not influence the latter. If no "-codec"
259 option is specified for a stream type, ffmpeg will select the default
260 encoder registered by the output file muxer.
261 An exception exists for subtitles. If a subtitle encoder is specified
263 for an output file, the first subtitle stream found of any type, text
264 or image, will be included. ffmpeg does not validate if the specified
265 encoder can convert the selected stream or if the converted stream is
266 acceptable within the output format. This applies generally as well:
267 when the user sets an encoder manually, the stream selection process
268 cannot check if the encoded stream can be muxed into the output file.
269 If it cannot, ffmpeg will abort and all output files will fail to be
270 processed.
271 Examples
273 The following examples illustrate the behavior, quirks and limitations
274 of ffmpeg's stream selection methods.
275 They assume the following three input files.
277 input file 'A.avi'
279 stream 0: video 640x360
280 stream 1: audio 2 channels
281 input file 'B.mp4'
283 stream 0: video 1920x1080
284 stream 1: audio 2 channels
285 stream 2: subtitles (text)
286 stream 3: audio 5.1 channels
287 stream 4: subtitles (text)
288 input file 'C.mkv'
290 stream 0: video 1280x720
291 stream 1: audio 2 channels
292 stream 2: subtitles (image)
293 Example: automatic stream selection
295 ffmpeg -i A.avi -i B.mp4 out1.mkv out2.wav -map 1:a -c:a copy out3.mov
297 There are three output files specified, and for the first two, no
299 "-map" options are set, so ffmpeg will select streams for these two
300 files automatically.
301 out1.mkv is a Matroska container file and accepts video, audio and
303 subtitle streams, so ffmpeg will try to select one of each type.For
304 video, it will select "stream 0" from B.mp4, which has the highest
305 resolution among all the input video streams.For audio, it will select
306 "stream 3" from B.mp4, since it has the greatest number of channels.For
307 subtitles, it will select "stream 2" from B.mp4, which is the first
308 subtitle stream from among A.avi and B.mp4.
309 out2.wav accepts only audio streams, so only "stream 3" from B.mp4 is
311 selected.
312 For out3.mov, since a "-map" option is set, no automatic stream
314 selection will occur. The "-map 1:a" option will select all audio
315 streams from the second input B.mp4. No other streams will be included
316 in this output file.
317 For the first two outputs, all included streams will be transcoded. The
319 encoders chosen will be the default ones registered by each output
320 format, which may not match the codec of the selected input streams.
321 For the third output, codec option for audio streams has been set to
323 "copy", so no decoding-filtering-encoding operations will occur, or can
324 occur. Packets of selected streams shall be conveyed from the input
325 file and muxed within the output file.
326 Example: automatic subtitles selection
328 ffmpeg -i C.mkv out1.mkv -c:s dvdsub -an out2.mkv
330 Although out1.mkv is a Matroska container file which accepts subtitle
332 streams, only a video and audio stream shall be selected. The subtitle
333 stream of C.mkv is image-based and the default subtitle encoder of the
334 Matroska muxer is text-based, so a transcode operation for the
335 subtitles is expected to fail and hence the stream isn't selected.
336 However, in out2.mkv, a subtitle encoder is specified in the command
337 and so, the subtitle stream is selected, in addition to the video
338 stream. The presence of "-an" disables audio stream selection for
339 out2.mkv.
340 Example: unlabeled filtergraph outputs
342 ffmpeg -i A.avi -i C.mkv -i B.mp4 -filter_complex "overlay" out1.mp4 out2.srt
344 A filtergraph is setup here using the "-filter_complex" option and
346 consists of a single video filter. The "overlay" filter requires
347 exactly two video inputs, but none are specified, so the first two
348 available video streams are used, those of A.avi and C.mkv. The output
349 pad of the filter has no label and so is sent to the first output file
350 out1.mp4. Due to this, automatic selection of the video stream is
351 skipped, which would have selected the stream in B.mp4. The audio
352 stream with most channels viz. "stream 3" in B.mp4, is chosen
353 automatically. No subtitle stream is chosen however, since the MP4
354 format has no default subtitle encoder registered, and the user hasn't
355 specified a subtitle encoder.
356 The 2nd output file, out2.srt, only accepts text-based subtitle
358 streams. So, even though the first subtitle stream available belongs to
359 C.mkv, it is image-based and hence skipped. The selected stream,
360 "stream 2" in B.mp4, is the first text-based subtitle stream.
361 Example: labeled filtergraph outputs
363 ffmpeg -i A.avi -i B.mp4 -i C.mkv -filter_complex "[1:v]hue=s=0[outv];overlay;aresample" \
365 -map '[outv]' -an out1.mp4 \
366 out2.mkv \
367 -map '[outv]' -map 1:a:0 out3.mkv
368 The above command will fail, as the output pad labelled "[outv]" has
370 been mapped twice. None of the output files shall be processed.
371 ffmpeg -i A.avi -i B.mp4 -i C.mkv -filter_complex "[1:v]hue=s=0[outv];overlay;aresample" \
373 -an out1.mp4 \
374 out2.mkv \
375 -map 1:a:0 out3.mkv
376 This command above will also fail as the hue filter output has a label,
378 "[outv]", and hasn't been mapped anywhere.
379 The command should be modified as follows,
381 ffmpeg -i A.avi -i B.mp4 -i C.mkv -filter_complex "[1:v]hue=s=0,split=2[outv1][outv2];overlay;aresample" \
383 -map '[outv1]' -an out1.mp4 \
384 out2.mkv \
385 -map '[outv2]' -map 1:a:0 out3.mkv
386 The video stream from B.mp4 is sent to the hue filter, whose output is
388 cloned once using the split filter, and both outputs labelled. Then a
389 copy each is mapped to the first and third output files.
390 The overlay filter, requiring two video inputs, uses the first two
392 unused video streams. Those are the streams from A.avi and C.mkv. The
393 overlay output isn't labelled, so it is sent to the first output file
394 out1.mp4, regardless of the presence of the "-map" option.
395 The aresample filter is sent the first unused audio stream, that of
397 A.avi. Since this filter output is also unlabelled, it too is mapped to
398 the first output file. The presence of "-an" only suppresses automatic
399 or manual stream selection of audio streams, not outputs sent from
400 filtergraphs. Both these mapped streams shall be ordered before the
401 mapped stream in out1.mp4.
402 The video, audio and subtitle streams mapped to "out2.mkv" are entirely
404 determined by automatic stream selection.
405 out3.mkv consists of the cloned video output from the hue filter and
407 the first audio stream from B.mp4.
408
410 All the numerical options, if not specified otherwise, accept a string
411 representing a number as input, which may be followed by one of the SI
412 unit prefixes, for example: 'K', 'M', or 'G'.
413 If 'i' is appended to the SI unit prefix, the complete prefix will be
415 interpreted as a unit prefix for binary multiples, which are based on
416 powers of 1024 instead of powers of 1000. Appending 'B' to the SI unit
417 prefix multiplies the value by 8. This allows using, for example: 'KB',
418 'MiB', 'G' and 'B' as number suffixes.
419 Options which do not take arguments are boolean options, and set the
421 corresponding value to true. They can be set to false by prefixing the
422 option name with "no". For example using "-nofoo" will set the boolean
423 option with name "foo" to false.
424 Stream specifiers
426 Some options are applied per-stream, e.g. bitrate or codec. Stream
427 specifiers are used to precisely specify which stream(s) a given option
428 belongs to.
429 A stream specifier is a string generally appended to the option name
431 and separated from it by a colon. E.g. "-codec:a:1 ac3" contains the
432 "a:1" stream specifier, which matches the second audio stream.
433 Therefore, it would select the ac3 codec for the second audio stream.
434 A stream specifier can match several streams, so that the option is
436 applied to all of them. E.g. the stream specifier in "-b:a 128k"
437 matches all audio streams.
438 An empty stream specifier matches all streams. For example, "-codec
440 copy" or "-codec: copy" would copy all the streams without reencoding.
441 Possible forms of stream specifiers are:
443 stream_index
445 Matches the stream with this index. E.g. "-threads:1 4" would set
446 the thread count for the second stream to 4. If stream_index is
447 used as an additional stream specifier (see below), then it selects
448 stream number stream_index from the matching streams. Stream
449 numbering is based on the order of the streams as detected by
450 libavformat except when a program ID is also specified. In this
451 case it is based on the ordering of the streams in the program.
452 stream_type[:additional_stream_specifier]
454 stream_type is one of following: 'v' or 'V' for video, 'a' for
455 audio, 's' for subtitle, 'd' for data, and 't' for attachments. 'v'
456 matches all video streams, 'V' only matches video streams which are
457 not attached pictures, video thumbnails or cover arts. If
458 additional_stream_specifier is used, then it matches streams which
459 both have this type and match the additional_stream_specifier.
460 Otherwise, it matches all streams of the specified type.
461 p:program_id[:additional_stream_specifier]
463 Matches streams which are in the program with the id program_id. If
464 additional_stream_specifier is used, then it matches streams which
465 both are part of the program and match the
466 additional_stream_specifier.
467 #stream_id or i:stream_id
469 Match the stream by stream id (e.g. PID in MPEG-TS container).
470 m:key[:value]
472 Matches streams with the metadata tag key having the specified
473 value. If value is not given, matches streams that contain the
474 given tag with any value.
475 u Matches streams with usable configuration, the codec must be
477 defined and the essential information such as video dimension or
478 audio sample rate must be present.
479 Note that in ffmpeg, matching by metadata will only work properly
481 for input files.
482 Generic options
484 These options are shared amongst the ff* tools.
485 -L Show license.
487 -h, -?, -help, --help [arg]
489 Show help. An optional parameter may be specified to print help
490 about a specific item. If no argument is specified, only basic (non
491 advanced) tool options are shown.
492 Possible values of arg are:
494 long
496 Print advanced tool options in addition to the basic tool
497 options.
498 full
500 Print complete list of options, including shared and private
501 options for encoders, decoders, demuxers, muxers, filters, etc.
502 decoder=decoder_name
504 Print detailed information about the decoder named
505 decoder_name. Use the -decoders option to get a list of all
506 decoders.
507 encoder=encoder_name
509 Print detailed information about the encoder named
510 encoder_name. Use the -encoders option to get a list of all
511 encoders.
512 demuxer=demuxer_name
514 Print detailed information about the demuxer named
515 demuxer_name. Use the -formats option to get a list of all
516 demuxers and muxers.
517 muxer=muxer_name
519 Print detailed information about the muxer named muxer_name.
520 Use the -formats option to get a list of all muxers and
521 demuxers.
522 filter=filter_name
524 Print detailed information about the filter named filter_name.
525 Use the -filters option to get a list of all filters.
526 bsf=bitstream_filter_name
528 Print detailed information about the bitstream filter named
529 bitstream_filter_name. Use the -bsfs option to get a list of
530 all bitstream filters.
531 protocol=protocol_name
533 Print detailed information about the protocol named
534 protocol_name. Use the -protocols option to get a list of all
535 protocols.
536 -version
538 Show version.
539 -buildconf
541 Show the build configuration, one option per line.
542 -formats
544 Show available formats (including devices).
545 -demuxers
547 Show available demuxers.
548 -muxers
550 Show available muxers.
551 -devices
553 Show available devices.
554 -codecs
556 Show all codecs known to libavcodec.
557 Note that the term 'codec' is used throughout this documentation as
559 a shortcut for what is more correctly called a media bitstream
560 format.
561 -decoders
563 Show available decoders.
564 -encoders
566 Show all available encoders.
567 -bsfs
569 Show available bitstream filters.
570 -protocols
572 Show available protocols.
573 -filters
575 Show available libavfilter filters.
576 -pix_fmts
578 Show available pixel formats.
579 -sample_fmts
581 Show available sample formats.
582 -layouts
584 Show channel names and standard channel layouts.
585 -dispositions
587 Show stream dispositions.
588 -colors
590 Show recognized color names.
591 -sources device[,opt1=val1[,opt2=val2]...]
593 Show autodetected sources of the input device. Some devices may
594 provide system-dependent source names that cannot be autodetected.
595 The returned list cannot be assumed to be always complete.
596 ffmpeg -sources pulse,server=192.168.0.4
598 -sinks device[,opt1=val1[,opt2=val2]...]
600 Show autodetected sinks of the output device. Some devices may
601 provide system-dependent sink names that cannot be autodetected.
602 The returned list cannot be assumed to be always complete.
603 ffmpeg -sinks pulse,server=192.168.0.4
605 -loglevel [flags+]loglevel | -v [flags+]loglevel
607 Set logging level and flags used by the library.
608 The optional flags prefix can consist of the following values:
610 repeat
612 Indicates that repeated log output should not be compressed to
613 the first line and the "Last message repeated n times" line
614 will be omitted.
615 level
617 Indicates that log output should add a "[level]" prefix to each
618 message line. This can be used as an alternative to log
619 coloring, e.g. when dumping the log to file.
620 Flags can also be used alone by adding a '+'/'-' prefix to
622 set/reset a single flag without affecting other flags or changing
623 loglevel. When setting both flags and loglevel, a '+' separator is
624 expected between the last flags value and before loglevel.
625 loglevel is a string or a number containing one of the following
627 values:
628 quiet, -8
630 Show nothing at all; be silent.
631 panic, 0
633 Only show fatal errors which could lead the process to crash,
634 such as an assertion failure. This is not currently used for
635 anything.
636 fatal, 8
638 Only show fatal errors. These are errors after which the
639 process absolutely cannot continue.
640 error, 16
642 Show all errors, including ones which can be recovered from.
643 warning, 24
645 Show all warnings and errors. Any message related to possibly
646 incorrect or unexpected events will be shown.
647 info, 32
649 Show informative messages during processing. This is in
650 addition to warnings and errors. This is the default value.
651 verbose, 40
653 Same as "info", except more verbose.
654 debug, 48
656 Show everything, including debugging information.
657 trace, 56
659 For example to enable repeated log output, add the "level" prefix,
661 and set loglevel to "verbose":
662 ffmpeg -loglevel repeat+level+verbose -i input output
664 Another example that enables repeated log output without affecting
666 current state of "level" prefix flag or loglevel:
667 ffmpeg [...] -loglevel +repeat
669 By default the program logs to stderr. If coloring is supported by
671 the terminal, colors are used to mark errors and warnings. Log
672 coloring can be disabled setting the environment variable
673 AV_LOG_FORCE_NOCOLOR, or can be forced setting the environment
674 variable AV_LOG_FORCE_COLOR.
675 -report
677 Dump full command line and log output to a file named
678 "program-YYYYMMDD-HHMMSS.log" in the current directory. This file
679 can be useful for bug reports. It also implies "-loglevel debug".
680 Setting the environment variable FFREPORT to any value has the same
682 effect. If the value is a ':'-separated key=value sequence, these
683 options will affect the report; option values must be escaped if
684 they contain special characters or the options delimiter ':' (see
685 the ``Quoting and escaping'' section in the ffmpeg-utils manual).
686 The following options are recognized:
688 file
690 set the file name to use for the report; %p is expanded to the
691 name of the program, %t is expanded to a timestamp, "%%" is
692 expanded to a plain "%"
693 level
695 set the log verbosity level using a numerical value (see
696 "-loglevel").
697 For example, to output a report to a file named ffreport.log using
699 a log level of 32 (alias for log level "info"):
700 FFREPORT=file=ffreport.log:level=32 ffmpeg -i input output
702 Errors in parsing the environment variable are not fatal, and will
704 not appear in the report.
705 -hide_banner
707 Suppress printing banner.
708 All FFmpeg tools will normally show a copyright notice, build
710 options and library versions. This option can be used to suppress
711 printing this information.
712 -cpuflags flags (global)
714 Allows setting and clearing cpu flags. This option is intended for
715 testing. Do not use it unless you know what you're doing.
716 ffmpeg -cpuflags -sse+mmx ...
718 ffmpeg -cpuflags mmx ...
719 ffmpeg -cpuflags 0 ...
720 Possible flags for this option are:
722 x86
724 mmx
725 mmxext
726 sse
727 sse2
728 sse2slow
729 sse3
730 sse3slow
731 ssse3
732 atom
733 sse4.1
734 sse4.2
735 avx
736 avx2
737 xop
738 fma3
739 fma4
740 3dnow
741 3dnowext
742 bmi1
743 bmi2
744 cmov
745 ARM
746 armv5te
747 armv6
748 armv6t2
749 vfp
750 vfpv3
751 neon
752 setend
753 AArch64
754 armv8
755 vfp
756 neon
757 PowerPC
758 altivec
759 Specific Processors
760 pentium2
761 pentium3
762 pentium4
763 k6
764 k62
765 athlon
766 athlonxp
767 k8
768 -cpucount count (global)
769 Override detection of CPU count. This option is intended for
770 testing. Do not use it unless you know what you're doing.
771 ffmpeg -cpucount 2
773 -max_alloc bytes
775 Set the maximum size limit for allocating a block on the heap by
776 ffmpeg's family of malloc functions. Exercise extreme caution when
777 using this option. Don't use if you do not understand the full
778 consequence of doing so. Default is INT_MAX.
779 AVOptions
781 These options are provided directly by the libavformat, libavdevice and
782 libavcodec libraries. To see the list of available AVOptions, use the
783 -help option. They are separated into two categories:
784 generic
786 These options can be set for any container, codec or device.
787 Generic options are listed under AVFormatContext options for
788 containers/devices and under AVCodecContext options for codecs.
789 private
791 These options are specific to the given container, device or codec.
792 Private options are listed under their corresponding
793 containers/devices/codecs.
794 For example to write an ID3v2.3 header instead of a default ID3v2.4 to
796 an MP3 file, use the id3v2_version private option of the MP3 muxer:
797 ffmpeg -i input.flac -id3v2_version 3 out.mp3
799 All codec AVOptions are per-stream, and thus a stream specifier should
801 be attached to them:
802 ffmpeg -i multichannel.mxf -map 0:v:0 -map 0:a:0 -map 0:a:0 -c:a:0 ac3 -b:a:0 640k -ac:a:1 2 -c:a:1 aac -b:2 128k out.mp4
804 In the above example, a multichannel audio stream is mapped twice for
806 output. The first instance is encoded with codec ac3 and bitrate 640k.
807 The second instance is downmixed to 2 channels and encoded with codec
808 aac. A bitrate of 128k is specified for it using absolute index of the
809 output stream.
810 Note: the -nooption syntax cannot be used for boolean AVOptions, use
812 -option 0/-option 1.
813 Note: the old undocumented way of specifying per-stream AVOptions by
815 prepending v/a/s to the options name is now obsolete and will be
816 removed soon.
817 Main options
819 -f fmt (input/output)
820 Force input or output file format. The format is normally auto
821 detected for input files and guessed from the file extension for
822 output files, so this option is not needed in most cases.
823 -i url (input)
825 input file url
826 -y (global)
828 Overwrite output files without asking.
829 -n (global)
831 Do not overwrite output files, and exit immediately if a specified
832 output file already exists.
833 -stream_loop number (input)
835 Set number of times input stream shall be looped. Loop 0 means no
836 loop, loop -1 means infinite loop.
837 -recast_media (global)
839 Allow forcing a decoder of a different media type than the one
840 detected or designated by the demuxer. Useful for decoding media
841 data muxed as data streams.
842 -c[:stream_specifier] codec (input/output,per-stream)
844 -codec[:stream_specifier] codec (input/output,per-stream)
845 Select an encoder (when used before an output file) or a decoder
846 (when used before an input file) for one or more streams. codec is
847 the name of a decoder/encoder or a special value "copy" (output
848 only) to indicate that the stream is not to be re-encoded.
849 For example
851 ffmpeg -i INPUT -map 0 -c:v libx264 -c:a copy OUTPUT
853 encodes all video streams with libx264 and copies all audio
855 streams.
856 For each stream, the last matching "c" option is applied, so
858 ffmpeg -i INPUT -map 0 -c copy -c:v:1 libx264 -c:a:137 libvorbis OUTPUT
860 will copy all the streams except the second video, which will be
862 encoded with libx264, and the 138th audio, which will be encoded
863 with libvorbis.
864 -t duration (input/output)
866 When used as an input option (before "-i"), limit the duration of
867 data read from the input file.
868 When used as an output option (before an output url), stop writing
870 the output after its duration reaches duration.
871 duration must be a time duration specification, see the Time
873 duration section in the ffmpeg-utils(1) manual.
874 -to and -t are mutually exclusive and -t has priority.
876 -to position (input/output)
878 Stop writing the output or reading the input at position. position
879 must be a time duration specification, see the Time duration
880 section in the ffmpeg-utils(1) manual.
881 -to and -t are mutually exclusive and -t has priority.
883 -fs limit_size (output)
885 Set the file size limit, expressed in bytes. No further chunk of
886 bytes is written after the limit is exceeded. The size of the
887 output file is slightly more than the requested file size.
888 -ss position (input/output)
890 When used as an input option (before "-i"), seeks in this input
891 file to position. Note that in most formats it is not possible to
892 seek exactly, so ffmpeg will seek to the closest seek point before
893 position. When transcoding and -accurate_seek is enabled (the
894 default), this extra segment between the seek point and position
895 will be decoded and discarded. When doing stream copy or when
896 -noaccurate_seek is used, it will be preserved.
897 When used as an output option (before an output url), decodes but
899 discards input until the timestamps reach position.
900 position must be a time duration specification, see the Time
902 duration section in the ffmpeg-utils(1) manual.
903 -sseof position (input)
905 Like the "-ss" option but relative to the "end of file". That is
906 negative values are earlier in the file, 0 is at EOF.
907 -itsoffset offset (input)
909 Set the input time offset.
910 offset must be a time duration specification, see the Time duration
912 section in the ffmpeg-utils(1) manual.
913 The offset is added to the timestamps of the input files.
915 Specifying a positive offset means that the corresponding streams
916 are delayed by the time duration specified in offset.
917 -itsscale scale (input,per-stream)
919 Rescale input timestamps. scale should be a floating point number.
920 -timestamp date (output)
922 Set the recording timestamp in the container.
923 date must be a date specification, see the Date section in the
925 ffmpeg-utils(1) manual.
926 -metadata[:metadata_specifier] key=value (output,per-metadata)
928 Set a metadata key/value pair.
929 An optional metadata_specifier may be given to set metadata on
931 streams, chapters or programs. See "-map_metadata" documentation
932 for details.
933 This option overrides metadata set with "-map_metadata". It is also
935 possible to delete metadata by using an empty value.
936 For example, for setting the title in the output file:
938 ffmpeg -i in.avi -metadata title="my title" out.flv
940 To set the language of the first audio stream:
942 ffmpeg -i INPUT -metadata:s:a:0 language=eng OUTPUT
944 -disposition[:stream_specifier] value (output,per-stream)
946 Sets the disposition for a stream.
947 By default, the disposition is copied from the input stream, unless
949 the output stream this option applies to is fed by a complex
950 filtergraph - in that case the disposition is unset by default.
951 value is a sequence of items separated by '+' or '-'. The first
953 item may also be prefixed with '+' or '-', in which case this
954 option modifies the default value. Otherwise (the first item is not
955 prefixed) this options overrides the default value. A '+' prefix
956 adds the given disposition, '-' removes it. It is also possible to
957 clear the disposition by setting it to 0.
958 If no "-disposition" options were specified for an output file,
960 ffmpeg will automatically set the 'default' disposition on the
961 first stream of each type, when there are multiple streams of this
962 type in the output file and no stream of that type is already
963 marked as default.
964 The "-dispositions" option lists the known dispositions.
966 For example, to make the second audio stream the default stream:
968 ffmpeg -i in.mkv -c copy -disposition:a:1 default out.mkv
970 To make the second subtitle stream the default stream and remove
972 the default disposition from the first subtitle stream:
973 ffmpeg -i in.mkv -c copy -disposition:s:0 0 -disposition:s:1 default out.mkv
975 To add an embedded cover/thumbnail:
977 ffmpeg -i in.mp4 -i IMAGE -map 0 -map 1 -c copy -c:v:1 png -disposition:v:1 attached_pic out.mp4
979 Not all muxers support embedded thumbnails, and those who do, only
981 support a few formats, like JPEG or PNG.
982 -program
984 [title=title:][program_num=program_num:]st=stream[:st=stream...]
985 (output)
986 Creates a program with the specified title, program_num and adds
987 the specified stream(s) to it.
988 -target type (output)
990 Specify target file type ("vcd", "svcd", "dvd", "dv", "dv50"). type
991 may be prefixed with "pal-", "ntsc-" or "film-" to use the
992 corresponding standard. All the format options (bitrate, codecs,
993 buffer sizes) are then set automatically. You can just type:
994 ffmpeg -i myfile.avi -target vcd /tmp/vcd.mpg
996 Nevertheless you can specify additional options as long as you know
998 they do not conflict with the standard, as in:
999 ffmpeg -i myfile.avi -target vcd -bf 2 /tmp/vcd.mpg
1001 The parameters set for each target are as follows.
1003 VCD
1005 <pal>:
1007 -f vcd -muxrate 1411200 -muxpreload 0.44 -packetsize 2324
1008 -s 352x288 -r 25
1009 -codec:v mpeg1video -g 15 -b:v 1150k -maxrate:v 1150v -minrate:v 1150k -bufsize:v 327680
1010 -ar 44100 -ac 2
1011 -codec:a mp2 -b:a 224k
1012 <ntsc>:
1014 -f vcd -muxrate 1411200 -muxpreload 0.44 -packetsize 2324
1015 -s 352x240 -r 30000/1001
1016 -codec:v mpeg1video -g 18 -b:v 1150k -maxrate:v 1150v -minrate:v 1150k -bufsize:v 327680
1017 -ar 44100 -ac 2
1018 -codec:a mp2 -b:a 224k
1019 <film>:
1021 -f vcd -muxrate 1411200 -muxpreload 0.44 -packetsize 2324
1022 -s 352x240 -r 24000/1001
1023 -codec:v mpeg1video -g 18 -b:v 1150k -maxrate:v 1150v -minrate:v 1150k -bufsize:v 327680
1024 -ar 44100 -ac 2
1025 -codec:a mp2 -b:a 224k
1026 SVCD
1028 <pal>:
1030 -f svcd -packetsize 2324
1031 -s 480x576 -pix_fmt yuv420p -r 25
1032 -codec:v mpeg2video -g 15 -b:v 2040k -maxrate:v 2516k -minrate:v 0 -bufsize:v 1835008 -scan_offset 1
1033 -ar 44100
1034 -codec:a mp2 -b:a 224k
1035 <ntsc>:
1037 -f svcd -packetsize 2324
1038 -s 480x480 -pix_fmt yuv420p -r 30000/1001
1039 -codec:v mpeg2video -g 18 -b:v 2040k -maxrate:v 2516k -minrate:v 0 -bufsize:v 1835008 -scan_offset 1
1040 -ar 44100
1041 -codec:a mp2 -b:a 224k
1042 <film>:
1044 -f svcd -packetsize 2324
1045 -s 480x480 -pix_fmt yuv420p -r 24000/1001
1046 -codec:v mpeg2video -g 18 -b:v 2040k -maxrate:v 2516k -minrate:v 0 -bufsize:v 1835008 -scan_offset 1
1047 -ar 44100
1048 -codec:a mp2 -b:a 224k
1049 DVD
1051 <pal>:
1053 -f dvd -muxrate 10080k -packetsize 2048
1054 -s 720x576 -pix_fmt yuv420p -r 25
1055 -codec:v mpeg2video -g 15 -b:v 6000k -maxrate:v 9000k -minrate:v 0 -bufsize:v 1835008
1056 -ar 48000
1057 -codec:a ac3 -b:a 448k
1058 <ntsc>:
1060 -f dvd -muxrate 10080k -packetsize 2048
1061 -s 720x480 -pix_fmt yuv420p -r 30000/1001
1062 -codec:v mpeg2video -g 18 -b:v 6000k -maxrate:v 9000k -minrate:v 0 -bufsize:v 1835008
1063 -ar 48000
1064 -codec:a ac3 -b:a 448k
1065 <film>:
1067 -f dvd -muxrate 10080k -packetsize 2048
1068 -s 720x480 -pix_fmt yuv420p -r 24000/1001
1069 -codec:v mpeg2video -g 18 -b:v 6000k -maxrate:v 9000k -minrate:v 0 -bufsize:v 1835008
1070 -ar 48000
1071 -codec:a ac3 -b:a 448k
1072 DV
1074 <pal>:
1076 -f dv
1077 -s 720x576 -pix_fmt yuv420p -r 25
1078 -ar 48000 -ac 2
1079 <ntsc>:
1081 -f dv
1082 -s 720x480 -pix_fmt yuv411p -r 30000/1001
1083 -ar 48000 -ac 2
1084 <film>:
1086 -f dv
1087 -s 720x480 -pix_fmt yuv411p -r 24000/1001
1088 -ar 48000 -ac 2
1089 The "dv50" target is identical to the "dv" target except that the
1091 pixel format set is "yuv422p" for all three standards.
1092 Any user-set value for a parameter above will override the target
1094 preset value. In that case, the output may not comply with the
1095 target standard.
1096 -dn (input/output)
1098 As an input option, blocks all data streams of a file from being
1099 filtered or being automatically selected or mapped for any output.
1100 See "-discard" option to disable streams individually.
1101 As an output option, disables data recording i.e. automatic
1103 selection or mapping of any data stream. For full manual control
1104 see the "-map" option.
1105 -dframes number (output)
1107 Set the number of data frames to output. This is an obsolete alias
1108 for "-frames:d", which you should use instead.
1109 -frames[:stream_specifier] framecount (output,per-stream)
1111 Stop writing to the stream after framecount frames.
1112 -q[:stream_specifier] q (output,per-stream)
1114 -qscale[:stream_specifier] q (output,per-stream)
1115 Use fixed quality scale (VBR). The meaning of q/qscale is codec-
1116 dependent. If qscale is used without a stream_specifier then it
1117 applies only to the video stream, this is to maintain compatibility
1118 with previous behavior and as specifying the same codec specific
1119 value to 2 different codecs that is audio and video generally is
1120 not what is intended when no stream_specifier is used.
1121 -filter[:stream_specifier] filtergraph (output,per-stream)
1123 Create the filtergraph specified by filtergraph and use it to
1124 filter the stream.
1125 filtergraph is a description of the filtergraph to apply to the
1127 stream, and must have a single input and a single output of the
1128 same type of the stream. In the filtergraph, the input is
1129 associated to the label "in", and the output to the label "out".
1130 See the ffmpeg-filters manual for more information about the
1131 filtergraph syntax.
1132 See the -filter_complex option if you want to create filtergraphs
1134 with multiple inputs and/or outputs.
1135 -filter_script[:stream_specifier] filename (output,per-stream)
1137 This option is similar to -filter, the only difference is that its
1138 argument is the name of the file from which a filtergraph
1139 description is to be read.
1140 -reinit_filter[:stream_specifier] integer (input,per-stream)
1142 This boolean option determines if the filtergraph(s) to which this
1143 stream is fed gets reinitialized when input frame parameters change
1144 mid-stream. This option is enabled by default as most video and all
1145 audio filters cannot handle deviation in input frame properties.
1146 Upon reinitialization, existing filter state is lost, like e.g. the
1147 frame count "n" reference available in some filters. Any frames
1148 buffered at time of reinitialization are lost. The properties
1149 where a change triggers reinitialization are, for video, frame
1150 resolution or pixel format; for audio, sample format, sample rate,
1151 channel count or channel layout.
1152 -filter_threads nb_threads (global)
1154 Defines how many threads are used to process a filter pipeline.
1155 Each pipeline will produce a thread pool with this many threads
1156 available for parallel processing. The default is the number of
1157 available CPUs.
1158 -pre[:stream_specifier] preset_name (output,per-stream)
1160 Specify the preset for matching stream(s).
1161 -stats (global)
1163 Print encoding progress/statistics. It is on by default, to
1164 explicitly disable it you need to specify "-nostats".
1165 -stats_period time (global)
1167 Set period at which encoding progress/statistics are updated.
1168 Default is 0.5 seconds.
1169 -progress url (global)
1171 Send program-friendly progress information to url.
1172 Progress information is written periodically and at the end of the
1174 encoding process. It is made of "key=value" lines. key consists of
1175 only alphanumeric characters. The last key of a sequence of
1176 progress information is always "progress".
1177 The update period is set using "-stats_period".
1179 -stdin
1181 Enable interaction on standard input. On by default unless standard
1182 input is used as an input. To explicitly disable interaction you
1183 need to specify "-nostdin".
1184 Disabling interaction on standard input is useful, for example, if
1186 ffmpeg is in the background process group. Roughly the same result
1187 can be achieved with "ffmpeg ... < /dev/null" but it requires a
1188 shell.
1189 -debug_ts (global)
1191 Print timestamp information. It is off by default. This option is
1192 mostly useful for testing and debugging purposes, and the output
1193 format may change from one version to another, so it should not be
1194 employed by portable scripts.
1195 See also the option "-fdebug ts".
1197 -attach filename (output)
1199 Add an attachment to the output file. This is supported by a few
1200 formats like Matroska for e.g. fonts used in rendering subtitles.
1201 Attachments are implemented as a specific type of stream, so this
1202 option will add a new stream to the file. It is then possible to
1203 use per-stream options on this stream in the usual way. Attachment
1204 streams created with this option will be created after all the
1205 other streams (i.e. those created with "-map" or automatic
1206 mappings).
1207 Note that for Matroska you also have to set the mimetype metadata
1209 tag:
1210 ffmpeg -i INPUT -attach DejaVuSans.ttf -metadata:s:2 mimetype=application/x-truetype-font out.mkv
1212 (assuming that the attachment stream will be third in the output
1214 file).
1215 -dump_attachment[:stream_specifier] filename (input,per-stream)
1217 Extract the matching attachment stream into a file named filename.
1218 If filename is empty, then the value of the "filename" metadata tag
1219 will be used.
1220 E.g. to extract the first attachment to a file named 'out.ttf':
1222 ffmpeg -dump_attachment:t:0 out.ttf -i INPUT
1224 To extract all attachments to files determined by the "filename"
1226 tag:
1227 ffmpeg -dump_attachment:t "" -i INPUT
1229 Technical note -- attachments are implemented as codec extradata,
1231 so this option can actually be used to extract extradata from any
1232 stream, not just attachments.
1233 Video Options
1235 -vframes number (output)
1236 Set the number of video frames to output. This is an obsolete alias
1237 for "-frames:v", which you should use instead.
1238 -r[:stream_specifier] fps (input/output,per-stream)
1240 Set frame rate (Hz value, fraction or abbreviation).
1241 As an input option, ignore any timestamps stored in the file and
1243 instead generate timestamps assuming constant frame rate fps. This
1244 is not the same as the -framerate option used for some input
1245 formats like image2 or v4l2 (it used to be the same in older
1246 versions of FFmpeg). If in doubt use -framerate instead of the
1247 input option -r.
1248 As an output option, duplicate or drop input frames to achieve
1250 constant output frame rate fps.
1251 -fpsmax[:stream_specifier] fps (output,per-stream)
1253 Set maximum frame rate (Hz value, fraction or abbreviation).
1254 Clamps output frame rate when output framerate is auto-set and is
1256 higher than this value. Useful in batch processing or when input
1257 framerate is wrongly detected as very high. It cannot be set
1258 together with "-r". It is ignored during streamcopy.
1259 -s[:stream_specifier] size (input/output,per-stream)
1261 Set frame size.
1262 As an input option, this is a shortcut for the video_size private
1264 option, recognized by some demuxers for which the frame size is
1265 either not stored in the file or is configurable -- e.g. raw video
1266 or video grabbers.
1267 As an output option, this inserts the "scale" video filter to the
1269 end of the corresponding filtergraph. Please use the "scale" filter
1270 directly to insert it at the beginning or some other place.
1271 The format is wxh (default - same as source).
1273 -aspect[:stream_specifier] aspect (output,per-stream)
1275 Set the video display aspect ratio specified by aspect.
1276 aspect can be a floating point number string, or a string of the
1278 form num:den, where num and den are the numerator and denominator
1279 of the aspect ratio. For example "4:3", "16:9", "1.3333", and
1280 "1.7777" are valid argument values.
1281 If used together with -vcodec copy, it will affect the aspect ratio
1283 stored at container level, but not the aspect ratio stored in
1284 encoded frames, if it exists.
1285 -vn (input/output)
1287 As an input option, blocks all video streams of a file from being
1288 filtered or being automatically selected or mapped for any output.
1289 See "-discard" option to disable streams individually.
1290 As an output option, disables video recording i.e. automatic
1292 selection or mapping of any video stream. For full manual control
1293 see the "-map" option.
1294 -vcodec codec (output)
1296 Set the video codec. This is an alias for "-codec:v".
1297 -pass[:stream_specifier] n (output,per-stream)
1299 Select the pass number (1 or 2). It is used to do two-pass video
1300 encoding. The statistics of the video are recorded in the first
1301 pass into a log file (see also the option -passlogfile), and in the
1302 second pass that log file is used to generate the video at the
1303 exact requested bitrate. On pass 1, you may just deactivate audio
1304 and set output to null, examples for Windows and Unix:
1305 ffmpeg -i foo.mov -c:v libxvid -pass 1 -an -f rawvideo -y NUL
1307 ffmpeg -i foo.mov -c:v libxvid -pass 1 -an -f rawvideo -y /dev/null
1308 -passlogfile[:stream_specifier] prefix (output,per-stream)
1310 Set two-pass log file name prefix to prefix, the default file name
1311 prefix is ``ffmpeg2pass''. The complete file name will be
1312 PREFIX-N.log, where N is a number specific to the output stream
1313 -vf filtergraph (output)
1315 Create the filtergraph specified by filtergraph and use it to
1316 filter the stream.
1317 This is an alias for "-filter:v", see the -filter option.
1319 -autorotate
1321 Automatically rotate the video according to file metadata. Enabled
1322 by default, use -noautorotate to disable it.
1323 -autoscale
1325 Automatically scale the video according to the resolution of first
1326 frame. Enabled by default, use -noautoscale to disable it. When
1327 autoscale is disabled, all output frames of filter graph might not
1328 be in the same resolution and may be inadequate for some
1329 encoder/muxer. Therefore, it is not recommended to disable it
1330 unless you really know what you are doing. Disable autoscale at
1331 your own risk.
1332 Advanced Video options
1334 -pix_fmt[:stream_specifier] format (input/output,per-stream)
1335 Set pixel format. Use "-pix_fmts" to show all the supported pixel
1336 formats. If the selected pixel format can not be selected, ffmpeg
1337 will print a warning and select the best pixel format supported by
1338 the encoder. If pix_fmt is prefixed by a "+", ffmpeg will exit
1339 with an error if the requested pixel format can not be selected,
1340 and automatic conversions inside filtergraphs are disabled. If
1341 pix_fmt is a single "+", ffmpeg selects the same pixel format as
1342 the input (or graph output) and automatic conversions are disabled.
1343 -sws_flags flags (input/output)
1345 Set SwScaler flags.
1346 -rc_override[:stream_specifier] override (output,per-stream)
1348 Rate control override for specific intervals, formatted as
1349 "int,int,int" list separated with slashes. Two first values are the
1350 beginning and end frame numbers, last one is quantizer to use if
1351 positive, or quality factor if negative.
1352 -ilme
1354 Force interlacing support in encoder (MPEG-2 and MPEG-4 only). Use
1355 this option if your input file is interlaced and you want to keep
1356 the interlaced format for minimum losses. The alternative is to
1357 deinterlace the input stream by use of a filter such as "yadif" or
1358 "bwdif", but deinterlacing introduces losses.
1359 -psnr
1361 Calculate PSNR of compressed frames.
1362 -vstats
1364 Dump video coding statistics to vstats_HHMMSS.log.
1365 -vstats_file file
1367 Dump video coding statistics to file.
1368 -vstats_version file
1370 Specifies which version of the vstats format to use. Default is 2.
1371 version = 1 :
1373 "frame= %5d q= %2.1f PSNR= %6.2f f_size= %6d s_size= %8.0fkB time=
1375 %0.3f br= %7.1fkbits/s avg_br= %7.1fkbits/s"
1376 version > 1:
1378 "out= %2d st= %2d frame= %5d q= %2.1f PSNR= %6.2f f_size= %6d
1380 s_size= %8.0fkB time= %0.3f br= %7.1fkbits/s avg_br= %7.1fkbits/s"
1381 -top[:stream_specifier] n (output,per-stream)
1383 top=1/bottom=0/auto=-1 field first
1384 -dc precision
1386 Intra_dc_precision.
1387 -vtag fourcc/tag (output)
1389 Force video tag/fourcc. This is an alias for "-tag:v".
1390 -qphist (global)
1392 Show QP histogram
1393 -vbsf bitstream_filter
1395 Deprecated see -bsf
1396 -force_key_frames[:stream_specifier] time[,time...] (output,per-stream)
1398 -force_key_frames[:stream_specifier] expr:expr (output,per-stream)
1399 -force_key_frames[:stream_specifier] source (output,per-stream)
1400 -force_key_frames[:stream_specifier] source_no_drop (output,per-stream)
1401 force_key_frames can take arguments of the following form:
1402 time[,time...]
1404 If the argument consists of timestamps, ffmpeg will round the
1405 specified times to the nearest output timestamp as per the
1406 encoder time base and force a keyframe at the first frame
1407 having timestamp equal or greater than the computed timestamp.
1408 Note that if the encoder time base is too coarse, then the
1409 keyframes may be forced on frames with timestamps lower than
1410 the specified time. The default encoder time base is the
1411 inverse of the output framerate but may be set otherwise via
1412 "-enc_time_base".
1413 If one of the times is ""chapters"[delta]", it is expanded into
1415 the time of the beginning of all chapters in the file, shifted
1416 by delta, expressed as a time in seconds. This option can be
1417 useful to ensure that a seek point is present at a chapter mark
1418 or any other designated place in the output file.
1419 For example, to insert a key frame at 5 minutes, plus key
1421 frames 0.1 second before the beginning of every chapter:
1422 -force_key_frames 0:05:00,chapters-0.1
1424 expr:expr
1426 If the argument is prefixed with "expr:", the string expr is
1427 interpreted like an expression and is evaluated for each frame.
1428 A key frame is forced in case the evaluation is non-zero.
1429 The expression in expr can contain the following constants:
1431 n the number of current processed frame, starting from 0
1433 n_forced
1435 the number of forced frames
1436 prev_forced_n
1438 the number of the previous forced frame, it is "NAN" when
1439 no keyframe was forced yet
1440 prev_forced_t
1442 the time of the previous forced frame, it is "NAN" when no
1443 keyframe was forced yet
1444 t the time of the current processed frame
1446 For example to force a key frame every 5 seconds, you can
1448 specify:
1449 -force_key_frames expr:gte(t,n_forced*5)
1451 To force a key frame 5 seconds after the time of the last
1453 forced one, starting from second 13:
1454 -force_key_frames expr:if(isnan(prev_forced_t),gte(t,13),gte(t,prev_forced_t+5))
1456 source
1458 If the argument is "source", ffmpeg will force a key frame if
1459 the current frame being encoded is marked as a key frame in its
1460 source.
1461 source_no_drop
1463 If the argument is "source_no_drop", ffmpeg will force a key
1464 frame if the current frame being encoded is marked as a key
1465 frame in its source. In cases where this particular source
1466 frame has to be dropped, enforce the next available frame to
1467 become a key frame instead.
1468 Note that forcing too many keyframes is very harmful for the
1470 lookahead algorithms of certain encoders: using fixed-GOP options
1471 or similar would be more efficient.
1472 -copyinkf[:stream_specifier] (output,per-stream)
1474 When doing stream copy, copy also non-key frames found at the
1475 beginning.
1476 -init_hw_device type[=name][:device[,key=value...]]
1478 Initialise a new hardware device of type type called name, using
1479 the given device parameters. If no name is specified it will
1480 receive a default name of the form "type%d".
1481 The meaning of device and the following arguments depends on the
1483 device type:
1484 cuda
1486 device is the number of the CUDA device.
1487 The following options are recognized:
1489 primary_ctx
1491 If set to 1, uses the primary device context instead of
1492 creating a new one.
1493 Examples:
1495 -init_hw_device cuda:1
1497 Choose the second device on the system.
1498 -init_hw_device cuda:0,primary_ctx=1
1500 Choose the first device and use the primary device context.
1501 dxva2
1503 device is the number of the Direct3D 9 display adapter.
1504 d3d11va
1506 device is the number of the Direct3D 11 display adapter.
1507 vaapi
1509 device is either an X11 display name or a DRM render node. If
1510 not specified, it will attempt to open the default X11 display
1511 ($DISPLAY) and then the first DRM render node
1512 (/dev/dri/renderD128).
1513 vdpau
1515 device is an X11 display name. If not specified, it will
1516 attempt to open the default X11 display ($DISPLAY).
1517 qsv device selects a value in MFX_IMPL_*. Allowed values are:
1519 auto
1521 sw
1522 hw
1523 auto_any
1524 hw_any
1525 hw2
1526 hw3
1527 hw4
1528 If not specified, auto_any is used. (Note that it may be
1530 easier to achieve the desired result for QSV by creating the
1531 platform-appropriate subdevice (dxva2 or d3d11va or vaapi) and
1532 then deriving a QSV device from that.)
1533 Alternatively, child_device_type helps to choose platform-
1535 appropriate subdevice type. On Windows d3d11va is used as
1536 default subdevice type.
1537 Examples:
1539 -init_hw_device qsv:hw,child_device_type=d3d11va
1541 Choose the GPU subdevice with type d3d11va and create QSV
1542 device with MFX_IMPL_HARDWARE.
1543 -init_hw_device qsv:hw,child_device_type=dxva2
1545 Choose the GPU subdevice with type dxva2 and create QSV
1546 device with MFX_IMPL_HARDWARE.
1547 opencl
1549 device selects the platform and device as
1550 platform_index.device_index.
1551 The set of devices can also be filtered using the key-value
1553 pairs to find only devices matching particular platform or
1554 device strings.
1555 The strings usable as filters are:
1557 platform_profile
1559 platform_version
1560 platform_name
1561 platform_vendor
1562 platform_extensions
1563 device_name
1564 device_vendor
1565 driver_version
1566 device_version
1567 device_profile
1568 device_extensions
1569 device_type
1570 The indices and filters must together uniquely select a device.
1572 Examples:
1574 -init_hw_device opencl:0.1
1576 Choose the second device on the first platform.
1577 -init_hw_device opencl:,device_name=Foo9000
1579 Choose the device with a name containing the string
1580 Foo9000.
1581 -init_hw_device
1583 opencl:1,device_type=gpu,device_extensions=cl_khr_fp16
1584 Choose the GPU device on the second platform supporting the
1585 cl_khr_fp16 extension.
1586 vulkan
1588 If device is an integer, it selects the device by its index in
1589 a system-dependent list of devices. If device is any other
1590 string, it selects the first device with a name containing that
1591 string as a substring.
1592 The following options are recognized:
1594 debug
1596 If set to 1, enables the validation layer, if installed.
1597 linear_images
1599 If set to 1, images allocated by the hwcontext will be
1600 linear and locally mappable.
1601 instance_extensions
1603 A plus separated list of additional instance extensions to
1604 enable.
1605 device_extensions
1607 A plus separated list of additional device extensions to
1608 enable.
1609 Examples:
1611 -init_hw_device vulkan:1
1613 Choose the second device on the system.
1614 -init_hw_device vulkan:RADV
1616 Choose the first device with a name containing the string
1617 RADV.
1618 -init_hw_device
1620 vulkan:0,instance_extensions=VK_KHR_wayland_surface+VK_KHR_xcb_surface
1621 Choose the first device and enable the Wayland and XCB
1622 instance extensions.
1623 -init_hw_device type[=name]@source
1625 Initialise a new hardware device of type type called name, deriving
1626 it from the existing device with the name source.
1627 -init_hw_device list
1629 List all hardware device types supported in this build of ffmpeg.
1630 -filter_hw_device name
1632 Pass the hardware device called name to all filters in any filter
1633 graph. This can be used to set the device to upload to with the
1634 "hwupload" filter, or the device to map to with the "hwmap" filter.
1635 Other filters may also make use of this parameter when they require
1636 a hardware device. Note that this is typically only required when
1637 the input is not already in hardware frames - when it is, filters
1638 will derive the device they require from the context of the frames
1639 they receive as input.
1640 This is a global setting, so all filters will receive the same
1642 device.
1643 -hwaccel[:stream_specifier] hwaccel (input,per-stream)
1645 Use hardware acceleration to decode the matching stream(s). The
1646 allowed values of hwaccel are:
1647 none
1649 Do not use any hardware acceleration (the default).
1650 auto
1652 Automatically select the hardware acceleration method.
1653 vdpau
1655 Use VDPAU (Video Decode and Presentation API for Unix) hardware
1656 acceleration.
1657 dxva2
1659 Use DXVA2 (DirectX Video Acceleration) hardware acceleration.
1660 d3d11va
1662 Use D3D11VA (DirectX Video Acceleration) hardware acceleration.
1663 vaapi
1665 Use VAAPI (Video Acceleration API) hardware acceleration.
1666 qsv Use the Intel QuickSync Video acceleration for video
1668 transcoding.
1669 Unlike most other values, this option does not enable
1671 accelerated decoding (that is used automatically whenever a qsv
1672 decoder is selected), but accelerated transcoding, without
1673 copying the frames into the system memory.
1674 For it to work, both the decoder and the encoder must support
1676 QSV acceleration and no filters must be used.
1677 This option has no effect if the selected hwaccel is not available
1679 or not supported by the chosen decoder.
1680 Note that most acceleration methods are intended for playback and
1682 will not be faster than software decoding on modern CPUs.
1683 Additionally, ffmpeg will usually need to copy the decoded frames
1684 from the GPU memory into the system memory, resulting in further
1685 performance loss. This option is thus mainly useful for testing.
1686 -hwaccel_device[:stream_specifier] hwaccel_device (input,per-stream)
1688 Select a device to use for hardware acceleration.
1689 This option only makes sense when the -hwaccel option is also
1691 specified. It can either refer to an existing device created with
1692 -init_hw_device by name, or it can create a new device as if
1693 -init_hw_device type:hwaccel_device were called immediately before.
1694 -hwaccels
1696 List all hardware acceleration components enabled in this build of
1697 ffmpeg. Actual runtime availability depends on the hardware and
1698 its suitable driver being installed.
1699 Audio Options
1701 -aframes number (output)
1702 Set the number of audio frames to output. This is an obsolete alias
1703 for "-frames:a", which you should use instead.
1704 -ar[:stream_specifier] freq (input/output,per-stream)
1706 Set the audio sampling frequency. For output streams it is set by
1707 default to the frequency of the corresponding input stream. For
1708 input streams this option only makes sense for audio grabbing
1709 devices and raw demuxers and is mapped to the corresponding demuxer
1710 options.
1711 -aq q (output)
1713 Set the audio quality (codec-specific, VBR). This is an alias for
1714 -q:a.
1715 -ac[:stream_specifier] channels (input/output,per-stream)
1717 Set the number of audio channels. For output streams it is set by
1718 default to the number of input audio channels. For input streams
1719 this option only makes sense for audio grabbing devices and raw
1720 demuxers and is mapped to the corresponding demuxer options.
1721 -an (input/output)
1723 As an input option, blocks all audio streams of a file from being
1724 filtered or being automatically selected or mapped for any output.
1725 See "-discard" option to disable streams individually.
1726 As an output option, disables audio recording i.e. automatic
1728 selection or mapping of any audio stream. For full manual control
1729 see the "-map" option.
1730 -acodec codec (input/output)
1732 Set the audio codec. This is an alias for "-codec:a".
1733 -sample_fmt[:stream_specifier] sample_fmt (output,per-stream)
1735 Set the audio sample format. Use "-sample_fmts" to get a list of
1736 supported sample formats.
1737 -af filtergraph (output)
1739 Create the filtergraph specified by filtergraph and use it to
1740 filter the stream.
1741 This is an alias for "-filter:a", see the -filter option.
1743 Advanced Audio options
1745 -atag fourcc/tag (output)
1746 Force audio tag/fourcc. This is an alias for "-tag:a".
1747 -absf bitstream_filter
1749 Deprecated, see -bsf
1750 -guess_layout_max channels (input,per-stream)
1752 If some input channel layout is not known, try to guess only if it
1753 corresponds to at most the specified number of channels. For
1754 example, 2 tells to ffmpeg to recognize 1 channel as mono and 2
1755 channels as stereo but not 6 channels as 5.1. The default is to
1756 always try to guess. Use 0 to disable all guessing.
1757 Subtitle options
1759 -scodec codec (input/output)
1760 Set the subtitle codec. This is an alias for "-codec:s".
1761 -sn (input/output)
1763 As an input option, blocks all subtitle streams of a file from
1764 being filtered or being automatically selected or mapped for any
1765 output. See "-discard" option to disable streams individually.
1766 As an output option, disables subtitle recording i.e. automatic
1768 selection or mapping of any subtitle stream. For full manual
1769 control see the "-map" option.
1770 -sbsf bitstream_filter
1772 Deprecated, see -bsf
1773 Advanced Subtitle options
1775 -fix_sub_duration
1776 Fix subtitles durations. For each subtitle, wait for the next
1777 packet in the same stream and adjust the duration of the first to
1778 avoid overlap. This is necessary with some subtitles codecs,
1779 especially DVB subtitles, because the duration in the original
1780 packet is only a rough estimate and the end is actually marked by
1781 an empty subtitle frame. Failing to use this option when necessary
1782 can result in exaggerated durations or muxing failures due to non-
1783 monotonic timestamps.
1784 Note that this option will delay the output of all data until the
1786 next subtitle packet is decoded: it may increase memory consumption
1787 and latency a lot.
1788 -canvas_size size
1790 Set the size of the canvas used to render subtitles.
1791 Advanced options
1793 -map
1794 [-]input_file_id[:stream_specifier][?][,sync_file_id[:stream_specifier]]
1795 | [linklabel] (output)
1796 Designate one or more input streams as a source for the output
1797 file. Each input stream is identified by the input file index
1798 input_file_id and the input stream index input_stream_id within the
1799 input file. Both indices start at 0. If specified,
1800 sync_file_id:stream_specifier sets which input stream is used as a
1801 presentation sync reference.
1802 The first "-map" option on the command line specifies the source
1804 for output stream 0, the second "-map" option specifies the source
1805 for output stream 1, etc.
1806 A "-" character before the stream identifier creates a "negative"
1808 mapping. It disables matching streams from already created
1809 mappings.
1810 A trailing "?" after the stream index will allow the map to be
1812 optional: if the map matches no streams the map will be ignored
1813 instead of failing. Note the map will still fail if an invalid
1814 input file index is used; such as if the map refers to a non-
1815 existent input.
1816 An alternative [linklabel] form will map outputs from complex
1818 filter graphs (see the -filter_complex option) to the output file.
1819 linklabel must correspond to a defined output link label in the
1820 graph.
1821 For example, to map ALL streams from the first input file to output
1823 ffmpeg -i INPUT -map 0 output
1825 For example, if you have two audio streams in the first input file,
1827 these streams are identified by "0:0" and "0:1". You can use "-map"
1828 to select which streams to place in an output file. For example:
1829 ffmpeg -i INPUT -map 0:1 out.wav
1831 will map the input stream in INPUT identified by "0:1" to the
1833 (single) output stream in out.wav.
1834 For example, to select the stream with index 2 from input file
1836 a.mov (specified by the identifier "0:2"), and stream with index 6
1837 from input b.mov (specified by the identifier "1:6"), and copy them
1838 to the output file out.mov:
1839 ffmpeg -i a.mov -i b.mov -c copy -map 0:2 -map 1:6 out.mov
1841 To select all video and the third audio stream from an input file:
1843 ffmpeg -i INPUT -map 0:v -map 0:a:2 OUTPUT
1845 To map all the streams except the second audio, use negative
1847 mappings
1848 ffmpeg -i INPUT -map 0 -map -0:a:1 OUTPUT
1850 To map the video and audio streams from the first input, and using
1852 the trailing "?", ignore the audio mapping if no audio streams
1853 exist in the first input:
1854 ffmpeg -i INPUT -map 0:v -map 0:a? OUTPUT
1856 To pick the English audio stream:
1858 ffmpeg -i INPUT -map 0:m:language:eng OUTPUT
1860 Note that using this option disables the default mappings for this
1862 output file.
1863 -ignore_unknown
1865 Ignore input streams with unknown type instead of failing if
1866 copying such streams is attempted.
1867 -copy_unknown
1869 Allow input streams with unknown type to be copied instead of
1870 failing if copying such streams is attempted.
1871 -map_channel
1873 [input_file_id.stream_specifier.channel_id|-1][?][:output_file_id.stream_specifier]
1874 Map an audio channel from a given input to an output. If
1875 output_file_id.stream_specifier is not set, the audio channel will
1876 be mapped on all the audio streams.
1877 Using "-1" instead of input_file_id.stream_specifier.channel_id
1879 will map a muted channel.
1880 A trailing "?" will allow the map_channel to be optional: if the
1882 map_channel matches no channel the map_channel will be ignored
1883 instead of failing.
1884 For example, assuming INPUT is a stereo audio file, you can switch
1886 the two audio channels with the following command:
1887 ffmpeg -i INPUT -map_channel 0.0.1 -map_channel 0.0.0 OUTPUT
1889 If you want to mute the first channel and keep the second:
1891 ffmpeg -i INPUT -map_channel -1 -map_channel 0.0.1 OUTPUT
1893 The order of the "-map_channel" option specifies the order of the
1895 channels in the output stream. The output channel layout is guessed
1896 from the number of channels mapped (mono if one "-map_channel",
1897 stereo if two, etc.). Using "-ac" in combination of "-map_channel"
1898 makes the channel gain levels to be updated if input and output
1899 channel layouts don't match (for instance two "-map_channel"
1900 options and "-ac 6").
1901 You can also extract each channel of an input to specific outputs;
1903 the following command extracts two channels of the INPUT audio
1904 stream (file 0, stream 0) to the respective OUTPUT_CH0 and
1905 OUTPUT_CH1 outputs:
1906 ffmpeg -i INPUT -map_channel 0.0.0 OUTPUT_CH0 -map_channel 0.0.1 OUTPUT_CH1
1908 The following example splits the channels of a stereo input into
1910 two separate streams, which are put into the same output file:
1911 ffmpeg -i stereo.wav -map 0:0 -map 0:0 -map_channel 0.0.0:0.0 -map_channel 0.0.1:0.1 -y out.ogg
1913 Note that currently each output stream can only contain channels
1915 from a single input stream; you can't for example use
1916 "-map_channel" to pick multiple input audio channels contained in
1917 different streams (from the same or different files) and merge them
1918 into a single output stream. It is therefore not currently
1919 possible, for example, to turn two separate mono streams into a
1920 single stereo stream. However splitting a stereo stream into two
1921 single channel mono streams is possible.
1922 If you need this feature, a possible workaround is to use the
1924 amerge filter. For example, if you need to merge a media (here
1925 input.mkv) with 2 mono audio streams into one single stereo channel
1926 audio stream (and keep the video stream), you can use the following
1927 command:
1928 ffmpeg -i input.mkv -filter_complex "[0:1] [0:2] amerge" -c:a pcm_s16le -c:v copy output.mkv
1930 To map the first two audio channels from the first input, and using
1932 the trailing "?", ignore the audio channel mapping if the first
1933 input is mono instead of stereo:
1934 ffmpeg -i INPUT -map_channel 0.0.0 -map_channel 0.0.1? OUTPUT
1936 -map_metadata[:metadata_spec_out] infile[:metadata_spec_in]
1938 (output,per-metadata)
1939 Set metadata information of the next output file from infile. Note
1940 that those are file indices (zero-based), not filenames. Optional
1941 metadata_spec_in/out parameters specify, which metadata to copy. A
1942 metadata specifier can have the following forms:
1943 g global metadata, i.e. metadata that applies to the whole file
1945 s[:stream_spec]
1947 per-stream metadata. stream_spec is a stream specifier as
1948 described in the Stream specifiers chapter. In an input
1949 metadata specifier, the first matching stream is copied from.
1950 In an output metadata specifier, all matching streams are
1951 copied to.
1952 c:chapter_index
1954 per-chapter metadata. chapter_index is the zero-based chapter
1955 index.
1956 p:program_index
1958 per-program metadata. program_index is the zero-based program
1959 index.
1960 If metadata specifier is omitted, it defaults to global.
1962 By default, global metadata is copied from the first input file,
1964 per-stream and per-chapter metadata is copied along with
1965 streams/chapters. These default mappings are disabled by creating
1966 any mapping of the relevant type. A negative file index can be used
1967 to create a dummy mapping that just disables automatic copying.
1968 For example to copy metadata from the first stream of the input
1970 file to global metadata of the output file:
1971 ffmpeg -i in.ogg -map_metadata 0:s:0 out.mp3
1973 To do the reverse, i.e. copy global metadata to all audio streams:
1975 ffmpeg -i in.mkv -map_metadata:s:a 0:g out.mkv
1977 Note that simple 0 would work as well in this example, since global
1979 metadata is assumed by default.
1980 -map_chapters input_file_index (output)
1982 Copy chapters from input file with index input_file_index to the
1983 next output file. If no chapter mapping is specified, then chapters
1984 are copied from the first input file with at least one chapter. Use
1985 a negative file index to disable any chapter copying.
1986 -benchmark (global)
1988 Show benchmarking information at the end of an encode. Shows real,
1989 system and user time used and maximum memory consumption. Maximum
1990 memory consumption is not supported on all systems, it will usually
1991 display as 0 if not supported.
1992 -benchmark_all (global)
1994 Show benchmarking information during the encode. Shows real,
1995 system and user time used in various steps (audio/video
1996 encode/decode).
1997 -timelimit duration (global)
1999 Exit after ffmpeg has been running for duration seconds in CPU user
2000 time.
2001 -dump (global)
2003 Dump each input packet to stderr.
2004 -hex (global)
2006 When dumping packets, also dump the payload.
2007 -readrate speed (input)
2009 Limit input read speed.
2010 Its value is a floating-point positive number which represents the
2012 maximum duration of media, in seconds, that should be ingested in
2013 one second of wallclock time. Default value is zero and represents
2014 no imposed limitation on speed of ingestion. Value 1 represents
2015 real-time speed and is equivalent to "-re".
2016 Mainly used to simulate a capture device or live input stream (e.g.
2018 when reading from a file). Should not be used with a low value
2019 when input is an actual capture device or live stream as it may
2020 cause packet loss.
2021 It is useful for when flow speed of output packets is important,
2023 such as live streaming.
2024 -re (input)
2026 Read input at native frame rate. This is equivalent to setting
2027 "-readrate 1".
2028 -vsync parameter
2030 Video sync method.
2031 For compatibility reasons some of the values can be specified as
2033 numbers (shown in parentheses in the following table). This is
2034 deprecated and will stop working in the future.
2035 passthrough (0)
2037 Each frame is passed with its timestamp from the demuxer to the
2038 muxer.
2039 cfr (1)
2041 Frames will be duplicated and dropped to achieve exactly the
2042 requested constant frame rate.
2043 vfr (2)
2045 Frames are passed through with their timestamp or dropped so as
2046 to prevent 2 frames from having the same timestamp.
2047 drop
2049 As passthrough but destroys all timestamps, making the muxer
2050 generate fresh timestamps based on frame-rate.
2051 auto (-1)
2053 Chooses between cfr and vfr depending on muxer capabilities.
2054 This is the default method.
2055 Note that the timestamps may be further modified by the muxer,
2057 after this. For example, in the case that the format option
2058 avoid_negative_ts is enabled.
2059 With -map you can select from which stream the timestamps should be
2061 taken. You can leave either video or audio unchanged and sync the
2062 remaining stream(s) to the unchanged one.
2063 -frame_drop_threshold parameter
2065 Frame drop threshold, which specifies how much behind video frames
2066 can be before they are dropped. In frame rate units, so 1.0 is one
2067 frame. The default is -1.1. One possible usecase is to avoid
2068 framedrops in case of noisy timestamps or to increase frame drop
2069 precision in case of exact timestamps.
2070 -async samples_per_second
2072 Audio sync method. "Stretches/squeezes" the audio stream to match
2073 the timestamps, the parameter is the maximum samples per second by
2074 which the audio is changed. -async 1 is a special case where only
2075 the start of the audio stream is corrected without any later
2076 correction.
2077 Note that the timestamps may be further modified by the muxer,
2079 after this. For example, in the case that the format option
2080 avoid_negative_ts is enabled.
2081 This option has been deprecated. Use the "aresample" audio filter
2083 instead.
2084 -adrift_threshold time
2086 Set the minimum difference between timestamps and audio data (in
2087 seconds) to trigger adding/dropping samples to make it match the
2088 timestamps. This option effectively is a threshold to select
2089 between hard (add/drop) and soft (squeeze/stretch) compensation.
2090 "-async" must be set to a positive value.
2091 -apad parameters (output,per-stream)
2093 Pad the output audio stream(s). This is the same as applying "-af
2094 apad". Argument is a string of filter parameters composed the same
2095 as with the "apad" filter. "-shortest" must be set for this output
2096 for the option to take effect.
2097 -copyts
2099 Do not process input timestamps, but keep their values without
2100 trying to sanitize them. In particular, do not remove the initial
2101 start time offset value.
2102 Note that, depending on the vsync option or on specific muxer
2104 processing (e.g. in case the format option avoid_negative_ts is
2105 enabled) the output timestamps may mismatch with the input
2106 timestamps even when this option is selected.
2107 -start_at_zero
2109 When used with copyts, shift input timestamps so they start at
2110 zero.
2111 This means that using e.g. "-ss 50" will make output timestamps
2113 start at 50 seconds, regardless of what timestamp the input file
2114 started at.
2115 -copytb mode
2117 Specify how to set the encoder timebase when stream copying. mode
2118 is an integer numeric value, and can assume one of the following
2119 values:
2120 1 Use the demuxer timebase.
2122 The time base is copied to the output encoder from the
2124 corresponding input demuxer. This is sometimes required to
2125 avoid non monotonically increasing timestamps when copying
2126 video streams with variable frame rate.
2127 0 Use the decoder timebase.
2129 The time base is copied to the output encoder from the
2131 corresponding input decoder.
2132 -1 Try to make the choice automatically, in order to generate a
2134 sane output.
2135 Default value is -1.
2137 -enc_time_base[:stream_specifier] timebase (output,per-stream)
2139 Set the encoder timebase. timebase is a floating point number, and
2140 can assume one of the following values:
2141 0 Assign a default value according to the media type.
2143 For video - use 1/framerate, for audio - use 1/samplerate.
2145 -1 Use the input stream timebase when possible.
2147 If an input stream is not available, the default timebase will
2149 be used.
2150 >0 Use the provided number as the timebase.
2152 This field can be provided as a ratio of two integers (e.g.
2154 1:24, 1:48000) or as a floating point number (e.g. 0.04166,
2155 2.0833e-5)
2156 Default value is 0.
2158 -bitexact (input/output)
2160 Enable bitexact mode for (de)muxer and (de/en)coder
2161 -shortest (output)
2163 Finish encoding when the shortest input stream ends.
2164 -dts_delta_threshold
2166 Timestamp discontinuity delta threshold.
2167 -dts_error_threshold seconds
2169 Timestamp error delta threshold. This threshold use to discard
2170 crazy/damaged timestamps and the default is 30 hours which is
2171 arbitrarily picked and quite conservative.
2172 -muxdelay seconds (output)
2174 Set the maximum demux-decode delay.
2175 -muxpreload seconds (output)
2177 Set the initial demux-decode delay.
2178 -streamid output-stream-index:new-value (output)
2180 Assign a new stream-id value to an output stream. This option
2181 should be specified prior to the output filename to which it
2182 applies. For the situation where multiple output files exist, a
2183 streamid may be reassigned to a different value.
2184 For example, to set the stream 0 PID to 33 and the stream 1 PID to
2186 36 for an output mpegts file:
2187 ffmpeg -i inurl -streamid 0:33 -streamid 1:36 out.ts
2189 -bsf[:stream_specifier] bitstream_filters (output,per-stream)
2191 Set bitstream filters for matching streams. bitstream_filters is a
2192 comma-separated list of bitstream filters. Use the "-bsfs" option
2193 to get the list of bitstream filters.
2194 ffmpeg -i h264.mp4 -c:v copy -bsf:v h264_mp4toannexb -an out.h264
2196 ffmpeg -i file.mov -an -vn -bsf:s mov2textsub -c:s copy -f rawvideo sub.txt
2199 -tag[:stream_specifier] codec_tag (input/output,per-stream)
2201 Force a tag/fourcc for matching streams.
2202 -timecode hh:mm:ssSEPff
2204 Specify Timecode for writing. SEP is ':' for non drop timecode and
2205 ';' (or '.') for drop.
2206 ffmpeg -i input.mpg -timecode 01:02:03.04 -r 30000/1001 -s ntsc output.mpg
2208 -filter_complex filtergraph (global)
2210 Define a complex filtergraph, i.e. one with arbitrary number of
2211 inputs and/or outputs. For simple graphs -- those with one input
2212 and one output of the same type -- see the -filter options.
2213 filtergraph is a description of the filtergraph, as described in
2214 the ``Filtergraph syntax'' section of the ffmpeg-filters manual.
2215 Input link labels must refer to input streams using the
2217 "[file_index:stream_specifier]" syntax (i.e. the same as -map
2218 uses). If stream_specifier matches multiple streams, the first one
2219 will be used. An unlabeled input will be connected to the first
2220 unused input stream of the matching type.
2221 Output link labels are referred to with -map. Unlabeled outputs are
2223 added to the first output file.
2224 Note that with this option it is possible to use only lavfi sources
2226 without normal input files.
2227 For example, to overlay an image over video
2229 ffmpeg -i video.mkv -i image.png -filter_complex '[0:v][1:v]overlay[out]' -map
2231 '[out]' out.mkv
2232 Here "[0:v]" refers to the first video stream in the first input
2234 file, which is linked to the first (main) input of the overlay
2235 filter. Similarly the first video stream in the second input is
2236 linked to the second (overlay) input of overlay.
2237 Assuming there is only one video stream in each input file, we can
2239 omit input labels, so the above is equivalent to
2240 ffmpeg -i video.mkv -i image.png -filter_complex 'overlay[out]' -map
2242 '[out]' out.mkv
2243 Furthermore we can omit the output label and the single output from
2245 the filter graph will be added to the output file automatically, so
2246 we can simply write
2247 ffmpeg -i video.mkv -i image.png -filter_complex 'overlay' out.mkv
2249 As a special exception, you can use a bitmap subtitle stream as
2251 input: it will be converted into a video with the same size as the
2252 largest video in the file, or 720x576 if no video is present. Note
2253 that this is an experimental and temporary solution. It will be
2254 removed once libavfilter has proper support for subtitles.
2255 For example, to hardcode subtitles on top of a DVB-T recording
2257 stored in MPEG-TS format, delaying the subtitles by 1 second:
2258 ffmpeg -i input.ts -filter_complex \
2260 '[#0x2ef] setpts=PTS+1/TB [sub] ; [#0x2d0] [sub] overlay' \
2261 -sn -map '#0x2dc' output.mkv
2262 (0x2d0, 0x2dc and 0x2ef are the MPEG-TS PIDs of respectively the
2264 video, audio and subtitles streams; 0:0, 0:3 and 0:7 would have
2265 worked too)
2266 To generate 5 seconds of pure red video using lavfi "color" source:
2268 ffmpeg -filter_complex 'color=c=red' -t 5 out.mkv
2270 -filter_complex_threads nb_threads (global)
2272 Defines how many threads are used to process a filter_complex
2273 graph. Similar to filter_threads but used for "-filter_complex"
2274 graphs only. The default is the number of available CPUs.
2275 -lavfi filtergraph (global)
2277 Define a complex filtergraph, i.e. one with arbitrary number of
2278 inputs and/or outputs. Equivalent to -filter_complex.
2279 -filter_complex_script filename (global)
2281 This option is similar to -filter_complex, the only difference is
2282 that its argument is the name of the file from which a complex
2283 filtergraph description is to be read.
2284 -accurate_seek (input)
2286 This option enables or disables accurate seeking in input files
2287 with the -ss option. It is enabled by default, so seeking is
2288 accurate when transcoding. Use -noaccurate_seek to disable it,
2289 which may be useful e.g. when copying some streams and transcoding
2290 the others.
2291 -seek_timestamp (input)
2293 This option enables or disables seeking by timestamp in input files
2294 with the -ss option. It is disabled by default. If enabled, the
2295 argument to the -ss option is considered an actual timestamp, and
2296 is not offset by the start time of the file. This matters only for
2297 files which do not start from timestamp 0, such as transport
2298 streams.
2299 -thread_queue_size size (input)
2301 This option sets the maximum number of queued packets when reading
2302 from the file or device. With low latency / high rate live streams,
2303 packets may be discarded if they are not read in a timely manner;
2304 setting this value can force ffmpeg to use a separate input thread
2305 and read packets as soon as they arrive. By default ffmpeg only do
2306 this if multiple inputs are specified.
2307 -sdp_file file (global)
2309 Print sdp information for an output stream to file. This allows
2310 dumping sdp information when at least one output isn't an rtp
2311 stream. (Requires at least one of the output formats to be rtp).
2312 -discard (input)
2314 Allows discarding specific streams or frames from streams. Any
2315 input stream can be fully discarded, using value "all" whereas
2316 selective discarding of frames from a stream occurs at the demuxer
2317 and is not supported by all demuxers.
2318 none
2320 Discard no frame.
2321 default
2323 Default, which discards no frames.
2324 noref
2326 Discard all non-reference frames.
2327 bidir
2329 Discard all bidirectional frames.
2330 nokey
2332 Discard all frames excepts keyframes.
2333 all Discard all frames.
2335 -abort_on flags (global)
2337 Stop and abort on various conditions. The following flags are
2338 available:
2339 empty_output
2341 No packets were passed to the muxer, the output is empty.
2342 empty_output_stream
2344 No packets were passed to the muxer in some of the output
2345 streams.
2346 -max_error_rate (global)
2348 Set fraction of decoding frame failures across all inputs which
2349 when crossed ffmpeg will return exit code 69. Crossing this
2350 threshold does not terminate processing. Range is a floating-point
2351 number between 0 to 1. Default is 2/3.
2352 -xerror (global)
2354 Stop and exit on error
2355 -max_muxing_queue_size packets (output,per-stream)
2357 When transcoding audio and/or video streams, ffmpeg will not begin
2358 writing into the output until it has one packet for each such
2359 stream. While waiting for that to happen, packets for other streams
2360 are buffered. This option sets the size of this buffer, in packets,
2361 for the matching output stream.
2362 The default value of this option should be high enough for most
2364 uses, so only touch this option if you are sure that you need it.
2365 -muxing_queue_data_threshold bytes (output,per-stream)
2367 This is a minimum threshold until which the muxing queue size is
2368 not taken into account. Defaults to 50 megabytes per stream, and is
2369 based on the overall size of packets passed to the muxer.
2370 -auto_conversion_filters (global)
2372 Enable automatically inserting format conversion filters in all
2373 filter graphs, including those defined by -vf, -af, -filter_complex
2374 and -lavfi. If filter format negotiation requires a conversion, the
2375 initialization of the filters will fail. Conversions can still be
2376 performed by inserting the relevant conversion filter (scale,
2377 aresample) in the graph. On by default, to explicitly disable it
2378 you need to specify "-noauto_conversion_filters".
2379 -bits_per_raw_sample[:stream_specifier] value (output,per-stream)
2381 Declare the number of bits per raw sample in the given output
2382 stream to be value. Note that this option sets the information
2383 provided to the encoder/muxer, it does not change the stream to
2384 conform to this value. Setting values that do not match the stream
2385 properties may result in encoding failures or invalid output files.
2386 Preset files
2388 A preset file contains a sequence of option=value pairs, one for each
2389 line, specifying a sequence of options which would be awkward to
2390 specify on the command line. Lines starting with the hash ('#')
2391 character are ignored and are used to provide comments. Check the
2392 presets directory in the FFmpeg source tree for examples.
2393 There are two types of preset files: ffpreset and avpreset files.
2395 ffpreset files
2397 ffpreset files are specified with the "vpre", "apre", "spre", and
2399 "fpre" options. The "fpre" option takes the filename of the preset
2400 instead of a preset name as input and can be used for any kind of
2401 codec. For the "vpre", "apre", and "spre" options, the options
2402 specified in a preset file are applied to the currently selected codec
2403 of the same type as the preset option.
2404 The argument passed to the "vpre", "apre", and "spre" preset options
2406 identifies the preset file to use according to the following rules:
2407 First ffmpeg searches for a file named arg.ffpreset in the directories
2409 $FFMPEG_DATADIR (if set), and $HOME/.ffmpeg, and in the datadir defined
2410 at configuration time (usually PREFIX/share/ffmpeg) or in a ffpresets
2411 folder along the executable on win32, in that order. For example, if
2412 the argument is "libvpx-1080p", it will search for the file
2413 libvpx-1080p.ffpreset.
2414 If no such file is found, then ffmpeg will search for a file named
2416 codec_name-arg.ffpreset in the above-mentioned directories, where
2417 codec_name is the name of the codec to which the preset file options
2418 will be applied. For example, if you select the video codec with
2419 "-vcodec libvpx" and use "-vpre 1080p", then it will search for the
2420 file libvpx-1080p.ffpreset.
2421 avpreset files
2423 avpreset files are specified with the "pre" option. They work similar
2425 to ffpreset files, but they only allow encoder- specific options.
2426 Therefore, an option=value pair specifying an encoder cannot be used.
2427 When the "pre" option is specified, ffmpeg will look for files with the
2429 suffix .avpreset in the directories $AVCONV_DATADIR (if set), and
2430 $HOME/.avconv, and in the datadir defined at configuration time
2431 (usually PREFIX/share/ffmpeg), in that order.
2432 First ffmpeg searches for a file named codec_name-arg.avpreset in the
2434 above-mentioned directories, where codec_name is the name of the codec
2435 to which the preset file options will be applied. For example, if you
2436 select the video codec with "-vcodec libvpx" and use "-pre 1080p", then
2437 it will search for the file libvpx-1080p.avpreset.
2438 If no such file is found, then ffmpeg will search for a file named
2440 arg.avpreset in the same directories.
2441
2443 Video and Audio grabbing
2444 If you specify the input format and device then ffmpeg can grab video
2445 and audio directly.
2446 ffmpeg -f oss -i /dev/dsp -f video4linux2 -i /dev/video0 /tmp/out.mpg
2448 Or with an ALSA audio source (mono input, card id 1) instead of OSS:
2450 ffmpeg -f alsa -ac 1 -i hw:1 -f video4linux2 -i /dev/video0 /tmp/out.mpg
2452 Note that you must activate the right video source and channel before
2454 launching ffmpeg with any TV viewer such as
2455 <http://linux.bytesex.org/xawtv/> by Gerd Knorr. You also have to set
2456 the audio recording levels correctly with a standard mixer.
2457 X11 grabbing
2459 Grab the X11 display with ffmpeg via
2460 ffmpeg -f x11grab -video_size cif -framerate 25 -i :0.0 /tmp/out.mpg
2462 0.0 is display.screen number of your X11 server, same as the DISPLAY
2464 environment variable.
2465 ffmpeg -f x11grab -video_size cif -framerate 25 -i :0.0+10,20 /tmp/out.mpg
2467 0.0 is display.screen number of your X11 server, same as the DISPLAY
2469 environment variable. 10 is the x-offset and 20 the y-offset for the
2470 grabbing.
2471 Video and Audio file format conversion
2473 Any supported file format and protocol can serve as input to ffmpeg:
2474 Examples:
2476 • You can use YUV files as input:
2478 ffmpeg -i /tmp/test%d.Y /tmp/out.mpg
2480 It will use the files:
2482 /tmp/test0.Y, /tmp/test0.U, /tmp/test0.V,
2484 /tmp/test1.Y, /tmp/test1.U, /tmp/test1.V, etc...
2485 The Y files use twice the resolution of the U and V files. They are
2487 raw files, without header. They can be generated by all decent
2488 video decoders. You must specify the size of the image with the -s
2489 option if ffmpeg cannot guess it.
2490 • You can input from a raw YUV420P file:
2492 ffmpeg -i /tmp/test.yuv /tmp/out.avi
2494 test.yuv is a file containing raw YUV planar data. Each frame is
2496 composed of the Y plane followed by the U and V planes at half
2497 vertical and horizontal resolution.
2498 • You can output to a raw YUV420P file:
2500 ffmpeg -i mydivx.avi hugefile.yuv
2502 • You can set several input files and output files:
2504 ffmpeg -i /tmp/a.wav -s 640x480 -i /tmp/a.yuv /tmp/a.mpg
2506 Converts the audio file a.wav and the raw YUV video file a.yuv to
2508 MPEG file a.mpg.
2509 • You can also do audio and video conversions at the same time:
2511 ffmpeg -i /tmp/a.wav -ar 22050 /tmp/a.mp2
2513 Converts a.wav to MPEG audio at 22050 Hz sample rate.
2515 • You can encode to several formats at the same time and define a
2517 mapping from input stream to output streams:
2518 ffmpeg -i /tmp/a.wav -map 0:a -b:a 64k /tmp/a.mp2 -map 0:a -b:a 128k /tmp/b.mp2
2520 Converts a.wav to a.mp2 at 64 kbits and to b.mp2 at 128 kbits.
2522 '-map file:index' specifies which input stream is used for each
2523 output stream, in the order of the definition of output streams.
2524 • You can transcode decrypted VOBs:
2526 ffmpeg -i snatch_1.vob -f avi -c:v mpeg4 -b:v 800k -g 300 -bf 2 -c:a libmp3lame -b:a 128k snatch.avi
2528 This is a typical DVD ripping example; the input is a VOB file, the
2530 output an AVI file with MPEG-4 video and MP3 audio. Note that in
2531 this command we use B-frames so the MPEG-4 stream is DivX5
2532 compatible, and GOP size is 300 which means one intra frame every
2533 10 seconds for 29.97fps input video. Furthermore, the audio stream
2534 is MP3-encoded so you need to enable LAME support by passing
2535 "--enable-libmp3lame" to configure. The mapping is particularly
2536 useful for DVD transcoding to get the desired audio language.
2537 NOTE: To see the supported input formats, use "ffmpeg -demuxers".
2539 • You can extract images from a video, or create a video from many
2541 images:
2542 For extracting images from a video:
2544 ffmpeg -i foo.avi -r 1 -s WxH -f image2 foo-%03d.jpeg
2546 This will extract one video frame per second from the video and
2548 will output them in files named foo-001.jpeg, foo-002.jpeg, etc.
2549 Images will be rescaled to fit the new WxH values.
2550 If you want to extract just a limited number of frames, you can use
2552 the above command in combination with the "-frames:v" or "-t"
2553 option, or in combination with -ss to start extracting from a
2554 certain point in time.
2555 For creating a video from many images:
2557 ffmpeg -f image2 -framerate 12 -i foo-%03d.jpeg -s WxH foo.avi
2559 The syntax "foo-%03d.jpeg" specifies to use a decimal number
2561 composed of three digits padded with zeroes to express the sequence
2562 number. It is the same syntax supported by the C printf function,
2563 but only formats accepting a normal integer are suitable.
2564 When importing an image sequence, -i also supports expanding shell-
2566 like wildcard patterns (globbing) internally, by selecting the
2567 image2-specific "-pattern_type glob" option.
2568 For example, for creating a video from filenames matching the glob
2570 pattern "foo-*.jpeg":
2571 ffmpeg -f image2 -pattern_type glob -framerate 12 -i 'foo-*.jpeg' -s WxH foo.avi
2573 • You can put many streams of the same type in the output:
2575 ffmpeg -i test1.avi -i test2.avi -map 1:1 -map 1:0 -map 0:1 -map 0:0 -c copy -y test12.nut
2577 The resulting output file test12.nut will contain the first four
2579 streams from the input files in reverse order.
2580 • To force CBR video output:
2582 ffmpeg -i myfile.avi -b 4000k -minrate 4000k -maxrate 4000k -bufsize 1835k out.m2v
2584 • The four options lmin, lmax, mblmin and mblmax use 'lambda' units,
2586 but you may use the QP2LAMBDA constant to easily convert from 'q'
2587 units:
2588 ffmpeg -i src.ext -lmax 21*QP2LAMBDA dst.ext
2590
2592 This section documents the syntax and formats employed by the FFmpeg
2593 libraries and tools.
2594 Quoting and escaping
2596 FFmpeg adopts the following quoting and escaping mechanism, unless
2597 explicitly specified. The following rules are applied:
2598 • ' and \ are special characters (respectively used for quoting and
2600 escaping). In addition to them, there might be other special
2601 characters depending on the specific syntax where the escaping and
2602 quoting are employed.
2603 • A special character is escaped by prefixing it with a \.
2605 • All characters enclosed between '' are included literally in the
2607 parsed string. The quote character ' itself cannot be quoted, so
2608 you may need to close the quote and escape it.
2609 • Leading and trailing whitespaces, unless escaped or quoted, are
2611 removed from the parsed string.
2612 Note that you may need to add a second level of escaping when using the
2614 command line or a script, which depends on the syntax of the adopted
2615 shell language.
2616 The function "av_get_token" defined in libavutil/avstring.h can be used
2618 to parse a token quoted or escaped according to the rules defined
2619 above.
2620 The tool tools/ffescape in the FFmpeg source tree can be used to
2622 automatically quote or escape a string in a script.
2623 Examples
2625 • Escape the string "Crime d'Amour" containing the "'" special
2627 character:
2628 Crime d\'Amour
2630 • The string above contains a quote, so the "'" needs to be escaped
2632 when quoting it:
2633 'Crime d'\''Amour'
2635 • Include leading or trailing whitespaces using quoting:
2637 ' this string starts and ends with whitespaces '
2639 • Escaping and quoting can be mixed together:
2641 ' The string '\'string\'' is a string '
2643 • To include a literal \ you can use either escaping or quoting:
2645 'c:\foo' can be written as c:\\foo
2647 Date
2649 The accepted syntax is:
2650 [(YYYY-MM-DD|YYYYMMDD)[T|t| ]]((HH:MM:SS[.m...]]])|(HHMMSS[.m...]]]))[Z]
2652 now
2653 If the value is "now" it takes the current time.
2655 Time is local time unless Z is appended, in which case it is
2657 interpreted as UTC. If the year-month-day part is not specified it
2658 takes the current year-month-day.
2659 Time duration
2661 There are two accepted syntaxes for expressing time duration.
2662 [-][<HH>:]<MM>:<SS>[.<m>...]
2664 HH expresses the number of hours, MM the number of minutes for a
2666 maximum of 2 digits, and SS the number of seconds for a maximum of 2
2667 digits. The m at the end expresses decimal value for SS.
2668 or
2670 [-]<S>+[.<m>...][s|ms|us]
2672 S expresses the number of seconds, with the optional decimal part m.
2674 The optional literal suffixes s, ms or us indicate to interpret the
2675 value as seconds, milliseconds or microseconds, respectively.
2676 In both expressions, the optional - indicates negative duration.
2678 Examples
2680 The following examples are all valid time duration:
2682 55 55 seconds
2684 0.2 0.2 seconds
2686 200ms
2688 200 milliseconds, that's 0.2s
2689 200000us
2691 200000 microseconds, that's 0.2s
2692 12:03:45
2694 12 hours, 03 minutes and 45 seconds
2695 23.189
2697 23.189 seconds
2698 Video size
2700 Specify the size of the sourced video, it may be a string of the form
2701 widthxheight, or the name of a size abbreviation.
2702 The following abbreviations are recognized:
2704 ntsc
2706 720x480
2707 pal 720x576
2709 qntsc
2711 352x240
2712 qpal
2714 352x288
2715 sntsc
2717 640x480
2718 spal
2720 768x576
2721 film
2723 352x240
2724 ntsc-film
2726 352x240
2727 sqcif
2729 128x96
2730 qcif
2732 176x144
2733 cif 352x288
2735 4cif
2737 704x576
2738 16cif
2740 1408x1152
2741 qqvga
2743 160x120
2744 qvga
2746 320x240
2747 vga 640x480
2749 svga
2751 800x600
2752 xga 1024x768
2754 uxga
2756 1600x1200
2757 qxga
2759 2048x1536
2760 sxga
2762 1280x1024
2763 qsxga
2765 2560x2048
2766 hsxga
2768 5120x4096
2769 wvga
2771 852x480
2772 wxga
2774 1366x768
2775 wsxga
2777 1600x1024
2778 wuxga
2780 1920x1200
2781 woxga
2783 2560x1600
2784 wqsxga
2786 3200x2048
2787 wquxga
2789 3840x2400
2790 whsxga
2792 6400x4096
2793 whuxga
2795 7680x4800
2796 cga 320x200
2798 ega 640x350
2800 hd480
2802 852x480
2803 hd720
2805 1280x720
2806 hd1080
2808 1920x1080
2809 2k 2048x1080
2811 2kflat
2813 1998x1080
2814 2kscope
2816 2048x858
2817 4k 4096x2160
2819 4kflat
2821 3996x2160
2822 4kscope
2824 4096x1716
2825 nhd 640x360
2827 hqvga
2829 240x160
2830 wqvga
2832 400x240
2833 fwqvga
2835 432x240
2836 hvga
2838 480x320
2839 qhd 960x540
2841 2kdci
2843 2048x1080
2844 4kdci
2846 4096x2160
2847 uhd2160
2849 3840x2160
2850 uhd4320
2852 7680x4320
2853 Video rate
2855 Specify the frame rate of a video, expressed as the number of frames
2856 generated per second. It has to be a string in the format
2857 frame_rate_num/frame_rate_den, an integer number, a float number or a
2858 valid video frame rate abbreviation.
2859 The following abbreviations are recognized:
2861 ntsc
2863 30000/1001
2864 pal 25/1
2866 qntsc
2868 30000/1001
2869 qpal
2871 25/1
2872 sntsc
2874 30000/1001
2875 spal
2877 25/1
2878 film
2880 24/1
2881 ntsc-film
2883 24000/1001
2884 Ratio
2886 A ratio can be expressed as an expression, or in the form
2887 numerator:denominator.
2888 Note that a ratio with infinite (1/0) or negative value is considered
2890 valid, so you should check on the returned value if you want to exclude
2891 those values.
2892 The undefined value can be expressed using the "0:0" string.
2894 Color
2896 It can be the name of a color as defined below (case insensitive match)
2897 or a "[0x|#]RRGGBB[AA]" sequence, possibly followed by @ and a string
2898 representing the alpha component.
2899 The alpha component may be a string composed by "0x" followed by an
2901 hexadecimal number or a decimal number between 0.0 and 1.0, which
2902 represents the opacity value (0x00 or 0.0 means completely transparent,
2903 0xff or 1.0 completely opaque). If the alpha component is not specified
2904 then 0xff is assumed.
2905 The string random will result in a random color.
2907 The following names of colors are recognized:
2909 AliceBlue
2911 0xF0F8FF
2912 AntiqueWhite
2914 0xFAEBD7
2915 Aqua
2917 0x00FFFF
2918 Aquamarine
2920 0x7FFFD4
2921 Azure
2923 0xF0FFFF
2924 Beige
2926 0xF5F5DC
2927 Bisque
2929 0xFFE4C4
2930 Black
2932 0x000000
2933 BlanchedAlmond
2935 0xFFEBCD
2936 Blue
2938 0x0000FF
2939 BlueViolet
2941 0x8A2BE2
2942 Brown
2944 0xA52A2A
2945 BurlyWood
2947 0xDEB887
2948 CadetBlue
2950 0x5F9EA0
2951 Chartreuse
2953 0x7FFF00
2954 Chocolate
2956 0xD2691E
2957 Coral
2959 0xFF7F50
2960 CornflowerBlue
2962 0x6495ED
2963 Cornsilk
2965 0xFFF8DC
2966 Crimson
2968 0xDC143C
2969 Cyan
2971 0x00FFFF
2972 DarkBlue
2974 0x00008B
2975 DarkCyan
2977 0x008B8B
2978 DarkGoldenRod
2980 0xB8860B
2981 DarkGray
2983 0xA9A9A9
2984 DarkGreen
2986 0x006400
2987 DarkKhaki
2989 0xBDB76B
2990 DarkMagenta
2992 0x8B008B
2993 DarkOliveGreen
2995 0x556B2F
2996 Darkorange
2998 0xFF8C00
2999 DarkOrchid
3001 0x9932CC
3002 DarkRed
3004 0x8B0000
3005 DarkSalmon
3007 0xE9967A
3008 DarkSeaGreen
3010 0x8FBC8F
3011 DarkSlateBlue
3013 0x483D8B
3014 DarkSlateGray
3016 0x2F4F4F
3017 DarkTurquoise
3019 0x00CED1
3020 DarkViolet
3022 0x9400D3
3023 DeepPink
3025 0xFF1493
3026 DeepSkyBlue
3028 0x00BFFF
3029 DimGray
3031 0x696969
3032 DodgerBlue
3034 0x1E90FF
3035 FireBrick
3037 0xB22222
3038 FloralWhite
3040 0xFFFAF0
3041 ForestGreen
3043 0x228B22
3044 Fuchsia
3046 0xFF00FF
3047 Gainsboro
3049 0xDCDCDC
3050 GhostWhite
3052 0xF8F8FF
3053 Gold
3055 0xFFD700
3056 GoldenRod
3058 0xDAA520
3059 Gray
3061 0x808080
3062 Green
3064 0x008000
3065 GreenYellow
3067 0xADFF2F
3068 HoneyDew
3070 0xF0FFF0
3071 HotPink
3073 0xFF69B4
3074 IndianRed
3076 0xCD5C5C
3077 Indigo
3079 0x4B0082
3080 Ivory
3082 0xFFFFF0
3083 Khaki
3085 0xF0E68C
3086 Lavender
3088 0xE6E6FA
3089 LavenderBlush
3091 0xFFF0F5
3092 LawnGreen
3094 0x7CFC00
3095 LemonChiffon
3097 0xFFFACD
3098 LightBlue
3100 0xADD8E6
3101 LightCoral
3103 0xF08080
3104 LightCyan
3106 0xE0FFFF
3107 LightGoldenRodYellow
3109 0xFAFAD2
3110 LightGreen
3112 0x90EE90
3113 LightGrey
3115 0xD3D3D3
3116 LightPink
3118 0xFFB6C1
3119 LightSalmon
3121 0xFFA07A
3122 LightSeaGreen
3124 0x20B2AA
3125 LightSkyBlue
3127 0x87CEFA
3128 LightSlateGray
3130 0x778899
3131 LightSteelBlue
3133 0xB0C4DE
3134 LightYellow
3136 0xFFFFE0
3137 Lime
3139 0x00FF00
3140 LimeGreen
3142 0x32CD32
3143 Linen
3145 0xFAF0E6
3146 Magenta
3148 0xFF00FF
3149 Maroon
3151 0x800000
3152 MediumAquaMarine
3154 0x66CDAA
3155 MediumBlue
3157 0x0000CD
3158 MediumOrchid
3160 0xBA55D3
3161 MediumPurple
3163 0x9370D8
3164 MediumSeaGreen
3166 0x3CB371
3167 MediumSlateBlue
3169 0x7B68EE
3170 MediumSpringGreen
3172 0x00FA9A
3173 MediumTurquoise
3175 0x48D1CC
3176 MediumVioletRed
3178 0xC71585
3179 MidnightBlue
3181 0x191970
3182 MintCream
3184 0xF5FFFA
3185 MistyRose
3187 0xFFE4E1
3188 Moccasin
3190 0xFFE4B5
3191 NavajoWhite
3193 0xFFDEAD
3194 Navy
3196 0x000080
3197 OldLace
3199 0xFDF5E6
3200 Olive
3202 0x808000
3203 OliveDrab
3205 0x6B8E23
3206 Orange
3208 0xFFA500
3209 OrangeRed
3211 0xFF4500
3212 Orchid
3214 0xDA70D6
3215 PaleGoldenRod
3217 0xEEE8AA
3218 PaleGreen
3220 0x98FB98
3221 PaleTurquoise
3223 0xAFEEEE
3224 PaleVioletRed
3226 0xD87093
3227 PapayaWhip
3229 0xFFEFD5
3230 PeachPuff
3232 0xFFDAB9
3233 Peru
3235 0xCD853F
3236 Pink
3238 0xFFC0CB
3239 Plum
3241 0xDDA0DD
3242 PowderBlue
3244 0xB0E0E6
3245 Purple
3247 0x800080
3248 Red 0xFF0000
3250 RosyBrown
3252 0xBC8F8F
3253 RoyalBlue
3255 0x4169E1
3256 SaddleBrown
3258 0x8B4513
3259 Salmon
3261 0xFA8072
3262 SandyBrown
3264 0xF4A460
3265 SeaGreen
3267 0x2E8B57
3268 SeaShell
3270 0xFFF5EE
3271 Sienna
3273 0xA0522D
3274 Silver
3276 0xC0C0C0
3277 SkyBlue
3279 0x87CEEB
3280 SlateBlue
3282 0x6A5ACD
3283 SlateGray
3285 0x708090
3286 Snow
3288 0xFFFAFA
3289 SpringGreen
3291 0x00FF7F
3292 SteelBlue
3294 0x4682B4
3295 Tan 0xD2B48C
3297 Teal
3299 0x008080
3300 Thistle
3302 0xD8BFD8
3303 Tomato
3305 0xFF6347
3306 Turquoise
3308 0x40E0D0
3309 Violet
3311 0xEE82EE
3312 Wheat
3314 0xF5DEB3
3315 White
3317 0xFFFFFF
3318 WhiteSmoke
3320 0xF5F5F5
3321 Yellow
3323 0xFFFF00
3324 YellowGreen
3326 0x9ACD32
3327 Channel Layout
3329 A channel layout specifies the spatial disposition of the channels in a
3330 multi-channel audio stream. To specify a channel layout, FFmpeg makes
3331 use of a special syntax.
3332 Individual channels are identified by an id, as given by the table
3334 below:
3335 FL front left
3337 FR front right
3339 FC front center
3341 LFE low frequency
3343 BL back left
3345 BR back right
3347 FLC front left-of-center
3349 FRC front right-of-center
3351 BC back center
3353 SL side left
3355 SR side right
3357 TC top center
3359 TFL top front left
3361 TFC top front center
3363 TFR top front right
3365 TBL top back left
3367 TBC top back center
3369 TBR top back right
3371 DL downmix left
3373 DR downmix right
3375 WL wide left
3377 WR wide right
3379 SDL surround direct left
3381 SDR surround direct right
3383 LFE2
3385 low frequency 2
3386 Standard channel layout compositions can be specified by using the
3388 following identifiers:
3389 mono
3391 FC
3392 stereo
3394 FL+FR
3395 2.1 FL+FR+LFE
3397 3.0 FL+FR+FC
3399 3.0(back)
3401 FL+FR+BC
3402 4.0 FL+FR+FC+BC
3404 quad
3406 FL+FR+BL+BR
3407 quad(side)
3409 FL+FR+SL+SR
3410 3.1 FL+FR+FC+LFE
3412 5.0 FL+FR+FC+BL+BR
3414 5.0(side)
3416 FL+FR+FC+SL+SR
3417 4.1 FL+FR+FC+LFE+BC
3419 5.1 FL+FR+FC+LFE+BL+BR
3421 5.1(side)
3423 FL+FR+FC+LFE+SL+SR
3424 6.0 FL+FR+FC+BC+SL+SR
3426 6.0(front)
3428 FL+FR+FLC+FRC+SL+SR
3429 hexagonal
3431 FL+FR+FC+BL+BR+BC
3432 6.1 FL+FR+FC+LFE+BC+SL+SR
3434 6.1 FL+FR+FC+LFE+BL+BR+BC
3436 6.1(front)
3438 FL+FR+LFE+FLC+FRC+SL+SR
3439 7.0 FL+FR+FC+BL+BR+SL+SR
3441 7.0(front)
3443 FL+FR+FC+FLC+FRC+SL+SR
3444 7.1 FL+FR+FC+LFE+BL+BR+SL+SR
3446 7.1(wide)
3448 FL+FR+FC+LFE+BL+BR+FLC+FRC
3449 7.1(wide-side)
3451 FL+FR+FC+LFE+FLC+FRC+SL+SR
3452 octagonal
3454 FL+FR+FC+BL+BR+BC+SL+SR
3455 hexadecagonal
3457 FL+FR+FC+BL+BR+BC+SL+SR+WL+WR+TBL+TBR+TBC+TFC+TFL+TFR
3458 downmix
3460 DL+DR
3461 22.2
3463 FL+FR+FC+LFE+BL+BR+FLC+FRC+BC+SL+SR+TC+TFL+TFC+TFR+TBL+TBC+TBR+LFE2+TSL+TSR+BFC+BFL+BFR
3464 A custom channel layout can be specified as a sequence of terms,
3466 separated by '+' or '|'. Each term can be:
3467 • the name of a standard channel layout (e.g. mono, stereo, 4.0,
3469 quad, 5.0, etc.)
3470 • the name of a single channel (e.g. FL, FR, FC, LFE, etc.)
3472 • a number of channels, in decimal, followed by 'c', yielding the
3474 default channel layout for that number of channels (see the
3475 function "av_get_default_channel_layout"). Note that not all
3476 channel counts have a default layout.
3477 • a number of channels, in decimal, followed by 'C', yielding an
3479 unknown channel layout with the specified number of channels. Note
3480 that not all channel layout specification strings support unknown
3481 channel layouts.
3482 • a channel layout mask, in hexadecimal starting with "0x" (see the
3484 "AV_CH_*" macros in libavutil/channel_layout.h.
3485 Before libavutil version 53 the trailing character "c" to specify a
3487 number of channels was optional, but now it is required, while a
3488 channel layout mask can also be specified as a decimal number (if and
3489 only if not followed by "c" or "C").
3490 See also the function "av_get_channel_layout" defined in
3492 libavutil/channel_layout.h.
3493
3495 When evaluating an arithmetic expression, FFmpeg uses an internal
3496 formula evaluator, implemented through the libavutil/eval.h interface.
3497 An expression may contain unary, binary operators, constants, and
3499 functions.
3500 Two expressions expr1 and expr2 can be combined to form another
3502 expression "expr1;expr2". expr1 and expr2 are evaluated in turn, and
3503 the new expression evaluates to the value of expr2.
3504 The following binary operators are available: "+", "-", "*", "/", "^".
3506 The following unary operators are available: "+", "-".
3508 The following functions are available:
3510 abs(x)
3512 Compute absolute value of x.
3513 acos(x)
3515 Compute arccosine of x.
3516 asin(x)
3518 Compute arcsine of x.
3519 atan(x)
3521 Compute arctangent of x.
3522 atan2(x, y)
3524 Compute principal value of the arc tangent of y/x.
3525 between(x, min, max)
3527 Return 1 if x is greater than or equal to min and lesser than or
3528 equal to max, 0 otherwise.
3529 bitand(x, y)
3531 bitor(x, y)
3532 Compute bitwise and/or operation on x and y.
3533 The results of the evaluation of x and y are converted to integers
3535 before executing the bitwise operation.
3536 Note that both the conversion to integer and the conversion back to
3538 floating point can lose precision. Beware of unexpected results for
3539 large numbers (usually 2^53 and larger).
3540 ceil(expr)
3542 Round the value of expression expr upwards to the nearest integer.
3543 For example, "ceil(1.5)" is "2.0".
3544 clip(x, min, max)
3546 Return the value of x clipped between min and max.
3547 cos(x)
3549 Compute cosine of x.
3550 cosh(x)
3552 Compute hyperbolic cosine of x.
3553 eq(x, y)
3555 Return 1 if x and y are equivalent, 0 otherwise.
3556 exp(x)
3558 Compute exponential of x (with base "e", the Euler's number).
3559 floor(expr)
3561 Round the value of expression expr downwards to the nearest
3562 integer. For example, "floor(-1.5)" is "-2.0".
3563 gauss(x)
3565 Compute Gauss function of x, corresponding to "exp(-x*x/2) /
3566 sqrt(2*PI)".
3567 gcd(x, y)
3569 Return the greatest common divisor of x and y. If both x and y are
3570 0 or either or both are less than zero then behavior is undefined.
3571 gt(x, y)
3573 Return 1 if x is greater than y, 0 otherwise.
3574 gte(x, y)
3576 Return 1 if x is greater than or equal to y, 0 otherwise.
3577 hypot(x, y)
3579 This function is similar to the C function with the same name; it
3580 returns "sqrt(x*x + y*y)", the length of the hypotenuse of a right
3581 triangle with sides of length x and y, or the distance of the point
3582 (x, y) from the origin.
3583 if(x, y)
3585 Evaluate x, and if the result is non-zero return the result of the
3586 evaluation of y, return 0 otherwise.
3587 if(x, y, z)
3589 Evaluate x, and if the result is non-zero return the evaluation
3590 result of y, otherwise the evaluation result of z.
3591 ifnot(x, y)
3593 Evaluate x, and if the result is zero return the result of the
3594 evaluation of y, return 0 otherwise.
3595 ifnot(x, y, z)
3597 Evaluate x, and if the result is zero return the evaluation result
3598 of y, otherwise the evaluation result of z.
3599 isinf(x)
3601 Return 1.0 if x is +/-INFINITY, 0.0 otherwise.
3602 isnan(x)
3604 Return 1.0 if x is NAN, 0.0 otherwise.
3605 ld(var)
3607 Load the value of the internal variable with number var, which was
3608 previously stored with st(var, expr). The function returns the
3609 loaded value.
3610 lerp(x, y, z)
3612 Return linear interpolation between x and y by amount of z.
3613 log(x)
3615 Compute natural logarithm of x.
3616 lt(x, y)
3618 Return 1 if x is lesser than y, 0 otherwise.
3619 lte(x, y)
3621 Return 1 if x is lesser than or equal to y, 0 otherwise.
3622 max(x, y)
3624 Return the maximum between x and y.
3625 min(x, y)
3627 Return the minimum between x and y.
3628 mod(x, y)
3630 Compute the remainder of division of x by y.
3631 not(expr)
3633 Return 1.0 if expr is zero, 0.0 otherwise.
3634 pow(x, y)
3636 Compute the power of x elevated y, it is equivalent to "(x)^(y)".
3637 print(t)
3639 print(t, l)
3640 Print the value of expression t with loglevel l. If l is not
3641 specified then a default log level is used. Returns the value of
3642 the expression printed.
3643 Prints t with loglevel l
3645 random(x)
3647 Return a pseudo random value between 0.0 and 1.0. x is the index of
3648 the internal variable which will be used to save the seed/state.
3649 root(expr, max)
3651 Find an input value for which the function represented by expr with
3652 argument ld(0) is 0 in the interval 0..max.
3653 The expression in expr must denote a continuous function or the
3655 result is undefined.
3656 ld(0) is used to represent the function input value, which means
3658 that the given expression will be evaluated multiple times with
3659 various input values that the expression can access through ld(0).
3660 When the expression evaluates to 0 then the corresponding input
3661 value will be returned.
3662 round(expr)
3664 Round the value of expression expr to the nearest integer. For
3665 example, "round(1.5)" is "2.0".
3666 sgn(x)
3668 Compute sign of x.
3669 sin(x)
3671 Compute sine of x.
3672 sinh(x)
3674 Compute hyperbolic sine of x.
3675 sqrt(expr)
3677 Compute the square root of expr. This is equivalent to "(expr)^.5".
3678 squish(x)
3680 Compute expression "1/(1 + exp(4*x))".
3681 st(var, expr)
3683 Store the value of the expression expr in an internal variable. var
3684 specifies the number of the variable where to store the value, and
3685 it is a value ranging from 0 to 9. The function returns the value
3686 stored in the internal variable. Note, Variables are currently not
3687 shared between expressions.
3688 tan(x)
3690 Compute tangent of x.
3691 tanh(x)
3693 Compute hyperbolic tangent of x.
3694 taylor(expr, x)
3696 taylor(expr, x, id)
3697 Evaluate a Taylor series at x, given an expression representing the
3698 "ld(id)"-th derivative of a function at 0.
3699 When the series does not converge the result is undefined.
3701 ld(id) is used to represent the derivative order in expr, which
3703 means that the given expression will be evaluated multiple times
3704 with various input values that the expression can access through
3705 "ld(id)". If id is not specified then 0 is assumed.
3706 Note, when you have the derivatives at y instead of 0,
3708 "taylor(expr, x-y)" can be used.
3709 time(0)
3711 Return the current (wallclock) time in seconds.
3712 trunc(expr)
3714 Round the value of expression expr towards zero to the nearest
3715 integer. For example, "trunc(-1.5)" is "-1.0".
3716 while(cond, expr)
3718 Evaluate expression expr while the expression cond is non-zero, and
3719 returns the value of the last expr evaluation, or NAN if cond was
3720 always false.
3721 The following constants are available:
3723 PI area of the unit disc, approximately 3.14
3725 E exp(1) (Euler's number), approximately 2.718
3727 PHI golden ratio (1+sqrt(5))/2, approximately 1.618
3729 Assuming that an expression is considered "true" if it has a non-zero
3731 value, note that:
3732 "*" works like AND
3734 "+" works like OR
3736 For example the construct:
3738 if (A AND B) then C
3740 is equivalent to:
3742 if(A*B, C)
3744 In your C code, you can extend the list of unary and binary functions,
3746 and define recognized constants, so that they are available for your
3747 expressions.
3748 The evaluator also recognizes the International System unit prefixes.
3750 If 'i' is appended after the prefix, binary prefixes are used, which
3751 are based on powers of 1024 instead of powers of 1000. The 'B' postfix
3752 multiplies the value by 8, and can be appended after a unit prefix or
3753 used alone. This allows using for example 'KB', 'MiB', 'G' and 'B' as
3754 number postfix.
3755 The list of available International System prefixes follows, with
3757 indication of the corresponding powers of 10 and of 2.
3758 y 10^-24 / 2^-80
3760 z 10^-21 / 2^-70
3762 a 10^-18 / 2^-60
3764 f 10^-15 / 2^-50
3766 p 10^-12 / 2^-40
3768 n 10^-9 / 2^-30
3770 u 10^-6 / 2^-20
3772 m 10^-3 / 2^-10
3774 c 10^-2
3776 d 10^-1
3778 h 10^2
3780 k 10^3 / 2^10
3782 K 10^3 / 2^10
3784 M 10^6 / 2^20
3786 G 10^9 / 2^30
3788 T 10^12 / 2^40
3790 P 10^15 / 2^40
3792 E 10^18 / 2^50
3794 Z 10^21 / 2^60
3796 Y 10^24 / 2^70
3798
3800 libavcodec provides some generic global options, which can be set on
3801 all the encoders and decoders. In addition each codec may support so-
3802 called private options, which are specific for a given codec.
3803 Sometimes, a global option may only affect a specific kind of codec,
3805 and may be nonsensical or ignored by another, so you need to be aware
3806 of the meaning of the specified options. Also some options are meant
3807 only for decoding or encoding.
3808 Options may be set by specifying -option value in the FFmpeg tools, or
3810 by setting the value explicitly in the "AVCodecContext" options or
3811 using the libavutil/opt.h API for programmatic use.
3812 The list of supported options follow:
3814 b integer (encoding,audio,video)
3816 Set bitrate in bits/s. Default value is 200K.
3817 ab integer (encoding,audio)
3819 Set audio bitrate (in bits/s). Default value is 128K.
3820 bt integer (encoding,video)
3822 Set video bitrate tolerance (in bits/s). In 1-pass mode, bitrate
3823 tolerance specifies how far ratecontrol is willing to deviate from
3824 the target average bitrate value. This is not related to min/max
3825 bitrate. Lowering tolerance too much has an adverse effect on
3826 quality.
3827 flags flags (decoding/encoding,audio,video,subtitles)
3829 Set generic flags.
3830 Possible values:
3832 mv4 Use four motion vector by macroblock (mpeg4).
3834 qpel
3836 Use 1/4 pel motion compensation.
3837 loop
3839 Use loop filter.
3840 qscale
3842 Use fixed qscale.
3843 pass1
3845 Use internal 2pass ratecontrol in first pass mode.
3846 pass2
3848 Use internal 2pass ratecontrol in second pass mode.
3849 gray
3851 Only decode/encode grayscale.
3852 psnr
3854 Set error[?] variables during encoding.
3855 truncated
3857 Input bitstream might be randomly truncated.
3858 drop_changed
3860 Don't output frames whose parameters differ from first decoded
3861 frame in stream. Error AVERROR_INPUT_CHANGED is returned when
3862 a frame is dropped.
3863 ildct
3865 Use interlaced DCT.
3866 low_delay
3868 Force low delay.
3869 global_header
3871 Place global headers in extradata instead of every keyframe.
3872 bitexact
3874 Only write platform-, build- and time-independent data. (except
3875 (I)DCT). This ensures that file and data checksums are
3876 reproducible and match between platforms. Its primary use is
3877 for regression testing.
3878 aic Apply H263 advanced intra coding / mpeg4 ac prediction.
3880 ilme
3882 Apply interlaced motion estimation.
3883 cgop
3885 Use closed gop.
3886 output_corrupt
3888 Output even potentially corrupted frames.
3889 time_base rational number
3891 Set codec time base.
3892 It is the fundamental unit of time (in seconds) in terms of which
3894 frame timestamps are represented. For fixed-fps content, timebase
3895 should be "1 / frame_rate" and timestamp increments should be
3896 identically 1.
3897 g integer (encoding,video)
3899 Set the group of picture (GOP) size. Default value is 12.
3900 ar integer (decoding/encoding,audio)
3902 Set audio sampling rate (in Hz).
3903 ac integer (decoding/encoding,audio)
3905 Set number of audio channels.
3906 cutoff integer (encoding,audio)
3908 Set cutoff bandwidth. (Supported only by selected encoders, see
3909 their respective documentation sections.)
3910 frame_size integer (encoding,audio)
3912 Set audio frame size.
3913 Each submitted frame except the last must contain exactly
3915 frame_size samples per channel. May be 0 when the codec has
3916 CODEC_CAP_VARIABLE_FRAME_SIZE set, in that case the frame size is
3917 not restricted. It is set by some decoders to indicate constant
3918 frame size.
3919 frame_number integer
3921 Set the frame number.
3922 delay integer
3924 qcomp float (encoding,video)
3925 Set video quantizer scale compression (VBR). It is used as a
3926 constant in the ratecontrol equation. Recommended range for default
3927 rc_eq: 0.0-1.0.
3928 qblur float (encoding,video)
3930 Set video quantizer scale blur (VBR).
3931 qmin integer (encoding,video)
3933 Set min video quantizer scale (VBR). Must be included between -1
3934 and 69, default value is 2.
3935 qmax integer (encoding,video)
3937 Set max video quantizer scale (VBR). Must be included between -1
3938 and 1024, default value is 31.
3939 qdiff integer (encoding,video)
3941 Set max difference between the quantizer scale (VBR).
3942 bf integer (encoding,video)
3944 Set max number of B frames between non-B-frames.
3945 Must be an integer between -1 and 16. 0 means that B-frames are
3947 disabled. If a value of -1 is used, it will choose an automatic
3948 value depending on the encoder.
3949 Default value is 0.
3951 b_qfactor float (encoding,video)
3953 Set qp factor between P and B frames.
3954 codec_tag integer
3956 bug flags (decoding,video)
3957 Workaround not auto detected encoder bugs.
3958 Possible values:
3960 autodetect
3962 xvid_ilace
3963 Xvid interlacing bug (autodetected if fourcc==XVIX)
3964 ump4
3966 (autodetected if fourcc==UMP4)
3967 no_padding
3969 padding bug (autodetected)
3970 amv
3972 qpel_chroma
3973 std_qpel
3974 old standard qpel (autodetected per fourcc/version)
3975 qpel_chroma2
3977 direct_blocksize
3978 direct-qpel-blocksize bug (autodetected per fourcc/version)
3979 edge
3981 edge padding bug (autodetected per fourcc/version)
3982 hpel_chroma
3984 dc_clip
3985 ms Workaround various bugs in microsoft broken decoders.
3986 trunc
3988 trancated frames
3989 strict integer (decoding/encoding,audio,video)
3991 Specify how strictly to follow the standards.
3992 Possible values:
3994 very
3996 strictly conform to an older more strict version of the spec or
3997 reference software
3998 strict
4000 strictly conform to all the things in the spec no matter what
4001 consequences
4002 normal
4004 unofficial
4005 allow unofficial extensions
4006 experimental
4008 allow non standardized experimental things, experimental
4009 (unfinished/work in progress/not well tested) decoders and
4010 encoders. Note: experimental decoders can pose a security
4011 risk, do not use this for decoding untrusted input.
4012 b_qoffset float (encoding,video)
4014 Set QP offset between P and B frames.
4015 err_detect flags (decoding,audio,video)
4017 Set error detection flags.
4018 Possible values:
4020 crccheck
4022 verify embedded CRCs
4023 bitstream
4025 detect bitstream specification deviations
4026 buffer
4028 detect improper bitstream length
4029 explode
4031 abort decoding on minor error detection
4032 ignore_err
4034 ignore decoding errors, and continue decoding. This is useful
4035 if you want to analyze the content of a video and thus want
4036 everything to be decoded no matter what. This option will not
4037 result in a video that is pleasing to watch in case of errors.
4038 careful
4040 consider things that violate the spec and have not been seen in
4041 the wild as errors
4042 compliant
4044 consider all spec non compliancies as errors
4045 aggressive
4047 consider things that a sane encoder should not do as an error
4048 has_b_frames integer
4050 block_align integer
4051 rc_override_count integer
4052 maxrate integer (encoding,audio,video)
4053 Set max bitrate tolerance (in bits/s). Requires bufsize to be set.
4054 minrate integer (encoding,audio,video)
4056 Set min bitrate tolerance (in bits/s). Most useful in setting up a
4057 CBR encode. It is of little use elsewise.
4058 bufsize integer (encoding,audio,video)
4060 Set ratecontrol buffer size (in bits).
4061 i_qfactor float (encoding,video)
4063 Set QP factor between P and I frames.
4064 i_qoffset float (encoding,video)
4066 Set QP offset between P and I frames.
4067 dct integer (encoding,video)
4069 Set DCT algorithm.
4070 Possible values:
4072 auto
4074 autoselect a good one (default)
4075 fastint
4077 fast integer
4078 int accurate integer
4080 mmx
4082 altivec
4083 faan
4084 floating point AAN DCT
4085 lumi_mask float (encoding,video)
4087 Compress bright areas stronger than medium ones.
4088 tcplx_mask float (encoding,video)
4090 Set temporal complexity masking.
4091 scplx_mask float (encoding,video)
4093 Set spatial complexity masking.
4094 p_mask float (encoding,video)
4096 Set inter masking.
4097 dark_mask float (encoding,video)
4099 Compress dark areas stronger than medium ones.
4100 idct integer (decoding/encoding,video)
4102 Select IDCT implementation.
4103 Possible values:
4105 auto
4107 int
4108 simple
4109 simplemmx
4110 simpleauto
4111 Automatically pick a IDCT compatible with the simple one
4112 arm
4114 altivec
4115 sh4
4116 simplearm
4117 simplearmv5te
4118 simplearmv6
4119 simpleneon
4120 xvid
4121 faani
4122 floating point AAN IDCT
4123 slice_count integer
4125 ec flags (decoding,video)
4126 Set error concealment strategy.
4127 Possible values:
4129 guess_mvs
4131 iterative motion vector (MV) search (slow)
4132 deblock
4134 use strong deblock filter for damaged MBs
4135 favor_inter
4137 favor predicting from the previous frame instead of the current
4138 bits_per_coded_sample integer
4140 aspect rational number (encoding,video)
4141 Set sample aspect ratio.
4142 sar rational number (encoding,video)
4144 Set sample aspect ratio. Alias to aspect.
4145 debug flags (decoding/encoding,audio,video,subtitles)
4147 Print specific debug info.
4148 Possible values:
4150 pict
4152 picture info
4153 rc rate control
4155 bitstream
4157 mb_type
4158 macroblock (MB) type
4159 qp per-block quantization parameter (QP)
4161 dct_coeff
4163 green_metadata
4164 display complexity metadata for the upcoming frame, GoP or for
4165 a given duration.
4166 skip
4168 startcode
4169 er error recognition
4170 mmco
4172 memory management control operations (H.264)
4173 bugs
4175 buffers
4176 picture buffer allocations
4177 thread_ops
4179 threading operations
4180 nomc
4182 skip motion compensation
4183 cmp integer (encoding,video)
4185 Set full pel me compare function.
4186 Possible values:
4188 sad sum of absolute differences, fast (default)
4190 sse sum of squared errors
4192 satd
4194 sum of absolute Hadamard transformed differences
4195 dct sum of absolute DCT transformed differences
4197 psnr
4199 sum of squared quantization errors (avoid, low quality)
4200 bit number of bits needed for the block
4202 rd rate distortion optimal, slow
4204 zero
4206 0
4207 vsad
4209 sum of absolute vertical differences
4210 vsse
4212 sum of squared vertical differences
4213 nsse
4215 noise preserving sum of squared differences
4216 w53 5/3 wavelet, only used in snow
4218 w97 9/7 wavelet, only used in snow
4220 dctmax
4222 chroma
4223 subcmp integer (encoding,video)
4224 Set sub pel me compare function.
4225 Possible values:
4227 sad sum of absolute differences, fast (default)
4229 sse sum of squared errors
4231 satd
4233 sum of absolute Hadamard transformed differences
4234 dct sum of absolute DCT transformed differences
4236 psnr
4238 sum of squared quantization errors (avoid, low quality)
4239 bit number of bits needed for the block
4241 rd rate distortion optimal, slow
4243 zero
4245 0
4246 vsad
4248 sum of absolute vertical differences
4249 vsse
4251 sum of squared vertical differences
4252 nsse
4254 noise preserving sum of squared differences
4255 w53 5/3 wavelet, only used in snow
4257 w97 9/7 wavelet, only used in snow
4259 dctmax
4261 chroma
4262 mbcmp integer (encoding,video)
4263 Set macroblock compare function.
4264 Possible values:
4266 sad sum of absolute differences, fast (default)
4268 sse sum of squared errors
4270 satd
4272 sum of absolute Hadamard transformed differences
4273 dct sum of absolute DCT transformed differences
4275 psnr
4277 sum of squared quantization errors (avoid, low quality)
4278 bit number of bits needed for the block
4280 rd rate distortion optimal, slow
4282 zero
4284 0
4285 vsad
4287 sum of absolute vertical differences
4288 vsse
4290 sum of squared vertical differences
4291 nsse
4293 noise preserving sum of squared differences
4294 w53 5/3 wavelet, only used in snow
4296 w97 9/7 wavelet, only used in snow
4298 dctmax
4300 chroma
4301 ildctcmp integer (encoding,video)
4302 Set interlaced dct compare function.
4303 Possible values:
4305 sad sum of absolute differences, fast (default)
4307 sse sum of squared errors
4309 satd
4311 sum of absolute Hadamard transformed differences
4312 dct sum of absolute DCT transformed differences
4314 psnr
4316 sum of squared quantization errors (avoid, low quality)
4317 bit number of bits needed for the block
4319 rd rate distortion optimal, slow
4321 zero
4323 0
4324 vsad
4326 sum of absolute vertical differences
4327 vsse
4329 sum of squared vertical differences
4330 nsse
4332 noise preserving sum of squared differences
4333 w53 5/3 wavelet, only used in snow
4335 w97 9/7 wavelet, only used in snow
4337 dctmax
4339 chroma
4340 dia_size integer (encoding,video)
4341 Set diamond type & size for motion estimation.
4342 (1024, INT_MAX)
4344 full motion estimation(slowest)
4345 (768, 1024]
4347 umh motion estimation
4348 (512, 768]
4350 hex motion estimation
4351 (256, 512]
4353 l2s diamond motion estimation
4354 [2,256]
4356 var diamond motion estimation
4357 (-1, 2)
4359 small diamond motion estimation
4360 -1 funny diamond motion estimation
4362 (INT_MIN, -1)
4364 sab diamond motion estimation
4365 last_pred integer (encoding,video)
4367 Set amount of motion predictors from the previous frame.
4368 precmp integer (encoding,video)
4370 Set pre motion estimation compare function.
4371 Possible values:
4373 sad sum of absolute differences, fast (default)
4375 sse sum of squared errors
4377 satd
4379 sum of absolute Hadamard transformed differences
4380 dct sum of absolute DCT transformed differences
4382 psnr
4384 sum of squared quantization errors (avoid, low quality)
4385 bit number of bits needed for the block
4387 rd rate distortion optimal, slow
4389 zero
4391 0
4392 vsad
4394 sum of absolute vertical differences
4395 vsse
4397 sum of squared vertical differences
4398 nsse
4400 noise preserving sum of squared differences
4401 w53 5/3 wavelet, only used in snow
4403 w97 9/7 wavelet, only used in snow
4405 dctmax
4407 chroma
4408 pre_dia_size integer (encoding,video)
4409 Set diamond type & size for motion estimation pre-pass.
4410 subq integer (encoding,video)
4412 Set sub pel motion estimation quality.
4413 me_range integer (encoding,video)
4415 Set limit motion vectors range (1023 for DivX player).
4416 global_quality integer (encoding,audio,video)
4418 slice_flags integer
4419 mbd integer (encoding,video)
4420 Set macroblock decision algorithm (high quality mode).
4421 Possible values:
4423 simple
4425 use mbcmp (default)
4426 bits
4428 use fewest bits
4429 rd use best rate distortion
4431 rc_init_occupancy integer (encoding,video)
4433 Set number of bits which should be loaded into the rc buffer before
4434 decoding starts.
4435 flags2 flags (decoding/encoding,audio,video,subtitles)
4437 Possible values:
4438 fast
4440 Allow non spec compliant speedup tricks.
4441 noout
4443 Skip bitstream encoding.
4444 ignorecrop
4446 Ignore cropping information from sps.
4447 local_header
4449 Place global headers at every keyframe instead of in extradata.
4450 chunks
4452 Frame data might be split into multiple chunks.
4453 showall
4455 Show all frames before the first keyframe.
4456 export_mvs
4458 Export motion vectors into frame side-data (see
4459 "AV_FRAME_DATA_MOTION_VECTORS") for codecs that support it. See
4460 also doc/examples/export_mvs.c.
4461 skip_manual
4463 Do not skip samples and export skip information as frame side
4464 data.
4465 ass_ro_flush_noop
4467 Do not reset ASS ReadOrder field on flush.
4468 export_side_data flags (decoding/encoding,audio,video,subtitles)
4470 Possible values:
4471 mvs Export motion vectors into frame side-data (see
4473 "AV_FRAME_DATA_MOTION_VECTORS") for codecs that support it. See
4474 also doc/examples/export_mvs.c.
4475 prft
4477 Export encoder Producer Reference Time into packet side-data
4478 (see "AV_PKT_DATA_PRFT") for codecs that support it.
4479 venc_params
4481 Export video encoding parameters through frame side data (see
4482 "AV_FRAME_DATA_VIDEO_ENC_PARAMS") for codecs that support it.
4483 At present, those are H.264 and VP9.
4484 film_grain
4486 Export film grain parameters through frame side data (see
4487 "AV_FRAME_DATA_FILM_GRAIN_PARAMS"). Supported at present by
4488 AV1 decoders.
4489 threads integer (decoding/encoding,video)
4491 Set the number of threads to be used, in case the selected codec
4492 implementation supports multi-threading.
4493 Possible values:
4495 auto, 0
4497 automatically select the number of threads to set
4498 Default value is auto.
4500 dc integer (encoding,video)
4502 Set intra_dc_precision.
4503 nssew integer (encoding,video)
4505 Set nsse weight.
4506 skip_top integer (decoding,video)
4508 Set number of macroblock rows at the top which are skipped.
4509 skip_bottom integer (decoding,video)
4511 Set number of macroblock rows at the bottom which are skipped.
4512 profile integer (encoding,audio,video)
4514 Set encoder codec profile. Default value is unknown. Encoder
4515 specific profiles are documented in the relevant encoder
4516 documentation.
4517 level integer (encoding,audio,video)
4519 Possible values:
4520 unknown
4522 lowres integer (decoding,audio,video)
4523 Decode at 1= 1/2, 2=1/4, 3=1/8 resolutions.
4524 mblmin integer (encoding,video)
4526 Set min macroblock lagrange factor (VBR).
4527 mblmax integer (encoding,video)
4529 Set max macroblock lagrange factor (VBR).
4530 skip_loop_filter integer (decoding,video)
4532 skip_idct integer (decoding,video)
4533 skip_frame integer (decoding,video)
4534 Make decoder discard processing depending on the frame type
4535 selected by the option value.
4536 skip_loop_filter skips frame loop filtering, skip_idct skips frame
4538 IDCT/dequantization, skip_frame skips decoding.
4539 Possible values:
4541 none
4543 Discard no frame.
4544 default
4546 Discard useless frames like 0-sized frames.
4547 noref
4549 Discard all non-reference frames.
4550 bidir
4552 Discard all bidirectional frames.
4553 nokey
4555 Discard all frames excepts keyframes.
4556 nointra
4558 Discard all frames except I frames.
4559 all Discard all frames.
4561 Default value is default.
4563 bidir_refine integer (encoding,video)
4565 Refine the two motion vectors used in bidirectional macroblocks.
4566 keyint_min integer (encoding,video)
4568 Set minimum interval between IDR-frames.
4569 refs integer (encoding,video)
4571 Set reference frames to consider for motion compensation.
4572 trellis integer (encoding,audio,video)
4574 Set rate-distortion optimal quantization.
4575 mv0_threshold integer (encoding,video)
4577 compression_level integer (encoding,audio,video)
4578 bits_per_raw_sample integer
4579 channel_layout integer (decoding/encoding,audio)
4580 Possible values:
4581 request_channel_layout integer (decoding,audio)
4583 Possible values:
4584 rc_max_vbv_use float (encoding,video)
4586 rc_min_vbv_use float (encoding,video)
4587 ticks_per_frame integer (decoding/encoding,audio,video)
4588 color_primaries integer (decoding/encoding,video)
4589 Possible values:
4590 bt709
4592 BT.709
4593 bt470m
4595 BT.470 M
4596 bt470bg
4598 BT.470 BG
4599 smpte170m
4601 SMPTE 170 M
4602 smpte240m
4604 SMPTE 240 M
4605 film
4607 Film
4608 bt2020
4610 BT.2020
4611 smpte428
4613 smpte428_1
4614 SMPTE ST 428-1
4615 smpte431
4617 SMPTE 431-2
4618 smpte432
4620 SMPTE 432-1
4621 jedec-p22
4623 JEDEC P22
4624 color_trc integer (decoding/encoding,video)
4626 Possible values:
4627 bt709
4629 BT.709
4630 gamma22
4632 BT.470 M
4633 gamma28
4635 BT.470 BG
4636 smpte170m
4638 SMPTE 170 M
4639 smpte240m
4641 SMPTE 240 M
4642 linear
4644 Linear
4645 log
4647 log100
4648 Log
4649 log_sqrt
4651 log316
4652 Log square root
4653 iec61966_2_4
4655 iec61966-2-4
4656 IEC 61966-2-4
4657 bt1361
4659 bt1361e
4660 BT.1361
4661 iec61966_2_1
4663 iec61966-2-1
4664 IEC 61966-2-1
4665 bt2020_10
4667 bt2020_10bit
4668 BT.2020 - 10 bit
4669 bt2020_12
4671 bt2020_12bit
4672 BT.2020 - 12 bit
4673 smpte2084
4675 SMPTE ST 2084
4676 smpte428
4678 smpte428_1
4679 SMPTE ST 428-1
4680 arib-std-b67
4682 ARIB STD-B67
4683 colorspace integer (decoding/encoding,video)
4685 Possible values:
4686 rgb RGB
4688 bt709
4690 BT.709
4691 fcc FCC
4693 bt470bg
4695 BT.470 BG
4696 smpte170m
4698 SMPTE 170 M
4699 smpte240m
4701 SMPTE 240 M
4702 ycocg
4704 YCOCG
4705 bt2020nc
4707 bt2020_ncl
4708 BT.2020 NCL
4709 bt2020c
4711 bt2020_cl
4712 BT.2020 CL
4713 smpte2085
4715 SMPTE 2085
4716 chroma-derived-nc
4718 Chroma-derived NCL
4719 chroma-derived-c
4721 Chroma-derived CL
4722 ictcp
4724 ICtCp
4725 color_range integer (decoding/encoding,video)
4727 If used as input parameter, it serves as a hint to the decoder,
4728 which color_range the input has. Possible values:
4729 tv
4731 mpeg
4732 MPEG (219*2^(n-8))
4733 pc
4735 jpeg
4736 JPEG (2^n-1)
4737 chroma_sample_location integer (decoding/encoding,video)
4739 Possible values:
4740 left
4742 center
4743 topleft
4744 top
4745 bottomleft
4746 bottom
4747 log_level_offset integer
4748 Set the log level offset.
4749 slices integer (encoding,video)
4751 Number of slices, used in parallelized encoding.
4752 thread_type flags (decoding/encoding,video)
4754 Select which multithreading methods to use.
4755 Use of frame will increase decoding delay by one frame per thread,
4757 so clients which cannot provide future frames should not use it.
4758 Possible values:
4760 slice
4762 Decode more than one part of a single frame at once.
4763 Multithreading using slices works only when the video was
4765 encoded with slices.
4766 frame
4768 Decode more than one frame at once.
4769 Default value is slice+frame.
4771 audio_service_type integer (encoding,audio)
4773 Set audio service type.
4774 Possible values:
4776 ma Main Audio Service
4778 ef Effects
4780 vi Visually Impaired
4782 hi Hearing Impaired
4784 di Dialogue
4786 co Commentary
4788 em Emergency
4790 vo Voice Over
4792 ka Karaoke
4794 request_sample_fmt sample_fmt (decoding,audio)
4796 Set sample format audio decoders should prefer. Default value is
4797 "none".
4798 pkt_timebase rational number
4800 sub_charenc encoding (decoding,subtitles)
4801 Set the input subtitles character encoding.
4802 field_order field_order (video)
4804 Set/override the field order of the video. Possible values:
4805 progressive
4807 Progressive video
4808 tt Interlaced video, top field coded and displayed first
4810 bb Interlaced video, bottom field coded and displayed first
4812 tb Interlaced video, top coded first, bottom displayed first
4814 bt Interlaced video, bottom coded first, top displayed first
4816 skip_alpha bool (decoding,video)
4818 Set to 1 to disable processing alpha (transparency). This works
4819 like the gray flag in the flags option which skips chroma
4820 information instead of alpha. Default is 0.
4821 codec_whitelist list (input)
4823 "," separated list of allowed decoders. By default all are allowed.
4824 dump_separator string (input)
4826 Separator used to separate the fields printed on the command line
4827 about the Stream parameters. For example, to separate the fields
4828 with newlines and indentation:
4829 ffprobe -dump_separator "
4831 " -i ~/videos/matrixbench_mpeg2.mpg
4832 max_pixels integer (decoding/encoding,video)
4834 Maximum number of pixels per image. This value can be used to avoid
4835 out of memory failures due to large images.
4836 apply_cropping bool (decoding,video)
4838 Enable cropping if cropping parameters are multiples of the
4839 required alignment for the left and top parameters. If the
4840 alignment is not met the cropping will be partially applied to
4841 maintain alignment. Default is 1 (enabled). Note: The required
4842 alignment depends on if "AV_CODEC_FLAG_UNALIGNED" is set and the
4843 CPU. "AV_CODEC_FLAG_UNALIGNED" cannot be changed from the command
4844 line. Also hardware decoders will not apply left/top Cropping.
4845
4847 Decoders are configured elements in FFmpeg which allow the decoding of
4848 multimedia streams.
4849 When you configure your FFmpeg build, all the supported native decoders
4851 are enabled by default. Decoders requiring an external library must be
4852 enabled manually via the corresponding "--enable-lib" option. You can
4853 list all available decoders using the configure option
4854 "--list-decoders".
4855 You can disable all the decoders with the configure option
4857 "--disable-decoders" and selectively enable / disable single decoders
4858 with the options "--enable-decoder=DECODER" /
4859 "--disable-decoder=DECODER".
4860 The option "-decoders" of the ff* tools will display the list of
4862 enabled decoders.
4863
4865 A description of some of the currently available video decoders
4866 follows.
4867 av1
4869 AOMedia Video 1 (AV1) decoder.
4870 Options
4872 operating_point
4874 Select an operating point of a scalable AV1 bitstream (0 - 31).
4875 Default is 0.
4876 rawvideo
4878 Raw video decoder.
4879 This decoder decodes rawvideo streams.
4881 Options
4883 top top_field_first
4885 Specify the assumed field type of the input video.
4886 -1 the video is assumed to be progressive (default)
4888 0 bottom-field-first is assumed
4890 1 top-field-first is assumed
4892 libdav1d
4894 dav1d AV1 decoder.
4895 libdav1d allows libavcodec to decode the AOMedia Video 1 (AV1) codec.
4897 Requires the presence of the libdav1d headers and library during
4898 configuration. You need to explicitly configure the build with
4899 "--enable-libdav1d".
4900 Options
4902 The following options are supported by the libdav1d wrapper.
4904 framethreads
4906 Set amount of frame threads to use during decoding. The default
4907 value is 0 (autodetect). This option is deprecated for libdav1d >=
4908 1.0 and will be removed in the future. Use the global option
4909 "threads" instead.
4910 tilethreads
4912 Set amount of tile threads to use during decoding. The default
4913 value is 0 (autodetect). This option is deprecated for libdav1d >=
4914 1.0 and will be removed in the future. Use the global option
4915 "threads" instead.
4916 filmgrain
4918 Apply film grain to the decoded video if present in the bitstream.
4919 Defaults to the internal default of the library. This option is
4920 deprecated and will be removed in the future. See the global option
4921 "export_side_data" to export Film Grain parameters instead of
4922 applying it.
4923 oppoint
4925 Select an operating point of a scalable AV1 bitstream (0 - 31).
4926 Defaults to the internal default of the library.
4927 alllayers
4929 Output all spatial layers of a scalable AV1 bitstream. The default
4930 value is false.
4931 libdavs2
4933 AVS2-P2/IEEE1857.4 video decoder wrapper.
4934 This decoder allows libavcodec to decode AVS2 streams with davs2
4936 library.
4937 libuavs3d
4939 AVS3-P2/IEEE1857.10 video decoder.
4940 libuavs3d allows libavcodec to decode AVS3 streams. Requires the
4942 presence of the libuavs3d headers and library during configuration.
4943 You need to explicitly configure the build with "--enable-libuavs3d".
4944 Options
4946 The following option is supported by the libuavs3d wrapper.
4948 frame_threads
4950 Set amount of frame threads to use during decoding. The default
4951 value is 0 (autodetect).
4952
4954 A description of some of the currently available audio decoders
4955 follows.
4956 ac3
4958 AC-3 audio decoder.
4959 This decoder implements part of ATSC A/52:2010 and ETSI TS 102 366, as
4961 well as the undocumented RealAudio 3 (a.k.a. dnet).
4962 AC-3 Decoder Options
4964 -drc_scale value
4966 Dynamic Range Scale Factor. The factor to apply to dynamic range
4967 values from the AC-3 stream. This factor is applied exponentially.
4968 The default value is 1. There are 3 notable scale factor ranges:
4969 drc_scale == 0
4971 DRC disabled. Produces full range audio.
4972 0 < drc_scale <= 1
4974 DRC enabled. Applies a fraction of the stream DRC value.
4975 Audio reproduction is between full range and full compression.
4976 drc_scale > 1
4978 DRC enabled. Applies drc_scale asymmetrically. Loud sounds are
4979 fully compressed. Soft sounds are enhanced.
4980 flac
4982 FLAC audio decoder.
4983 This decoder aims to implement the complete FLAC specification from
4985 Xiph.
4986 FLAC Decoder options
4988 -use_buggy_lpc
4990 The lavc FLAC encoder used to produce buggy streams with high lpc
4991 values (like the default value). This option makes it possible to
4992 decode such streams correctly by using lavc's old buggy lpc logic
4993 for decoding.
4994 ffwavesynth
4996 Internal wave synthesizer.
4997 This decoder generates wave patterns according to predefined sequences.
4999 Its use is purely internal and the format of the data it accepts is not
5000 publicly documented.
5001 libcelt
5003 libcelt decoder wrapper.
5004 libcelt allows libavcodec to decode the Xiph CELT ultra-low delay audio
5006 codec. Requires the presence of the libcelt headers and library during
5007 configuration. You need to explicitly configure the build with
5008 "--enable-libcelt".
5009 libgsm
5011 libgsm decoder wrapper.
5012 libgsm allows libavcodec to decode the GSM full rate audio codec.
5014 Requires the presence of the libgsm headers and library during
5015 configuration. You need to explicitly configure the build with
5016 "--enable-libgsm".
5017 This decoder supports both the ordinary GSM and the Microsoft variant.
5019 libilbc
5021 libilbc decoder wrapper.
5022 libilbc allows libavcodec to decode the Internet Low Bitrate Codec
5024 (iLBC) audio codec. Requires the presence of the libilbc headers and
5025 library during configuration. You need to explicitly configure the
5026 build with "--enable-libilbc".
5027 Options
5029 The following option is supported by the libilbc wrapper.
5031 enhance
5033 Enable the enhancement of the decoded audio when set to 1. The
5034 default value is 0 (disabled).
5035 libopencore-amrnb
5037 libopencore-amrnb decoder wrapper.
5038 libopencore-amrnb allows libavcodec to decode the Adaptive Multi-Rate
5040 Narrowband audio codec. Using it requires the presence of the
5041 libopencore-amrnb headers and library during configuration. You need to
5042 explicitly configure the build with "--enable-libopencore-amrnb".
5043 An FFmpeg native decoder for AMR-NB exists, so users can decode AMR-NB
5045 without this library.
5046 libopencore-amrwb
5048 libopencore-amrwb decoder wrapper.
5049 libopencore-amrwb allows libavcodec to decode the Adaptive Multi-Rate
5051 Wideband audio codec. Using it requires the presence of the
5052 libopencore-amrwb headers and library during configuration. You need to
5053 explicitly configure the build with "--enable-libopencore-amrwb".
5054 An FFmpeg native decoder for AMR-WB exists, so users can decode AMR-WB
5056 without this library.
5057 libopus
5059 libopus decoder wrapper.
5060 libopus allows libavcodec to decode the Opus Interactive Audio Codec.
5062 Requires the presence of the libopus headers and library during
5063 configuration. You need to explicitly configure the build with
5064 "--enable-libopus".
5065 An FFmpeg native decoder for Opus exists, so users can decode Opus
5067 without this library.
5068
5070 libaribb24
5071 ARIB STD-B24 caption decoder.
5072 Implements profiles A and C of the ARIB STD-B24 standard.
5074 libaribb24 Decoder Options
5076 -aribb24-base-path path
5078 Sets the base path for the libaribb24 library. This is utilized for
5079 reading of configuration files (for custom unicode conversions),
5080 and for dumping of non-text symbols as images under that location.
5081 Unset by default.
5083 -aribb24-skip-ruby-text boolean
5085 Tells the decoder wrapper to skip text blocks that contain half-
5086 height ruby text.
5087 Enabled by default.
5089 dvbsub
5091 Options
5092 compute_clut
5094 -2 Compute clut once if no matching CLUT is in the stream.
5095 -1 Compute clut if no matching CLUT is in the stream.
5097 0 Never compute CLUT
5099 1 Always compute CLUT and override the one provided in the
5101 stream.
5102 dvb_substream
5104 Selects the dvb substream, or all substreams if -1 which is
5105 default.
5106 dvdsub
5108 This codec decodes the bitmap subtitles used in DVDs; the same
5109 subtitles can also be found in VobSub file pairs and in some Matroska
5110 files.
5111 Options
5113 palette
5115 Specify the global palette used by the bitmaps. When stored in
5116 VobSub, the palette is normally specified in the index file; in
5117 Matroska, the palette is stored in the codec extra-data in the same
5118 format as in VobSub. In DVDs, the palette is stored in the IFO
5119 file, and therefore not available when reading from dumped VOB
5120 files.
5121 The format for this option is a string containing 16 24-bits
5123 hexadecimal numbers (without 0x prefix) separated by commas, for
5124 example "0d00ee, ee450d, 101010, eaeaea, 0ce60b, ec14ed, ebff0b,
5125 0d617a, 7b7b7b, d1d1d1, 7b2a0e, 0d950c, 0f007b, cf0dec, cfa80c,
5126 7c127b".
5127 ifo_palette
5129 Specify the IFO file from which the global palette is obtained.
5130 (experimental)
5131 forced_subs_only
5133 Only decode subtitle entries marked as forced. Some titles have
5134 forced and non-forced subtitles in the same track. Setting this
5135 flag to 1 will only keep the forced subtitles. Default value is 0.
5136 libzvbi-teletext
5138 Libzvbi allows libavcodec to decode DVB teletext pages and DVB teletext
5139 subtitles. Requires the presence of the libzvbi headers and library
5140 during configuration. You need to explicitly configure the build with
5141 "--enable-libzvbi".
5142 Options
5144 txt_page
5146 List of teletext page numbers to decode. Pages that do not match
5147 the specified list are dropped. You may use the special "*" string
5148 to match all pages, or "subtitle" to match all subtitle pages.
5149 Default value is *.
5150 txt_default_region
5152 Set default character set used for decoding, a value between 0 and
5153 87 (see ETS 300 706, Section 15, Table 32). Default value is -1,
5154 which does not override the libzvbi default. This option is needed
5155 for some legacy level 1.0 transmissions which cannot signal the
5156 proper charset.
5157 txt_chop_top
5159 Discards the top teletext line. Default value is 1.
5160 txt_format
5162 Specifies the format of the decoded subtitles.
5163 bitmap
5165 The default format, you should use this for teletext pages,
5166 because certain graphics and colors cannot be expressed in
5167 simple text or even ASS.
5168 text
5170 Simple text based output without formatting.
5171 ass Formatted ASS output, subtitle pages and teletext pages are
5173 returned in different styles, subtitle pages are stripped down
5174 to text, but an effort is made to keep the text alignment and
5175 the formatting.
5176 txt_left
5178 X offset of generated bitmaps, default is 0.
5179 txt_top
5181 Y offset of generated bitmaps, default is 0.
5182 txt_chop_spaces
5184 Chops leading and trailing spaces and removes empty lines from the
5185 generated text. This option is useful for teletext based subtitles
5186 where empty spaces may be present at the start or at the end of the
5187 lines or empty lines may be present between the subtitle lines
5188 because of double-sized teletext characters. Default value is 1.
5189 txt_duration
5191 Sets the display duration of the decoded teletext pages or
5192 subtitles in milliseconds. Default value is -1 which means infinity
5193 or until the next subtitle event comes.
5194 txt_transparent
5196 Force transparent background of the generated teletext bitmaps.
5197 Default value is 0 which means an opaque background.
5198 txt_opacity
5200 Sets the opacity (0-255) of the teletext background. If
5201 txt_transparent is not set, it only affects characters between a
5202 start box and an end box, typically subtitles. Default value is 0
5203 if txt_transparent is set, 255 otherwise.
5204
5206 Encoders are configured elements in FFmpeg which allow the encoding of
5207 multimedia streams.
5208 When you configure your FFmpeg build, all the supported native encoders
5210 are enabled by default. Encoders requiring an external library must be
5211 enabled manually via the corresponding "--enable-lib" option. You can
5212 list all available encoders using the configure option
5213 "--list-encoders".
5214 You can disable all the encoders with the configure option
5216 "--disable-encoders" and selectively enable / disable single encoders
5217 with the options "--enable-encoder=ENCODER" /
5218 "--disable-encoder=ENCODER".
5219 The option "-encoders" of the ff* tools will display the list of
5221 enabled encoders.
5222
5224 A description of some of the currently available audio encoders
5225 follows.
5226 aac
5228 Advanced Audio Coding (AAC) encoder.
5229 This encoder is the default AAC encoder, natively implemented into
5231 FFmpeg.
5232 Options
5234 b Set bit rate in bits/s. Setting this automatically activates
5236 constant bit rate (CBR) mode. If this option is unspecified it is
5237 set to 128kbps.
5238 q Set quality for variable bit rate (VBR) mode. This option is valid
5240 only using the ffmpeg command-line tool. For library interface
5241 users, use global_quality.
5242 cutoff
5244 Set cutoff frequency. If unspecified will allow the encoder to
5245 dynamically adjust the cutoff to improve clarity on low bitrates.
5246 aac_coder
5248 Set AAC encoder coding method. Possible values:
5249 twoloop
5251 Two loop searching (TLS) method. This is the default method.
5252 This method first sets quantizers depending on band thresholds
5254 and then tries to find an optimal combination by adding or
5255 subtracting a specific value from all quantizers and adjusting
5256 some individual quantizer a little. Will tune itself based on
5257 whether aac_is, aac_ms and aac_pns are enabled.
5258 anmr
5260 Average noise to mask ratio (ANMR) trellis-based solution.
5261 This is an experimental coder which currently produces a lower
5263 quality, is more unstable and is slower than the default
5264 twoloop coder but has potential. Currently has no support for
5265 the aac_is or aac_pns options. Not currently recommended.
5266 fast
5268 Constant quantizer method.
5269 Uses a cheaper version of twoloop algorithm that doesn't try to
5271 do as many clever adjustments. Worse with low bitrates (less
5272 than 64kbps), but is better and much faster at higher bitrates.
5273 aac_ms
5275 Sets mid/side coding mode. The default value of "auto" will
5276 automatically use M/S with bands which will benefit from such
5277 coding. Can be forced for all bands using the value "enable", which
5278 is mainly useful for debugging or disabled using "disable".
5279 aac_is
5281 Sets intensity stereo coding tool usage. By default, it's enabled
5282 and will automatically toggle IS for similar pairs of stereo bands
5283 if it's beneficial. Can be disabled for debugging by setting the
5284 value to "disable".
5285 aac_pns
5287 Uses perceptual noise substitution to replace low entropy high
5288 frequency bands with imperceptible white noise during the decoding
5289 process. By default, it's enabled, but can be disabled for
5290 debugging purposes by using "disable".
5291 aac_tns
5293 Enables the use of a multitap FIR filter which spans through the
5294 high frequency bands to hide quantization noise during the encoding
5295 process and is reverted by the decoder. As well as decreasing
5296 unpleasant artifacts in the high range this also reduces the
5297 entropy in the high bands and allows for more bits to be used by
5298 the mid-low bands. By default it's enabled but can be disabled for
5299 debugging by setting the option to "disable".
5300 aac_ltp
5302 Enables the use of the long term prediction extension which
5303 increases coding efficiency in very low bandwidth situations such
5304 as encoding of voice or solo piano music by extending constant
5305 harmonic peaks in bands throughout frames. This option is implied
5306 by profile:a aac_low and is incompatible with aac_pred. Use in
5307 conjunction with -ar to decrease the samplerate.
5308 aac_pred
5310 Enables the use of a more traditional style of prediction where the
5311 spectral coefficients transmitted are replaced by the difference of
5312 the current coefficients minus the previous "predicted"
5313 coefficients. In theory and sometimes in practice this can improve
5314 quality for low to mid bitrate audio. This option implies the
5315 aac_main profile and is incompatible with aac_ltp.
5316 profile
5318 Sets the encoding profile, possible values:
5319 aac_low
5321 The default, AAC "Low-complexity" profile. Is the most
5322 compatible and produces decent quality.
5323 mpeg2_aac_low
5325 Equivalent to "-profile:a aac_low -aac_pns 0". PNS was
5326 introduced with the MPEG4 specifications.
5327 aac_ltp
5329 Long term prediction profile, is enabled by and will enable the
5330 aac_ltp option. Introduced in MPEG4.
5331 aac_main
5333 Main-type prediction profile, is enabled by and will enable the
5334 aac_pred option. Introduced in MPEG2.
5335 If this option is unspecified it is set to aac_low.
5337 ac3 and ac3_fixed
5339 AC-3 audio encoders.
5340 These encoders implement part of ATSC A/52:2010 and ETSI TS 102 366, as
5342 well as the undocumented RealAudio 3 (a.k.a. dnet).
5343 The ac3 encoder uses floating-point math, while the ac3_fixed encoder
5345 only uses fixed-point integer math. This does not mean that one is
5346 always faster, just that one or the other may be better suited to a
5347 particular system. The ac3_fixed encoder is not the default codec for
5348 any of the output formats, so it must be specified explicitly using the
5349 option "-acodec ac3_fixed" in order to use it.
5350 AC-3 Metadata
5352 The AC-3 metadata options are used to set parameters that describe the
5354 audio, but in most cases do not affect the audio encoding itself. Some
5355 of the options do directly affect or influence the decoding and
5356 playback of the resulting bitstream, while others are just for
5357 informational purposes. A few of the options will add bits to the
5358 output stream that could otherwise be used for audio data, and will
5359 thus affect the quality of the output. Those will be indicated
5360 accordingly with a note in the option list below.
5361 These parameters are described in detail in several publicly-available
5363 documents.
5364 *<<http://www.atsc.org/cms/standards/a_52-2010.pdf>>
5366 *<<http://www.atsc.org/cms/standards/a_54a_with_corr_1.pdf>>
5367 *<<http://www.dolby.com/uploadedFiles/zz-_Shared_Assets/English_PDFs/Professional/18_Metadata.Guide.pdf>>
5368 *<<http://www.dolby.com/uploadedFiles/zz-_Shared_Assets/English_PDFs/Professional/46_DDEncodingGuidelines.pdf>>
5369 Metadata Control Options
5371 -per_frame_metadata boolean
5373 Allow Per-Frame Metadata. Specifies if the encoder should check for
5374 changing metadata for each frame.
5375 0 The metadata values set at initialization will be used for
5377 every frame in the stream. (default)
5378 1 Metadata values can be changed before encoding each frame.
5380 Downmix Levels
5382 -center_mixlev level
5384 Center Mix Level. The amount of gain the decoder should apply to
5385 the center channel when downmixing to stereo. This field will only
5386 be written to the bitstream if a center channel is present. The
5387 value is specified as a scale factor. There are 3 valid values:
5388 0.707
5390 Apply -3dB gain
5391 0.595
5393 Apply -4.5dB gain (default)
5394 0.500
5396 Apply -6dB gain
5397 -surround_mixlev level
5399 Surround Mix Level. The amount of gain the decoder should apply to
5400 the surround channel(s) when downmixing to stereo. This field will
5401 only be written to the bitstream if one or more surround channels
5402 are present. The value is specified as a scale factor. There are 3
5403 valid values:
5404 0.707
5406 Apply -3dB gain
5407 0.500
5409 Apply -6dB gain (default)
5410 0.000
5412 Silence Surround Channel(s)
5413 Audio Production Information
5415 Audio Production Information is optional information describing the
5417 mixing environment. Either none or both of the fields are written to
5418 the bitstream.
5419 -mixing_level number
5421 Mixing Level. Specifies peak sound pressure level (SPL) in the
5422 production environment when the mix was mastered. Valid values are
5423 80 to 111, or -1 for unknown or not indicated. The default value is
5424 -1, but that value cannot be used if the Audio Production
5425 Information is written to the bitstream. Therefore, if the
5426 "room_type" option is not the default value, the "mixing_level"
5427 option must not be -1.
5428 -room_type type
5430 Room Type. Describes the equalization used during the final mixing
5431 session at the studio or on the dubbing stage. A large room is a
5432 dubbing stage with the industry standard X-curve equalization; a
5433 small room has flat equalization. This field will not be written
5434 to the bitstream if both the "mixing_level" option and the
5435 "room_type" option have the default values.
5436 0
5438 notindicated
5439 Not Indicated (default)
5440 1
5442 large
5443 Large Room
5444 2
5446 small
5447 Small Room
5448 Other Metadata Options
5450 -copyright boolean
5452 Copyright Indicator. Specifies whether a copyright exists for this
5453 audio.
5454 0
5456 off No Copyright Exists (default)
5457 1
5459 on Copyright Exists
5460 -dialnorm value
5462 Dialogue Normalization. Indicates how far the average dialogue
5463 level of the program is below digital 100% full scale (0 dBFS).
5464 This parameter determines a level shift during audio reproduction
5465 that sets the average volume of the dialogue to a preset level. The
5466 goal is to match volume level between program sources. A value of
5467 -31dB will result in no volume level change, relative to the source
5468 volume, during audio reproduction. Valid values are whole numbers
5469 in the range -31 to -1, with -31 being the default.
5470 -dsur_mode mode
5472 Dolby Surround Mode. Specifies whether the stereo signal uses Dolby
5473 Surround (Pro Logic). This field will only be written to the
5474 bitstream if the audio stream is stereo. Using this option does NOT
5475 mean the encoder will actually apply Dolby Surround processing.
5476 0
5478 notindicated
5479 Not Indicated (default)
5480 1
5482 off Not Dolby Surround Encoded
5483 2
5485 on Dolby Surround Encoded
5486 -original boolean
5488 Original Bit Stream Indicator. Specifies whether this audio is from
5489 the original source and not a copy.
5490 0
5492 off Not Original Source
5493 1
5495 on Original Source (default)
5496 Extended Bitstream Information
5498 The extended bitstream options are part of the Alternate Bit Stream
5500 Syntax as specified in Annex D of the A/52:2010 standard. It is grouped
5501 into 2 parts. If any one parameter in a group is specified, all values
5502 in that group will be written to the bitstream. Default values are
5503 used for those that are written but have not been specified. If the
5504 mixing levels are written, the decoder will use these values instead of
5505 the ones specified in the "center_mixlev" and "surround_mixlev" options
5506 if it supports the Alternate Bit Stream Syntax.
5507 Extended Bitstream Information - Part 1
5509 -dmix_mode mode
5511 Preferred Stereo Downmix Mode. Allows the user to select either
5512 Lt/Rt (Dolby Surround) or Lo/Ro (normal stereo) as the preferred
5513 stereo downmix mode.
5514 0
5516 notindicated
5517 Not Indicated (default)
5518 1
5520 ltrt
5521 Lt/Rt Downmix Preferred
5522 2
5524 loro
5525 Lo/Ro Downmix Preferred
5526 -ltrt_cmixlev level
5528 Lt/Rt Center Mix Level. The amount of gain the decoder should apply
5529 to the center channel when downmixing to stereo in Lt/Rt mode.
5530 1.414
5532 Apply +3dB gain
5533 1.189
5535 Apply +1.5dB gain
5536 1.000
5538 Apply 0dB gain
5539 0.841
5541 Apply -1.5dB gain
5542 0.707
5544 Apply -3.0dB gain
5545 0.595
5547 Apply -4.5dB gain (default)
5548 0.500
5550 Apply -6.0dB gain
5551 0.000
5553 Silence Center Channel
5554 -ltrt_surmixlev level
5556 Lt/Rt Surround Mix Level. The amount of gain the decoder should
5557 apply to the surround channel(s) when downmixing to stereo in Lt/Rt
5558 mode.
5559 0.841
5561 Apply -1.5dB gain
5562 0.707
5564 Apply -3.0dB gain
5565 0.595
5567 Apply -4.5dB gain
5568 0.500
5570 Apply -6.0dB gain (default)
5571 0.000
5573 Silence Surround Channel(s)
5574 -loro_cmixlev level
5576 Lo/Ro Center Mix Level. The amount of gain the decoder should apply
5577 to the center channel when downmixing to stereo in Lo/Ro mode.
5578 1.414
5580 Apply +3dB gain
5581 1.189
5583 Apply +1.5dB gain
5584 1.000
5586 Apply 0dB gain
5587 0.841
5589 Apply -1.5dB gain
5590 0.707
5592 Apply -3.0dB gain
5593 0.595
5595 Apply -4.5dB gain (default)
5596 0.500
5598 Apply -6.0dB gain
5599 0.000
5601 Silence Center Channel
5602 -loro_surmixlev level
5604 Lo/Ro Surround Mix Level. The amount of gain the decoder should
5605 apply to the surround channel(s) when downmixing to stereo in Lo/Ro
5606 mode.
5607 0.841
5609 Apply -1.5dB gain
5610 0.707
5612 Apply -3.0dB gain
5613 0.595
5615 Apply -4.5dB gain
5616 0.500
5618 Apply -6.0dB gain (default)
5619 0.000
5621 Silence Surround Channel(s)
5622 Extended Bitstream Information - Part 2
5624 -dsurex_mode mode
5626 Dolby Surround EX Mode. Indicates whether the stream uses Dolby
5627 Surround EX (7.1 matrixed to 5.1). Using this option does NOT mean
5628 the encoder will actually apply Dolby Surround EX processing.
5629 0
5631 notindicated
5632 Not Indicated (default)
5633 1
5635 on Dolby Surround EX Off
5636 2
5638 off Dolby Surround EX On
5639 -dheadphone_mode mode
5641 Dolby Headphone Mode. Indicates whether the stream uses Dolby
5642 Headphone encoding (multi-channel matrixed to 2.0 for use with
5643 headphones). Using this option does NOT mean the encoder will
5644 actually apply Dolby Headphone processing.
5645 0
5647 notindicated
5648 Not Indicated (default)
5649 1
5651 on Dolby Headphone Off
5652 2
5654 off Dolby Headphone On
5655 -ad_conv_type type
5657 A/D Converter Type. Indicates whether the audio has passed through
5658 HDCD A/D conversion.
5659 0
5661 standard
5662 Standard A/D Converter (default)
5663 1
5665 hdcd
5666 HDCD A/D Converter
5667 Other AC-3 Encoding Options
5669 -stereo_rematrixing boolean
5671 Stereo Rematrixing. Enables/Disables use of rematrixing for stereo
5672 input. This is an optional AC-3 feature that increases quality by
5673 selectively encoding the left/right channels as mid/side. This
5674 option is enabled by default, and it is highly recommended that it
5675 be left as enabled except for testing purposes.
5676 cutoff frequency
5678 Set lowpass cutoff frequency. If unspecified, the encoder selects a
5679 default determined by various other encoding parameters.
5680 Floating-Point-Only AC-3 Encoding Options
5682 These options are only valid for the floating-point encoder and do not
5684 exist for the fixed-point encoder due to the corresponding features not
5685 being implemented in fixed-point.
5686 -channel_coupling boolean
5688 Enables/Disables use of channel coupling, which is an optional AC-3
5689 feature that increases quality by combining high frequency
5690 information from multiple channels into a single channel. The per-
5691 channel high frequency information is sent with less accuracy in
5692 both the frequency and time domains. This allows more bits to be
5693 used for lower frequencies while preserving enough information to
5694 reconstruct the high frequencies. This option is enabled by default
5695 for the floating-point encoder and should generally be left as
5696 enabled except for testing purposes or to increase encoding speed.
5697 -1
5699 auto
5700 Selected by Encoder (default)
5701 0
5703 off Disable Channel Coupling
5704 1
5706 on Enable Channel Coupling
5707 -cpl_start_band number
5709 Coupling Start Band. Sets the channel coupling start band, from 1
5710 to 15. If a value higher than the bandwidth is used, it will be
5711 reduced to 1 less than the coupling end band. If auto is used, the
5712 start band will be determined by the encoder based on the bit rate,
5713 sample rate, and channel layout. This option has no effect if
5714 channel coupling is disabled.
5715 -1
5717 auto
5718 Selected by Encoder (default)
5719 flac
5721 FLAC (Free Lossless Audio Codec) Encoder
5722 Options
5724 The following options are supported by FFmpeg's flac encoder.
5726 compression_level
5728 Sets the compression level, which chooses defaults for many other
5729 options if they are not set explicitly. Valid values are from 0 to
5730 12, 5 is the default.
5731 frame_size
5733 Sets the size of the frames in samples per channel.
5734 lpc_coeff_precision
5736 Sets the LPC coefficient precision, valid values are from 1 to 15,
5737 15 is the default.
5738 lpc_type
5740 Sets the first stage LPC algorithm
5741 none
5743 LPC is not used
5744 fixed
5746 fixed LPC coefficients
5747 levinson
5749 cholesky
5750 lpc_passes
5751 Number of passes to use for Cholesky factorization during LPC
5752 analysis
5753 min_partition_order
5755 The minimum partition order
5756 max_partition_order
5758 The maximum partition order
5759 prediction_order_method
5761 estimation
5762 2level
5763 4level
5764 8level
5765 search
5766 Bruteforce search
5767 log
5769 ch_mode
5770 Channel mode
5771 auto
5773 The mode is chosen automatically for each frame
5774 indep
5776 Channels are independently coded
5777 left_side
5779 right_side
5780 mid_side
5781 exact_rice_parameters
5782 Chooses if rice parameters are calculated exactly or approximately.
5783 if set to 1 then they are chosen exactly, which slows the code down
5784 slightly and improves compression slightly.
5785 multi_dim_quant
5787 Multi Dimensional Quantization. If set to 1 then a 2nd stage LPC
5788 algorithm is applied after the first stage to finetune the
5789 coefficients. This is quite slow and slightly improves compression.
5790 opus
5792 Opus encoder.
5793 This is a native FFmpeg encoder for the Opus format. Currently its in
5795 development and only implements the CELT part of the codec. Its quality
5796 is usually worse and at best is equal to the libopus encoder.
5797 Options
5799 b Set bit rate in bits/s. If unspecified it uses the number of
5801 channels and the layout to make a good guess.
5802 opus_delay
5804 Sets the maximum delay in milliseconds. Lower delays than 20ms will
5805 very quickly decrease quality.
5806 libfdk_aac
5808 libfdk-aac AAC (Advanced Audio Coding) encoder wrapper.
5809 The libfdk-aac library is based on the Fraunhofer FDK AAC code from the
5811 Android project.
5812 Requires the presence of the libfdk-aac headers and library during
5814 configuration. You need to explicitly configure the build with
5815 "--enable-libfdk-aac". The library is also incompatible with GPL, so if
5816 you allow the use of GPL, you should configure with "--enable-gpl
5817 --enable-nonfree --enable-libfdk-aac".
5818 This encoder has support for the AAC-HE profiles.
5820 VBR encoding, enabled through the vbr or flags +qscale options, is
5822 experimental and only works with some combinations of parameters.
5823 Support for encoding 7.1 audio is only available with libfdk-aac 0.1.3
5825 or higher.
5826 For more information see the fdk-aac project at
5828 <http://sourceforge.net/p/opencore-amr/fdk-aac/>.
5829 Options
5831 The following options are mapped on the shared FFmpeg codec options.
5833 b Set bit rate in bits/s. If the bitrate is not explicitly specified,
5835 it is automatically set to a suitable value depending on the
5836 selected profile.
5837 In case VBR mode is enabled the option is ignored.
5839 ar Set audio sampling rate (in Hz).
5841 channels
5843 Set the number of audio channels.
5844 flags +qscale
5846 Enable fixed quality, VBR (Variable Bit Rate) mode. Note that VBR
5847 is implicitly enabled when the vbr value is positive.
5848 cutoff
5850 Set cutoff frequency. If not specified (or explicitly set to 0) it
5851 will use a value automatically computed by the library. Default
5852 value is 0.
5853 profile
5855 Set audio profile.
5856 The following profiles are recognized:
5858 aac_low
5860 Low Complexity AAC (LC)
5861 aac_he
5863 High Efficiency AAC (HE-AAC)
5864 aac_he_v2
5866 High Efficiency AAC version 2 (HE-AACv2)
5867 aac_ld
5869 Low Delay AAC (LD)
5870 aac_eld
5872 Enhanced Low Delay AAC (ELD)
5873 If not specified it is set to aac_low.
5875 The following are private options of the libfdk_aac encoder.
5877 afterburner
5879 Enable afterburner feature if set to 1, disabled if set to 0. This
5880 improves the quality but also the required processing power.
5881 Default value is 1.
5883 eld_sbr
5885 Enable SBR (Spectral Band Replication) for ELD if set to 1,
5886 disabled if set to 0.
5887 Default value is 0.
5889 eld_v2
5891 Enable ELDv2 (LD-MPS extension for ELD stereo signals) for ELDv2 if
5892 set to 1, disabled if set to 0.
5893 Note that option is available when fdk-aac version
5895 (AACENCODER_LIB_VL0.AACENCODER_LIB_VL1.AACENCODER_LIB_VL2) >
5896 (4.0.0).
5897 Default value is 0.
5899 signaling
5901 Set SBR/PS signaling style.
5902 It can assume one of the following values:
5904 default
5906 choose signaling implicitly (explicit hierarchical by default,
5907 implicit if global header is disabled)
5908 implicit
5910 implicit backwards compatible signaling
5911 explicit_sbr
5913 explicit SBR, implicit PS signaling
5914 explicit_hierarchical
5916 explicit hierarchical signaling
5917 Default value is default.
5919 latm
5921 Output LATM/LOAS encapsulated data if set to 1, disabled if set to
5922 0.
5923 Default value is 0.
5925 header_period
5927 Set StreamMuxConfig and PCE repetition period (in frames) for
5928 sending in-band configuration buffers within LATM/LOAS transport
5929 layer.
5930 Must be a 16-bits non-negative integer.
5932 Default value is 0.
5934 vbr Set VBR mode, from 1 to 5. 1 is lowest quality (though still pretty
5936 good) and 5 is highest quality. A value of 0 will disable VBR, and
5937 CBR (Constant Bit Rate) is enabled.
5938 Currently only the aac_low profile supports VBR encoding.
5940 VBR modes 1-5 correspond to roughly the following average bit
5942 rates:
5943 1 32 kbps/channel
5945 2 40 kbps/channel
5947 3 48-56 kbps/channel
5949 4 64 kbps/channel
5951 5 about 80-96 kbps/channel
5953 Default value is 0.
5955 Examples
5957 • Use ffmpeg to convert an audio file to VBR AAC in an M4A (MP4)
5959 container:
5960 ffmpeg -i input.wav -codec:a libfdk_aac -vbr 3 output.m4a
5962 • Use ffmpeg to convert an audio file to CBR 64k kbps AAC, using the
5964 High-Efficiency AAC profile:
5965 ffmpeg -i input.wav -c:a libfdk_aac -profile:a aac_he -b:a 64k output.m4a
5967 libmp3lame
5969 LAME (Lame Ain't an MP3 Encoder) MP3 encoder wrapper.
5970 Requires the presence of the libmp3lame headers and library during
5972 configuration. You need to explicitly configure the build with
5973 "--enable-libmp3lame".
5974 See libshine for a fixed-point MP3 encoder, although with a lower
5976 quality.
5977 Options
5979 The following options are supported by the libmp3lame wrapper. The
5981 lame-equivalent of the options are listed in parentheses.
5982 b (-b)
5984 Set bitrate expressed in bits/s for CBR or ABR. LAME "bitrate" is
5985 expressed in kilobits/s.
5986 q (-V)
5988 Set constant quality setting for VBR. This option is valid only
5989 using the ffmpeg command-line tool. For library interface users,
5990 use global_quality.
5991 compression_level (-q)
5993 Set algorithm quality. Valid arguments are integers in the 0-9
5994 range, with 0 meaning highest quality but slowest, and 9 meaning
5995 fastest while producing the worst quality.
5996 cutoff (--lowpass)
5998 Set lowpass cutoff frequency. If unspecified, the encoder
5999 dynamically adjusts the cutoff.
6000 reservoir
6002 Enable use of bit reservoir when set to 1. Default value is 1. LAME
6003 has this enabled by default, but can be overridden by use --nores
6004 option.
6005 joint_stereo (-m j)
6007 Enable the encoder to use (on a frame by frame basis) either L/R
6008 stereo or mid/side stereo. Default value is 1.
6009 abr (--abr)
6011 Enable the encoder to use ABR when set to 1. The lame --abr sets
6012 the target bitrate, while this options only tells FFmpeg to use ABR
6013 still relies on b to set bitrate.
6014 libopencore-amrnb
6016 OpenCORE Adaptive Multi-Rate Narrowband encoder.
6017 Requires the presence of the libopencore-amrnb headers and library
6019 during configuration. You need to explicitly configure the build with
6020 "--enable-libopencore-amrnb --enable-version3".
6021 This is a mono-only encoder. Officially it only supports 8000Hz sample
6023 rate, but you can override it by setting strict to unofficial or lower.
6024 Options
6026 b Set bitrate in bits per second. Only the following bitrates are
6028 supported, otherwise libavcodec will round to the nearest valid
6029 bitrate.
6030 4750
6032 5150
6033 5900
6034 6700
6035 7400
6036 7950
6037 10200
6038 12200
6039 dtx Allow discontinuous transmission (generate comfort noise) when set
6040 to 1. The default value is 0 (disabled).
6041 libopus
6043 libopus Opus Interactive Audio Codec encoder wrapper.
6044 Requires the presence of the libopus headers and library during
6046 configuration. You need to explicitly configure the build with
6047 "--enable-libopus".
6048 Option Mapping
6050 Most libopus options are modelled after the opusenc utility from opus-
6052 tools. The following is an option mapping chart describing options
6053 supported by the libopus wrapper, and their opusenc-equivalent in
6054 parentheses.
6055 b (bitrate)
6057 Set the bit rate in bits/s. FFmpeg's b option is expressed in
6058 bits/s, while opusenc's bitrate in kilobits/s.
6059 vbr (vbr, hard-cbr, and cvbr)
6061 Set VBR mode. The FFmpeg vbr option has the following valid
6062 arguments, with the opusenc equivalent options in parentheses:
6063 off (hard-cbr)
6065 Use constant bit rate encoding.
6066 on (vbr)
6068 Use variable bit rate encoding (the default).
6069 constrained (cvbr)
6071 Use constrained variable bit rate encoding.
6072 compression_level (comp)
6074 Set encoding algorithm complexity. Valid options are integers in
6075 the 0-10 range. 0 gives the fastest encodes but lower quality,
6076 while 10 gives the highest quality but slowest encoding. The
6077 default is 10.
6078 frame_duration (framesize)
6080 Set maximum frame size, or duration of a frame in milliseconds. The
6081 argument must be exactly the following: 2.5, 5, 10, 20, 40, 60.
6082 Smaller frame sizes achieve lower latency but less quality at a
6083 given bitrate. Sizes greater than 20ms are only interesting at
6084 fairly low bitrates. The default is 20ms.
6085 packet_loss (expect-loss)
6087 Set expected packet loss percentage. The default is 0.
6088 fec (n/a)
6090 Enable inband forward error correction. packet_loss must be non-
6091 zero to take advantage - frequency of FEC 'side-data' is
6092 proportional to expected packet loss. Default is disabled.
6093 application (N.A.)
6095 Set intended application type. Valid options are listed below:
6096 voip
6098 Favor improved speech intelligibility.
6099 audio
6101 Favor faithfulness to the input (the default).
6102 lowdelay
6104 Restrict to only the lowest delay modes.
6105 cutoff (N.A.)
6107 Set cutoff bandwidth in Hz. The argument must be exactly one of the
6108 following: 4000, 6000, 8000, 12000, or 20000, corresponding to
6109 narrowband, mediumband, wideband, super wideband, and fullband
6110 respectively. The default is 0 (cutoff disabled).
6111 mapping_family (mapping_family)
6113 Set channel mapping family to be used by the encoder. The default
6114 value of -1 uses mapping family 0 for mono and stereo inputs, and
6115 mapping family 1 otherwise. The default also disables the surround
6116 masking and LFE bandwidth optimzations in libopus, and requires
6117 that the input contains 8 channels or fewer.
6118 Other values include 0 for mono and stereo, 1 for surround sound
6120 with masking and LFE bandwidth optimizations, and 255 for
6121 independent streams with an unspecified channel layout.
6122 apply_phase_inv (N.A.) (requires libopus >= 1.2)
6124 If set to 0, disables the use of phase inversion for intensity
6125 stereo, improving the quality of mono downmixes, but slightly
6126 reducing normal stereo quality. The default is 1 (phase inversion
6127 enabled).
6128 libshine
6130 Shine Fixed-Point MP3 encoder wrapper.
6131 Shine is a fixed-point MP3 encoder. It has a far better performance on
6133 platforms without an FPU, e.g. armel CPUs, and some phones and tablets.
6134 However, as it is more targeted on performance than quality, it is not
6135 on par with LAME and other production-grade encoders quality-wise.
6136 Also, according to the project's homepage, this encoder may not be free
6137 of bugs as the code was written a long time ago and the project was
6138 dead for at least 5 years.
6139 This encoder only supports stereo and mono input. This is also CBR-
6141 only.
6142 The original project (last updated in early 2007) is at
6144 <http://sourceforge.net/projects/libshine-fxp/>. We only support the
6145 updated fork by the Savonet/Liquidsoap project at
6146 <https://github.com/savonet/shine>.
6147 Requires the presence of the libshine headers and library during
6149 configuration. You need to explicitly configure the build with
6150 "--enable-libshine".
6151 See also libmp3lame.
6153 Options
6155 The following options are supported by the libshine wrapper. The
6157 shineenc-equivalent of the options are listed in parentheses.
6158 b (-b)
6160 Set bitrate expressed in bits/s for CBR. shineenc -b option is
6161 expressed in kilobits/s.
6162 libtwolame
6164 TwoLAME MP2 encoder wrapper.
6165 Requires the presence of the libtwolame headers and library during
6167 configuration. You need to explicitly configure the build with
6168 "--enable-libtwolame".
6169 Options
6171 The following options are supported by the libtwolame wrapper. The
6173 twolame-equivalent options follow the FFmpeg ones and are in
6174 parentheses.
6175 b (-b)
6177 Set bitrate expressed in bits/s for CBR. twolame b option is
6178 expressed in kilobits/s. Default value is 128k.
6179 q (-V)
6181 Set quality for experimental VBR support. Maximum value range is
6182 from -50 to 50, useful range is from -10 to 10. The higher the
6183 value, the better the quality. This option is valid only using the
6184 ffmpeg command-line tool. For library interface users, use
6185 global_quality.
6186 mode (--mode)
6188 Set the mode of the resulting audio. Possible values:
6189 auto
6191 Choose mode automatically based on the input. This is the
6192 default.
6193 stereo
6195 Stereo
6196 joint_stereo
6198 Joint stereo
6199 dual_channel
6201 Dual channel
6202 mono
6204 Mono
6205 psymodel (--psyc-mode)
6207 Set psychoacoustic model to use in encoding. The argument must be
6208 an integer between -1 and 4, inclusive. The higher the value, the
6209 better the quality. The default value is 3.
6210 energy_levels (--energy)
6212 Enable energy levels extensions when set to 1. The default value is
6213 0 (disabled).
6214 error_protection (--protect)
6216 Enable CRC error protection when set to 1. The default value is 0
6217 (disabled).
6218 copyright (--copyright)
6220 Set MPEG audio copyright flag when set to 1. The default value is 0
6221 (disabled).
6222 original (--original)
6224 Set MPEG audio original flag when set to 1. The default value is 0
6225 (disabled).
6226 libvo-amrwbenc
6228 VisualOn Adaptive Multi-Rate Wideband encoder.
6229 Requires the presence of the libvo-amrwbenc headers and library during
6231 configuration. You need to explicitly configure the build with
6232 "--enable-libvo-amrwbenc --enable-version3".
6233 This is a mono-only encoder. Officially it only supports 16000Hz sample
6235 rate, but you can override it by setting strict to unofficial or lower.
6236 Options
6238 b Set bitrate in bits/s. Only the following bitrates are supported,
6240 otherwise libavcodec will round to the nearest valid bitrate.
6241 6600
6243 8850
6244 12650
6245 14250
6246 15850
6247 18250
6248 19850
6249 23050
6250 23850
6251 dtx Allow discontinuous transmission (generate comfort noise) when set
6252 to 1. The default value is 0 (disabled).
6253 libvorbis
6255 libvorbis encoder wrapper.
6256 Requires the presence of the libvorbisenc headers and library during
6258 configuration. You need to explicitly configure the build with
6259 "--enable-libvorbis".
6260 Options
6262 The following options are supported by the libvorbis wrapper. The
6264 oggenc-equivalent of the options are listed in parentheses.
6265 To get a more accurate and extensive documentation of the libvorbis
6267 options, consult the libvorbisenc's and oggenc's documentations. See
6268 <http://xiph.org/vorbis/>, <http://wiki.xiph.org/Vorbis-tools>, and
6269 oggenc(1).
6270 b (-b)
6272 Set bitrate expressed in bits/s for ABR. oggenc -b is expressed in
6273 kilobits/s.
6274 q (-q)
6276 Set constant quality setting for VBR. The value should be a float
6277 number in the range of -1.0 to 10.0. The higher the value, the
6278 better the quality. The default value is 3.0.
6279 This option is valid only using the ffmpeg command-line tool. For
6281 library interface users, use global_quality.
6282 cutoff (--advanced-encode-option lowpass_frequency=N)
6284 Set cutoff bandwidth in Hz, a value of 0 disables cutoff. oggenc's
6285 related option is expressed in kHz. The default value is 0 (cutoff
6286 disabled).
6287 minrate (-m)
6289 Set minimum bitrate expressed in bits/s. oggenc -m is expressed in
6290 kilobits/s.
6291 maxrate (-M)
6293 Set maximum bitrate expressed in bits/s. oggenc -M is expressed in
6294 kilobits/s. This only has effect on ABR mode.
6295 iblock (--advanced-encode-option impulse_noisetune=N)
6297 Set noise floor bias for impulse blocks. The value is a float
6298 number from -15.0 to 0.0. A negative bias instructs the encoder to
6299 pay special attention to the crispness of transients in the encoded
6300 audio. The tradeoff for better transient response is a higher
6301 bitrate.
6302 mjpeg
6304 Motion JPEG encoder.
6305 Options
6307 huffman
6309 Set the huffman encoding strategy. Possible values:
6310 default
6312 Use the default huffman tables. This is the default strategy.
6313 optimal
6315 Compute and use optimal huffman tables.
6316 wavpack
6318 WavPack lossless audio encoder.
6319 Options
6321 The equivalent options for wavpack command line utility are listed in
6323 parentheses.
6324 Shared options
6326 The following shared options are effective for this encoder. Only
6328 special notes about this particular encoder will be documented here.
6329 For the general meaning of the options, see the Codec Options chapter.
6330 frame_size (--blocksize)
6332 For this encoder, the range for this option is between 128 and
6333 131072. Default is automatically decided based on sample rate and
6334 number of channel.
6335 For the complete formula of calculating default, see
6337 libavcodec/wavpackenc.c.
6338 compression_level (-f, -h, -hh, and -x)
6340 Private options
6342 joint_stereo (-j)
6344 Set whether to enable joint stereo. Valid values are:
6345 on (1)
6347 Force mid/side audio encoding.
6348 off (0)
6350 Force left/right audio encoding.
6351 auto
6353 Let the encoder decide automatically.
6354 optimize_mono
6356 Set whether to enable optimization for mono. This option is only
6357 effective for non-mono streams. Available values:
6358 on enabled
6360 off disabled
6362
6364 A description of some of the currently available video encoders
6365 follows.
6366 a64_multi, a64_multi5
6368 A64 / Commodore 64 multicolor charset encoder. "a64_multi5" is extended
6369 with 5th color (colram).
6370 GIF
6372 GIF image/animation encoder.
6373 Options
6375 gifflags integer
6377 Sets the flags used for GIF encoding.
6378 offsetting
6380 Enables picture offsetting.
6381 Default is enabled.
6383 transdiff
6385 Enables transparency detection between frames.
6386 Default is enabled.
6388 gifimage integer
6390 Enables encoding one full GIF image per frame, rather than an
6391 animated GIF.
6392 Default value is 0.
6394 global_palette integer
6396 Writes a palette to the global GIF header where feasible.
6397 If disabled, every frame will always have a palette written, even
6399 if there is a global palette supplied.
6400 Default value is 1.
6402 Hap
6404 Vidvox Hap video encoder.
6405 Options
6407 format integer
6409 Specifies the Hap format to encode.
6410 hap
6412 hap_alpha
6413 hap_q
6414 Default value is hap.
6416 chunks integer
6418 Specifies the number of chunks to split frames into, between 1 and
6419 64. This permits multithreaded decoding of large frames,
6420 potentially at the cost of data-rate. The encoder may modify this
6421 value to divide frames evenly.
6422 Default value is 1.
6424 compressor integer
6426 Specifies the second-stage compressor to use. If set to none,
6427 chunks will be limited to 1, as chunked uncompressed frames offer
6428 no benefit.
6429 none
6431 snappy
6432 Default value is snappy.
6434 jpeg2000
6436 The native jpeg 2000 encoder is lossy by default, the "-q:v" option can
6437 be used to set the encoding quality. Lossless encoding can be selected
6438 with "-pred 1".
6439 Options
6441 format integer
6443 Can be set to either "j2k" or "jp2" (the default) that makes it
6444 possible to store non-rgb pix_fmts.
6445 tile_width integer
6447 Sets tile width. Range is 1 to 1073741824. Default is 256.
6448 tile_height integer
6450 Sets tile height. Range is 1 to 1073741824. Default is 256.
6451 pred integer
6453 Allows setting the discrete wavelet transform (DWT) type
6454 dwt97int (Lossy)
6456 dwt53 (Lossless)
6457 Default is "dwt97int"
6459 sop boolean
6461 Enable this to add SOP marker at the start of each packet. Disabled
6462 by default.
6463 eph boolean
6465 Enable this to add EPH marker at the end of each packet header.
6466 Disabled by default.
6467 prog integer
6469 Sets the progression order to be used by the encoder. Possible
6470 values are:
6471 lrcp
6473 rlcp
6474 rpcl
6475 pcrl
6476 cprl
6477 Set to "lrcp" by default.
6479 layer_rates string
6481 By default, when this option is not used, compression is done using
6482 the quality metric. This option allows for compression using
6483 compression ratio. The compression ratio for each level could be
6484 specified. The compression ratio of a layer "l" species the what
6485 ratio of total file size is contained in the first "l" layers.
6486 Example usage:
6488 ffmpeg -i input.bmp -c:v jpeg2000 -layer_rates "100,10,1" output.j2k
6490 This would compress the image to contain 3 layers, where the data
6492 contained in the first layer would be compressed by 1000 times,
6493 compressed by 100 in the first two layers, and shall contain all
6494 data while using all 3 layers.
6495 librav1e
6497 rav1e AV1 encoder wrapper.
6498 Requires the presence of the rav1e headers and library during
6500 configuration. You need to explicitly configure the build with
6501 "--enable-librav1e".
6502 Options
6504 qmax
6506 Sets the maximum quantizer to use when using bitrate mode.
6507 qmin
6509 Sets the minimum quantizer to use when using bitrate mode.
6510 qp Uses quantizer mode to encode at the given quantizer (0-255).
6512 speed
6514 Selects the speed preset (0-10) to encode with.
6515 tiles
6517 Selects how many tiles to encode with.
6518 tile-rows
6520 Selects how many rows of tiles to encode with.
6521 tile-columns
6523 Selects how many columns of tiles to encode with.
6524 rav1e-params
6526 Set rav1e options using a list of key=value pairs separated by ":".
6527 See rav1e --help for a list of options.
6528 For example to specify librav1e encoding options with
6530 -rav1e-params:
6531 ffmpeg -i input -c:v librav1e -b:v 500K -rav1e-params speed=5:low_latency=true output.mp4
6533 libaom-av1
6535 libaom AV1 encoder wrapper.
6536 Requires the presence of the libaom headers and library during
6538 configuration. You need to explicitly configure the build with
6539 "--enable-libaom".
6540 Options
6542 The wrapper supports the following standard libavcodec options:
6544 b Set bitrate target in bits/second. By default this will use
6546 variable-bitrate mode. If maxrate and minrate are also set to the
6547 same value then it will use constant-bitrate mode, otherwise if crf
6548 is set as well then it will use constrained-quality mode.
6549 g keyint_min
6551 Set key frame placement. The GOP size sets the maximum distance
6552 between key frames; if zero the output stream will be intra-only.
6553 The minimum distance is ignored unless it is the same as the GOP
6554 size, in which case key frames will always appear at a fixed
6555 interval. Not set by default, so without this option the library
6556 has completely free choice about where to place key frames.
6557 qmin qmax
6559 Set minimum/maximum quantisation values. Valid range is from 0 to
6560 63 (warning: this does not match the quantiser values actually used
6561 by AV1 - divide by four to map real quantiser values to this
6562 range). Defaults to min/max (no constraint).
6563 minrate maxrate bufsize rc_init_occupancy
6565 Set rate control buffering parameters. Not used if not set -
6566 defaults to unconstrained variable bitrate.
6567 threads
6569 Set the number of threads to use while encoding. This may require
6570 the tiles or row-mt options to also be set to actually use the
6571 specified number of threads fully. Defaults to the number of
6572 hardware threads supported by the host machine.
6573 profile
6575 Set the encoding profile. Defaults to using the profile which
6576 matches the bit depth and chroma subsampling of the input.
6577 The wrapper also has some specific options:
6579 cpu-used
6581 Set the quality/encoding speed tradeoff. Valid range is from 0 to
6582 8, higher numbers indicating greater speed and lower quality. The
6583 default value is 1, which will be slow and high quality.
6584 auto-alt-ref
6586 Enable use of alternate reference frames. Defaults to the internal
6587 default of the library.
6588 arnr-max-frames (frames)
6590 Set altref noise reduction max frame count. Default is -1.
6591 arnr-strength (strength)
6593 Set altref noise reduction filter strength. Range is -1 to 6.
6594 Default is -1.
6595 aq-mode (aq-mode)
6597 Set adaptive quantization mode. Possible values:
6598 none (0)
6600 Disabled.
6601 variance (1)
6603 Variance-based.
6604 complexity (2)
6606 Complexity-based.
6607 cyclic (3)
6609 Cyclic refresh.
6610 tune (tune)
6612 Set the distortion metric the encoder is tuned with. Default is
6613 "psnr".
6614 psnr (0)
6616 ssim (1)
6617 lag-in-frames
6618 Set the maximum number of frames which the encoder may keep in
6619 flight at any one time for lookahead purposes. Defaults to the
6620 internal default of the library.
6621 error-resilience
6623 Enable error resilience features:
6624 default
6626 Improve resilience against losses of whole frames.
6627 Not enabled by default.
6629 crf Set the quality/size tradeoff for constant-quality (no bitrate
6631 target) and constrained-quality (with maximum bitrate target)
6632 modes. Valid range is 0 to 63, higher numbers indicating lower
6633 quality and smaller output size. Only used if set; by default only
6634 the bitrate target is used.
6635 static-thresh
6637 Set a change threshold on blocks below which they will be skipped
6638 by the encoder. Defined in arbitrary units as a nonnegative
6639 integer, defaulting to zero (no blocks are skipped).
6640 drop-threshold
6642 Set a threshold for dropping frames when close to rate control
6643 bounds. Defined as a percentage of the target buffer - when the
6644 rate control buffer falls below this percentage, frames will be
6645 dropped until it has refilled above the threshold. Defaults to
6646 zero (no frames are dropped).
6647 denoise-noise-level (level)
6649 Amount of noise to be removed for grain synthesis. Grain synthesis
6650 is disabled if this option is not set or set to 0.
6651 denoise-block-size (pixels)
6653 Block size used for denoising for grain synthesis. If not set, AV1
6654 codec uses the default value of 32.
6655 undershoot-pct (pct)
6657 Set datarate undershoot (min) percentage of the target bitrate.
6658 Range is -1 to 100. Default is -1.
6659 overshoot-pct (pct)
6661 Set datarate overshoot (max) percentage of the target bitrate.
6662 Range is -1 to 1000. Default is -1.
6663 minsection-pct (pct)
6665 Minimum percentage variation of the GOP bitrate from the target
6666 bitrate. If minsection-pct is not set, the libaomenc wrapper
6667 computes it as follows: "(minrate * 100 / bitrate)". Range is -1
6668 to 100. Default is -1 (unset).
6669 maxsection-pct (pct)
6671 Maximum percentage variation of the GOP bitrate from the target
6672 bitrate. If maxsection-pct is not set, the libaomenc wrapper
6673 computes it as follows: "(maxrate * 100 / bitrate)". Range is -1
6674 to 5000. Default is -1 (unset).
6675 frame-parallel (boolean)
6677 Enable frame parallel decodability features. Default is true.
6678 tiles
6680 Set the number of tiles to encode the input video with, as columns
6681 x rows. Larger numbers allow greater parallelism in both encoding
6682 and decoding, but may decrease coding efficiency. Defaults to the
6683 minimum number of tiles required by the size of the input video
6684 (this is 1x1 (that is, a single tile) for sizes up to and including
6685 4K).
6686 tile-columns tile-rows
6688 Set the number of tiles as log2 of the number of tile rows and
6689 columns. Provided for compatibility with libvpx/VP9.
6690 row-mt (Requires libaom >= 1.0.0-759-g90a15f4f2)
6692 Enable row based multi-threading. Disabled by default.
6693 enable-cdef (boolean)
6695 Enable Constrained Directional Enhancement Filter. The libaom-av1
6696 encoder enables CDEF by default.
6697 enable-restoration (boolean)
6699 Enable Loop Restoration Filter. Default is true for libaom-av1.
6700 enable-global-motion (boolean)
6702 Enable the use of global motion for block prediction. Default is
6703 true.
6704 enable-intrabc (boolean)
6706 Enable block copy mode for intra block prediction. This mode is
6707 useful for screen content. Default is true.
6708 enable-rect-partitions (boolean) (Requires libaom >= v2.0.0)
6710 Enable rectangular partitions. Default is true.
6711 enable-1to4-partitions (boolean) (Requires libaom >= v2.0.0)
6713 Enable 1:4/4:1 partitions. Default is true.
6714 enable-ab-partitions (boolean) (Requires libaom >= v2.0.0)
6716 Enable AB shape partitions. Default is true.
6717 enable-angle-delta (boolean) (Requires libaom >= v2.0.0)
6719 Enable angle delta intra prediction. Default is true.
6720 enable-cfl-intra (boolean) (Requires libaom >= v2.0.0)
6722 Enable chroma predicted from luma intra prediction. Default is
6723 true.
6724 enable-filter-intra (boolean) (Requires libaom >= v2.0.0)
6726 Enable filter intra predictor. Default is true.
6727 enable-intra-edge-filter (boolean) (Requires libaom >= v2.0.0)
6729 Enable intra edge filter. Default is true.
6730 enable-smooth-intra (boolean) (Requires libaom >= v2.0.0)
6732 Enable smooth intra prediction mode. Default is true.
6733 enable-paeth-intra (boolean) (Requires libaom >= v2.0.0)
6735 Enable paeth predictor in intra prediction. Default is true.
6736 enable-palette (boolean) (Requires libaom >= v2.0.0)
6738 Enable palette prediction mode. Default is true.
6739 enable-flip-idtx (boolean) (Requires libaom >= v2.0.0)
6741 Enable extended transform type, including FLIPADST_DCT,
6742 DCT_FLIPADST, FLIPADST_FLIPADST, ADST_FLIPADST, FLIPADST_ADST,
6743 IDTX, V_DCT, H_DCT, V_ADST, H_ADST, V_FLIPADST, H_FLIPADST. Default
6744 is true.
6745 enable-tx64 (boolean) (Requires libaom >= v2.0.0)
6747 Enable 64-pt transform. Default is true.
6748 reduced-tx-type-set (boolean) (Requires libaom >= v2.0.0)
6750 Use reduced set of transform types. Default is false.
6751 use-intra-dct-only (boolean) (Requires libaom >= v2.0.0)
6753 Use DCT only for INTRA modes. Default is false.
6754 use-inter-dct-only (boolean) (Requires libaom >= v2.0.0)
6756 Use DCT only for INTER modes. Default is false.
6757 use-intra-default-tx-only (boolean) (Requires libaom >= v2.0.0)
6759 Use Default-transform only for INTRA modes. Default is false.
6760 enable-ref-frame-mvs (boolean) (Requires libaom >= v2.0.0)
6762 Enable temporal mv prediction. Default is true.
6763 enable-reduced-reference-set (boolean) (Requires libaom >= v2.0.0)
6765 Use reduced set of single and compound references. Default is
6766 false.
6767 enable-obmc (boolean) (Requires libaom >= v2.0.0)
6769 Enable obmc. Default is true.
6770 enable-dual-filter (boolean) (Requires libaom >= v2.0.0)
6772 Enable dual filter. Default is true.
6773 enable-diff-wtd-comp (boolean) (Requires libaom >= v2.0.0)
6775 Enable difference-weighted compound. Default is true.
6776 enable-dist-wtd-comp (boolean) (Requires libaom >= v2.0.0)
6778 Enable distance-weighted compound. Default is true.
6779 enable-onesided-comp (boolean) (Requires libaom >= v2.0.0)
6781 Enable one sided compound. Default is true.
6782 enable-interinter-wedge (boolean) (Requires libaom >= v2.0.0)
6784 Enable interinter wedge compound. Default is true.
6785 enable-interintra-wedge (boolean) (Requires libaom >= v2.0.0)
6787 Enable interintra wedge compound. Default is true.
6788 enable-masked-comp (boolean) (Requires libaom >= v2.0.0)
6790 Enable masked compound. Default is true.
6791 enable-interintra-comp (boolean) (Requires libaom >= v2.0.0)
6793 Enable interintra compound. Default is true.
6794 enable-smooth-interintra (boolean) (Requires libaom >= v2.0.0)
6796 Enable smooth interintra mode. Default is true.
6797 aom-params
6799 Set libaom options using a list of key=value pairs separated by
6800 ":". For a list of supported options, see aomenc --help under the
6801 section "AV1 Specific Options".
6802 For example to specify libaom encoding options with -aom-params:
6804 ffmpeg -i input -c:v libaom-av1 -b:v 500K -aom-params tune=psnr:enable-tpl-model=1 output.mp4
6806 libsvtav1
6808 SVT-AV1 encoder wrapper.
6809 Requires the presence of the SVT-AV1 headers and library during
6811 configuration. You need to explicitly configure the build with
6812 "--enable-libsvtav1".
6813 Options
6815 profile
6817 Set the encoding profile.
6818 main
6820 high
6821 professional
6822 level
6823 Set the operating point level. For example: '4.0'
6824 hielevel
6826 Set the Hierarchical prediction levels.
6827 3level
6829 4level
6830 This is the default.
6831 tier
6833 Set the operating point tier.
6834 main
6836 This is the default.
6837 high
6839 rc Set the rate control mode to use.
6840 Possible modes:
6842 cqp Constant quantizer: use fixed values of qindex (dependent on
6844 the frame type) throughout the stream. This mode is the
6845 default.
6846 vbr Variable bitrate: use a target bitrate for the whole stream.
6848 cvbr
6850 Constrained variable bitrate: use a target bitrate for each
6851 GOP.
6852 qmax
6854 Set the maximum quantizer to use when using a bitrate mode.
6855 qmin
6857 Set the minimum quantizer to use when using a bitrate mode.
6858 qp Set the quantizer used in cqp rate control mode (0-63).
6860 sc_detection
6862 Enable scene change detection.
6863 la_depth
6865 Set number of frames to look ahead (0-120).
6866 preset
6868 Set the quality-speed tradeoff, in the range 0 to 8. Higher values
6869 are faster but lower quality. Defaults to 8 (highest speed).
6870 tile_rows
6872 Set log2 of the number of rows of tiles to use (0-6).
6873 tile_columns
6875 Set log2 of the number of columns of tiles to use (0-4).
6876 libkvazaar
6878 Kvazaar H.265/HEVC encoder.
6879 Requires the presence of the libkvazaar headers and library during
6881 configuration. You need to explicitly configure the build with
6882 --enable-libkvazaar.
6883 Options
6885 b Set target video bitrate in bit/s and enable rate control.
6887 kvazaar-params
6889 Set kvazaar parameters as a list of name=value pairs separated by
6890 commas (,). See kvazaar documentation for a list of options.
6891 libopenh264
6893 Cisco libopenh264 H.264/MPEG-4 AVC encoder wrapper.
6894 This encoder requires the presence of the libopenh264 headers and
6896 library during configuration. You need to explicitly configure the
6897 build with "--enable-libopenh264". The library is detected using pkg-
6898 config.
6899 For more information about the library see <http://www.openh264.org>.
6901 Options
6903 The following FFmpeg global options affect the configurations of the
6905 libopenh264 encoder.
6906 b Set the bitrate (as a number of bits per second).
6908 g Set the GOP size.
6910 maxrate
6912 Set the max bitrate (as a number of bits per second).
6913 flags +global_header
6915 Set global header in the bitstream.
6916 slices
6918 Set the number of slices, used in parallelized encoding. Default
6919 value is 0. This is only used when slice_mode is set to fixed.
6920 slice_mode
6922 Set slice mode. Can assume one of the following possible values:
6923 fixed
6925 a fixed number of slices
6926 rowmb
6928 one slice per row of macroblocks
6929 auto
6931 automatic number of slices according to number of threads
6932 dyn dynamic slicing
6934 Default value is auto.
6936 loopfilter
6938 Enable loop filter, if set to 1 (automatically enabled). To disable
6939 set a value of 0.
6940 profile
6942 Set profile restrictions. If set to the value of main enable CABAC
6943 (set the "SEncParamExt.iEntropyCodingModeFlag" flag to 1).
6944 max_nal_size
6946 Set maximum NAL size in bytes.
6947 allow_skip_frames
6949 Allow skipping frames to hit the target bitrate if set to 1.
6950 libtheora
6952 libtheora Theora encoder wrapper.
6953 Requires the presence of the libtheora headers and library during
6955 configuration. You need to explicitly configure the build with
6956 "--enable-libtheora".
6957 For more information about the libtheora project see
6959 <http://www.theora.org/>.
6960 Options
6962 The following global options are mapped to internal libtheora options
6964 which affect the quality and the bitrate of the encoded stream.
6965 b Set the video bitrate in bit/s for CBR (Constant Bit Rate) mode.
6967 In case VBR (Variable Bit Rate) mode is enabled this option is
6968 ignored.
6969 flags
6971 Used to enable constant quality mode (VBR) encoding through the
6972 qscale flag, and to enable the "pass1" and "pass2" modes.
6973 g Set the GOP size.
6975 global_quality
6977 Set the global quality as an integer in lambda units.
6978 Only relevant when VBR mode is enabled with "flags +qscale". The
6980 value is converted to QP units by dividing it by "FF_QP2LAMBDA",
6981 clipped in the [0 - 10] range, and then multiplied by 6.3 to get a
6982 value in the native libtheora range [0-63]. A higher value
6983 corresponds to a higher quality.
6984 q Enable VBR mode when set to a non-negative value, and set constant
6986 quality value as a double floating point value in QP units.
6987 The value is clipped in the [0-10] range, and then multiplied by
6989 6.3 to get a value in the native libtheora range [0-63].
6990 This option is valid only using the ffmpeg command-line tool. For
6992 library interface users, use global_quality.
6993 Examples
6995 • Set maximum constant quality (VBR) encoding with ffmpeg:
6997 ffmpeg -i INPUT -codec:v libtheora -q:v 10 OUTPUT.ogg
6999 • Use ffmpeg to convert a CBR 1000 kbps Theora video stream:
7001 ffmpeg -i INPUT -codec:v libtheora -b:v 1000k OUTPUT.ogg
7003 libvpx
7005 VP8/VP9 format supported through libvpx.
7006 Requires the presence of the libvpx headers and library during
7008 configuration. You need to explicitly configure the build with
7009 "--enable-libvpx".
7010 Options
7012 The following options are supported by the libvpx wrapper. The
7014 vpxenc-equivalent options or values are listed in parentheses for easy
7015 migration.
7016 To reduce the duplication of documentation, only the private options
7018 and some others requiring special attention are documented here. For
7019 the documentation of the undocumented generic options, see the Codec
7020 Options chapter.
7021 To get more documentation of the libvpx options, invoke the command
7023 ffmpeg -h encoder=libvpx, ffmpeg -h encoder=libvpx-vp9 or vpxenc
7024 --help. Further information is available in the libvpx API
7025 documentation.
7026 b (target-bitrate)
7028 Set bitrate in bits/s. Note that FFmpeg's b option is expressed in
7029 bits/s, while vpxenc's target-bitrate is in kilobits/s.
7030 g (kf-max-dist)
7032 keyint_min (kf-min-dist)
7033 qmin (min-q)
7034 qmax (max-q)
7035 bufsize (buf-sz, buf-optimal-sz)
7036 Set ratecontrol buffer size (in bits). Note vpxenc's options are
7037 specified in milliseconds, the libvpx wrapper converts this value
7038 as follows: "buf-sz = bufsize * 1000 / bitrate", "buf-optimal-sz =
7039 bufsize * 1000 / bitrate * 5 / 6".
7040 rc_init_occupancy (buf-initial-sz)
7042 Set number of bits which should be loaded into the rc buffer before
7043 decoding starts. Note vpxenc's option is specified in milliseconds,
7044 the libvpx wrapper converts this value as follows:
7045 "rc_init_occupancy * 1000 / bitrate".
7046 undershoot-pct
7048 Set datarate undershoot (min) percentage of the target bitrate.
7049 overshoot-pct
7051 Set datarate overshoot (max) percentage of the target bitrate.
7052 skip_threshold (drop-frame)
7054 qcomp (bias-pct)
7055 maxrate (maxsection-pct)
7056 Set GOP max bitrate in bits/s. Note vpxenc's option is specified as
7057 a percentage of the target bitrate, the libvpx wrapper converts
7058 this value as follows: "(maxrate * 100 / bitrate)".
7059 minrate (minsection-pct)
7061 Set GOP min bitrate in bits/s. Note vpxenc's option is specified as
7062 a percentage of the target bitrate, the libvpx wrapper converts
7063 this value as follows: "(minrate * 100 / bitrate)".
7064 minrate, maxrate, b end-usage=cbr
7066 "(minrate == maxrate == bitrate)".
7067 crf (end-usage=cq, cq-level)
7069 tune (tune)
7070 psnr (psnr)
7071 ssim (ssim)
7072 quality, deadline (deadline)
7073 best
7074 Use best quality deadline. Poorly named and quite slow, this
7075 option should be avoided as it may give worse quality output
7076 than good.
7077 good
7079 Use good quality deadline. This is a good trade-off between
7080 speed and quality when used with the cpu-used option.
7081 realtime
7083 Use realtime quality deadline.
7084 speed, cpu-used (cpu-used)
7086 Set quality/speed ratio modifier. Higher values speed up the encode
7087 at the cost of quality.
7088 nr (noise-sensitivity)
7090 static-thresh
7091 Set a change threshold on blocks below which they will be skipped
7092 by the encoder.
7093 slices (token-parts)
7095 Note that FFmpeg's slices option gives the total number of
7096 partitions, while vpxenc's token-parts is given as
7097 "log2(partitions)".
7098 max-intra-rate
7100 Set maximum I-frame bitrate as a percentage of the target bitrate.
7101 A value of 0 means unlimited.
7102 force_key_frames
7104 "VPX_EFLAG_FORCE_KF"
7105 Alternate reference frame related
7107 auto-alt-ref
7108 Enable use of alternate reference frames (2-pass only). Values
7109 greater than 1 enable multi-layer alternate reference frames
7110 (VP9 only).
7111 arnr-maxframes
7113 Set altref noise reduction max frame count.
7114 arnr-type
7116 Set altref noise reduction filter type: backward, forward,
7117 centered.
7118 arnr-strength
7120 Set altref noise reduction filter strength.
7121 rc-lookahead, lag-in-frames (lag-in-frames)
7123 Set number of frames to look ahead for frametype and
7124 ratecontrol.
7125 error-resilient
7127 Enable error resiliency features.
7128 sharpness integer
7130 Increase sharpness at the expense of lower PSNR. The valid range
7131 is [0, 7].
7132 ts-parameters
7134 Sets the temporal scalability configuration using a :-separated
7135 list of key=value pairs. For example, to specify temporal
7136 scalability parameters with "ffmpeg":
7137 ffmpeg -i INPUT -c:v libvpx -ts-parameters ts_number_layers=3:\
7139 ts_target_bitrate=250,500,1000:ts_rate_decimator=4,2,1:\
7140 ts_periodicity=4:ts_layer_id=0,2,1,2:ts_layering_mode=3 OUTPUT
7141 Below is a brief explanation of each of the parameters, please
7143 refer to "struct vpx_codec_enc_cfg" in "vpx/vpx_encoder.h" for more
7144 details.
7145 ts_number_layers
7147 Number of temporal coding layers.
7148 ts_target_bitrate
7150 Target bitrate for each temporal layer (in kbps). (bitrate
7151 should be inclusive of the lower temporal layer).
7152 ts_rate_decimator
7154 Frame rate decimation factor for each temporal layer.
7155 ts_periodicity
7157 Length of the sequence defining frame temporal layer
7158 membership.
7159 ts_layer_id
7161 Template defining the membership of frames to temporal layers.
7162 ts_layering_mode
7164 (optional) Selecting the temporal structure from a set of pre-
7165 defined temporal layering modes. Currently supports the
7166 following options.
7167 0 No temporal layering flags are provided internally, relies
7169 on flags being passed in using "metadata" field in
7170 "AVFrame" with following keys.
7171 vp8-flags
7173 Sets the flags passed into the encoder to indicate the
7174 referencing scheme for the current frame. Refer to
7175 function "vpx_codec_encode" in "vpx/vpx_encoder.h" for
7176 more details.
7177 temporal_id
7179 Explicitly sets the temporal id of the current frame to
7180 encode.
7181 2 Two temporal layers. 0-1...
7183 3 Three temporal layers. 0-2-1-2...; with single reference
7185 frame.
7186 4 Same as option "3", except there is a dependency between
7188 the two temporal layer 2 frames within the temporal period.
7189 VP9-specific options
7191 lossless
7192 Enable lossless mode.
7193 tile-columns
7195 Set number of tile columns to use. Note this is given as
7196 "log2(tile_columns)". For example, 8 tile columns would be
7197 requested by setting the tile-columns option to 3.
7198 tile-rows
7200 Set number of tile rows to use. Note this is given as
7201 "log2(tile_rows)". For example, 4 tile rows would be requested
7202 by setting the tile-rows option to 2.
7203 frame-parallel
7205 Enable frame parallel decodability features.
7206 aq-mode
7208 Set adaptive quantization mode (0: off (default), 1: variance
7209 2: complexity, 3: cyclic refresh, 4: equator360).
7210 colorspace color-space
7212 Set input color space. The VP9 bitstream supports signaling the
7213 following colorspaces:
7214 rgb sRGB
7216 bt709 bt709
7217 unspecified unknown
7218 bt470bg bt601
7219 smpte170m smpte170
7220 smpte240m smpte240
7221 bt2020_ncl bt2020
7222 row-mt boolean
7223 Enable row based multi-threading.
7224 tune-content
7226 Set content type: default (0), screen (1), film (2).
7227 corpus-complexity
7229 Corpus VBR mode is a variant of standard VBR where the
7230 complexity distribution midpoint is passed in rather than
7231 calculated for a specific clip or chunk.
7232 The valid range is [0, 10000]. 0 (default) uses standard VBR.
7234 enable-tpl boolean
7236 Enable temporal dependency model.
7237 ref-frame-config
7239 Using per-frame metadata, set members of the structure
7240 "vpx_svc_ref_frame_config_t" in "vpx/vp8cx.h" to fine-control
7241 referencing schemes and frame buffer management. Use a
7242 :-separated list of key=value pairs. For example,
7243 av_dict_set(&av_frame->metadata, "ref-frame-config", \
7245 "rfc_update_buffer_slot=7:rfc_lst_fb_idx=0:rfc_gld_fb_idx=1:rfc_alt_fb_idx=2:rfc_reference_last=0:rfc_reference_golden=0:rfc_reference_alt_ref=0");
7246 rfc_update_buffer_slot
7248 Indicates the buffer slot number to update
7249 rfc_update_last
7251 Indicates whether to update the LAST frame
7252 rfc_update_golden
7254 Indicates whether to update GOLDEN frame
7255 rfc_update_alt_ref
7257 Indicates whether to update ALT_REF frame
7258 rfc_lst_fb_idx
7260 LAST frame buffer index
7261 rfc_gld_fb_idx
7263 GOLDEN frame buffer index
7264 rfc_alt_fb_idx
7266 ALT_REF frame buffer index
7267 rfc_reference_last
7269 Indicates whether to reference LAST frame
7270 rfc_reference_golden
7272 Indicates whether to reference GOLDEN frame
7273 rfc_reference_alt_ref
7275 Indicates whether to reference ALT_REF frame
7276 rfc_reference_duration
7278 Indicates frame duration
7279 For more information about libvpx see: <http://www.webmproject.org/>
7281 libwebp
7283 libwebp WebP Image encoder wrapper
7284 libwebp is Google's official encoder for WebP images. It can encode in
7286 either lossy or lossless mode. Lossy images are essentially a wrapper
7287 around a VP8 frame. Lossless images are a separate codec developed by
7288 Google.
7289 Pixel Format
7291 Currently, libwebp only supports YUV420 for lossy and RGB for lossless
7293 due to limitations of the format and libwebp. Alpha is supported for
7294 either mode. Because of API limitations, if RGB is passed in when
7295 encoding lossy or YUV is passed in for encoding lossless, the pixel
7296 format will automatically be converted using functions from libwebp.
7297 This is not ideal and is done only for convenience.
7298 Options
7300 -lossless boolean
7302 Enables/Disables use of lossless mode. Default is 0.
7303 -compression_level integer
7305 For lossy, this is a quality/speed tradeoff. Higher values give
7306 better quality for a given size at the cost of increased encoding
7307 time. For lossless, this is a size/speed tradeoff. Higher values
7308 give smaller size at the cost of increased encoding time. More
7309 specifically, it controls the number of extra algorithms and
7310 compression tools used, and varies the combination of these tools.
7311 This maps to the method option in libwebp. The valid range is 0 to
7312 6. Default is 4.
7313 -qscale float
7315 For lossy encoding, this controls image quality, 0 to 100. For
7316 lossless encoding, this controls the effort and time spent at
7317 compressing more. The default value is 75. Note that for usage via
7318 libavcodec, this option is called global_quality and must be
7319 multiplied by FF_QP2LAMBDA.
7320 -preset type
7322 Configuration preset. This does some automatic settings based on
7323 the general type of the image.
7324 none
7326 Do not use a preset.
7327 default
7329 Use the encoder default.
7330 picture
7332 Digital picture, like portrait, inner shot
7333 photo
7335 Outdoor photograph, with natural lighting
7336 drawing
7338 Hand or line drawing, with high-contrast details
7339 icon
7341 Small-sized colorful images
7342 text
7344 Text-like
7345 libx264, libx264rgb
7347 x264 H.264/MPEG-4 AVC encoder wrapper.
7348 This encoder requires the presence of the libx264 headers and library
7350 during configuration. You need to explicitly configure the build with
7351 "--enable-libx264".
7352 libx264 supports an impressive number of features, including 8x8 and
7354 4x4 adaptive spatial transform, adaptive B-frame placement, CAVLC/CABAC
7355 entropy coding, interlacing (MBAFF), lossless mode, psy optimizations
7356 for detail retention (adaptive quantization, psy-RD, psy-trellis).
7357 Many libx264 encoder options are mapped to FFmpeg global codec options,
7359 while unique encoder options are provided through private options.
7360 Additionally the x264opts and x264-params private options allows one to
7361 pass a list of key=value tuples as accepted by the libx264
7362 "x264_param_parse" function.
7363 The x264 project website is at
7365 <http://www.videolan.org/developers/x264.html>.
7366 The libx264rgb encoder is the same as libx264, except it accepts packed
7368 RGB pixel formats as input instead of YUV.
7369 Supported Pixel Formats
7371 x264 supports 8- to 10-bit color spaces. The exact bit depth is
7373 controlled at x264's configure time.
7374 Options
7376 The following options are supported by the libx264 wrapper. The
7378 x264-equivalent options or values are listed in parentheses for easy
7379 migration.
7380 To reduce the duplication of documentation, only the private options
7382 and some others requiring special attention are documented here. For
7383 the documentation of the undocumented generic options, see the Codec
7384 Options chapter.
7385 To get a more accurate and extensive documentation of the libx264
7387 options, invoke the command x264 --fullhelp or consult the libx264
7388 documentation.
7389 b (bitrate)
7391 Set bitrate in bits/s. Note that FFmpeg's b option is expressed in
7392 bits/s, while x264's bitrate is in kilobits/s.
7393 bf (bframes)
7395 g (keyint)
7396 qmin (qpmin)
7397 Minimum quantizer scale.
7398 qmax (qpmax)
7400 Maximum quantizer scale.
7401 qdiff (qpstep)
7403 Maximum difference between quantizer scales.
7404 qblur (qblur)
7406 Quantizer curve blur
7407 qcomp (qcomp)
7409 Quantizer curve compression factor
7410 refs (ref)
7412 Number of reference frames each P-frame can use. The range is from
7413 0-16.
7414 sc_threshold (scenecut)
7416 Sets the threshold for the scene change detection.
7417 trellis (trellis)
7419 Performs Trellis quantization to increase efficiency. Enabled by
7420 default.
7421 nr (nr)
7423 me_range (merange)
7424 Maximum range of the motion search in pixels.
7425 me_method (me)
7427 Set motion estimation method. Possible values in the decreasing
7428 order of speed:
7429 dia (dia)
7431 epzs (dia)
7432 Diamond search with radius 1 (fastest). epzs is an alias for
7433 dia.
7434 hex (hex)
7436 Hexagonal search with radius 2.
7437 umh (umh)
7439 Uneven multi-hexagon search.
7440 esa (esa)
7442 Exhaustive search.
7443 tesa (tesa)
7445 Hadamard exhaustive search (slowest).
7446 forced-idr
7448 Normally, when forcing a I-frame type, the encoder can select any
7449 type of I-frame. This option forces it to choose an IDR-frame.
7450 subq (subme)
7452 Sub-pixel motion estimation method.
7453 b_strategy (b-adapt)
7455 Adaptive B-frame placement decision algorithm. Use only on first-
7456 pass.
7457 keyint_min (min-keyint)
7459 Minimum GOP size.
7460 coder
7462 Set entropy encoder. Possible values:
7463 ac Enable CABAC.
7465 vlc Enable CAVLC and disable CABAC. It generates the same effect as
7467 x264's --no-cabac option.
7468 cmp Set full pixel motion estimation comparison algorithm. Possible
7470 values:
7471 chroma
7473 Enable chroma in motion estimation.
7474 sad Ignore chroma in motion estimation. It generates the same
7476 effect as x264's --no-chroma-me option.
7477 threads (threads)
7479 Number of encoding threads.
7480 thread_type
7482 Set multithreading technique. Possible values:
7483 slice
7485 Slice-based multithreading. It generates the same effect as
7486 x264's --sliced-threads option.
7487 frame
7489 Frame-based multithreading.
7490 flags
7492 Set encoding flags. It can be used to disable closed GOP and enable
7493 open GOP by setting it to "-cgop". The result is similar to the
7494 behavior of x264's --open-gop option.
7495 rc_init_occupancy (vbv-init)
7497 preset (preset)
7498 Set the encoding preset.
7499 tune (tune)
7501 Set tuning of the encoding params.
7502 profile (profile)
7504 Set profile restrictions.
7505 fastfirstpass
7507 Enable fast settings when encoding first pass, when set to 1. When
7508 set to 0, it has the same effect of x264's --slow-firstpass option.
7509 crf (crf)
7511 Set the quality for constant quality mode.
7512 crf_max (crf-max)
7514 In CRF mode, prevents VBV from lowering quality beyond this point.
7515 qp (qp)
7517 Set constant quantization rate control method parameter.
7518 aq-mode (aq-mode)
7520 Set AQ method. Possible values:
7521 none (0)
7523 Disabled.
7524 variance (1)
7526 Variance AQ (complexity mask).
7527 autovariance (2)
7529 Auto-variance AQ (experimental).
7530 aq-strength (aq-strength)
7532 Set AQ strength, reduce blocking and blurring in flat and textured
7533 areas.
7534 psy Use psychovisual optimizations when set to 1. When set to 0, it has
7536 the same effect as x264's --no-psy option.
7537 psy-rd (psy-rd)
7539 Set strength of psychovisual optimization, in psy-rd:psy-trellis
7540 format.
7541 rc-lookahead (rc-lookahead)
7543 Set number of frames to look ahead for frametype and ratecontrol.
7544 weightb
7546 Enable weighted prediction for B-frames when set to 1. When set to
7547 0, it has the same effect as x264's --no-weightb option.
7548 weightp (weightp)
7550 Set weighted prediction method for P-frames. Possible values:
7551 none (0)
7553 Disabled
7554 simple (1)
7556 Enable only weighted refs
7557 smart (2)
7559 Enable both weighted refs and duplicates
7560 ssim (ssim)
7562 Enable calculation and printing SSIM stats after the encoding.
7563 intra-refresh (intra-refresh)
7565 Enable the use of Periodic Intra Refresh instead of IDR frames when
7566 set to 1.
7567 avcintra-class (class)
7569 Configure the encoder to generate AVC-Intra. Valid values are
7570 50,100 and 200
7571 bluray-compat (bluray-compat)
7573 Configure the encoder to be compatible with the bluray standard.
7574 It is a shorthand for setting "bluray-compat=1 force-cfr=1".
7575 b-bias (b-bias)
7577 Set the influence on how often B-frames are used.
7578 b-pyramid (b-pyramid)
7580 Set method for keeping of some B-frames as references. Possible
7581 values:
7582 none (none)
7584 Disabled.
7585 strict (strict)
7587 Strictly hierarchical pyramid.
7588 normal (normal)
7590 Non-strict (not Blu-ray compatible).
7591 mixed-refs
7593 Enable the use of one reference per partition, as opposed to one
7594 reference per macroblock when set to 1. When set to 0, it has the
7595 same effect as x264's --no-mixed-refs option.
7596 8x8dct
7598 Enable adaptive spatial transform (high profile 8x8 transform) when
7599 set to 1. When set to 0, it has the same effect as x264's
7600 --no-8x8dct option.
7601 fast-pskip
7603 Enable early SKIP detection on P-frames when set to 1. When set to
7604 0, it has the same effect as x264's --no-fast-pskip option.
7605 aud (aud)
7607 Enable use of access unit delimiters when set to 1.
7608 mbtree
7610 Enable use macroblock tree ratecontrol when set to 1. When set to
7611 0, it has the same effect as x264's --no-mbtree option.
7612 deblock (deblock)
7614 Set loop filter parameters, in alpha:beta form.
7615 cplxblur (cplxblur)
7617 Set fluctuations reduction in QP (before curve compression).
7618 partitions (partitions)
7620 Set partitions to consider as a comma-separated list of. Possible
7621 values in the list:
7622 p8x8
7624 8x8 P-frame partition.
7625 p4x4
7627 4x4 P-frame partition.
7628 b8x8
7630 4x4 B-frame partition.
7631 i8x8
7633 8x8 I-frame partition.
7634 i4x4
7636 4x4 I-frame partition. (Enabling p4x4 requires p8x8 to be
7637 enabled. Enabling i8x8 requires adaptive spatial transform
7638 (8x8dct option) to be enabled.)
7639 none (none)
7641 Do not consider any partitions.
7642 all (all)
7644 Consider every partition.
7645 direct-pred (direct)
7647 Set direct MV prediction mode. Possible values:
7648 none (none)
7650 Disable MV prediction.
7651 spatial (spatial)
7653 Enable spatial predicting.
7654 temporal (temporal)
7656 Enable temporal predicting.
7657 auto (auto)
7659 Automatically decided.
7660 slice-max-size (slice-max-size)
7662 Set the limit of the size of each slice in bytes. If not specified
7663 but RTP payload size (ps) is specified, that is used.
7664 stats (stats)
7666 Set the file name for multi-pass stats.
7667 nal-hrd (nal-hrd)
7669 Set signal HRD information (requires vbv-bufsize to be set).
7670 Possible values:
7671 none (none)
7673 Disable HRD information signaling.
7674 vbr (vbr)
7676 Variable bit rate.
7677 cbr (cbr)
7679 Constant bit rate (not allowed in MP4 container).
7680 x264opts (N.A.)
7682 Set any x264 option, see x264 --fullhelp for a list.
7683 Argument is a list of key=value couples separated by ":". In filter
7685 and psy-rd options that use ":" as a separator themselves, use ","
7686 instead. They accept it as well since long ago but this is kept
7687 undocumented for some reason.
7688 For example to specify libx264 encoding options with ffmpeg:
7690 ffmpeg -i foo.mpg -c:v libx264 -x264opts keyint=123:min-keyint=20 -an out.mkv
7692 a53cc boolean
7694 Import closed captions (which must be ATSC compatible format) into
7695 output. Only the mpeg2 and h264 decoders provide these. Default is
7696 1 (on).
7697 udu_sei boolean
7699 Import user data unregistered SEI if available into output. Default
7700 is 0 (off).
7701 x264-params (N.A.)
7703 Override the x264 configuration using a :-separated list of
7704 key=value parameters.
7705 This option is functionally the same as the x264opts, but is
7707 duplicated for compatibility with the Libav fork.
7708 For example to specify libx264 encoding options with ffmpeg:
7710 ffmpeg -i INPUT -c:v libx264 -x264-params level=30:bframes=0:weightp=0:\
7712 cabac=0:ref=1:vbv-maxrate=768:vbv-bufsize=2000:analyse=all:me=umh:\
7713 no-fast-pskip=1:subq=6:8x8dct=0:trellis=0 OUTPUT
7714 Encoding ffpresets for common usages are provided so they can be used
7716 with the general presets system (e.g. passing the pre option).
7717 libx265
7719 x265 H.265/HEVC encoder wrapper.
7720 This encoder requires the presence of the libx265 headers and library
7722 during configuration. You need to explicitly configure the build with
7723 --enable-libx265.
7724 Options
7726 b Sets target video bitrate.
7728 bf
7730 g Set the GOP size.
7731 keyint_min
7733 Minimum GOP size.
7734 refs
7736 Number of reference frames each P-frame can use. The range is from
7737 1-16.
7738 preset
7740 Set the x265 preset.
7741 tune
7743 Set the x265 tune parameter.
7744 profile
7746 Set profile restrictions.
7747 crf Set the quality for constant quality mode.
7749 qp Set constant quantization rate control method parameter.
7751 qmin
7753 Minimum quantizer scale.
7754 qmax
7756 Maximum quantizer scale.
7757 qdiff
7759 Maximum difference between quantizer scales.
7760 qblur
7762 Quantizer curve blur
7763 qcomp
7765 Quantizer curve compression factor
7766 i_qfactor
7768 b_qfactor
7769 forced-idr
7770 Normally, when forcing a I-frame type, the encoder can select any
7771 type of I-frame. This option forces it to choose an IDR-frame.
7772 udu_sei boolean
7774 Import user data unregistered SEI if available into output. Default
7775 is 0 (off).
7776 x265-params
7778 Set x265 options using a list of key=value couples separated by
7779 ":". See x265 --help for a list of options.
7780 For example to specify libx265 encoding options with -x265-params:
7782 ffmpeg -i input -c:v libx265 -x265-params crf=26:psy-rd=1 output.mp4
7784 libxavs2
7786 xavs2 AVS2-P2/IEEE1857.4 encoder wrapper.
7787 This encoder requires the presence of the libxavs2 headers and library
7789 during configuration. You need to explicitly configure the build with
7790 --enable-libxavs2.
7791 The following standard libavcodec options are used:
7793 • b / bit_rate
7795 • g / gop_size
7797 • bf / max_b_frames
7799 The encoder also has its own specific options:
7801 Options
7803 lcu_row_threads
7805 Set the number of parallel threads for rows from 1 to 8 (default
7806 5).
7807 initial_qp
7809 Set the xavs2 quantization parameter from 1 to 63 (default 34).
7810 This is used to set the initial qp for the first frame.
7811 qp Set the xavs2 quantization parameter from 1 to 63 (default 34).
7813 This is used to set the qp value under constant-QP mode.
7814 max_qp
7816 Set the max qp for rate control from 1 to 63 (default 55).
7817 min_qp
7819 Set the min qp for rate control from 1 to 63 (default 20).
7820 speed_level
7822 Set the Speed level from 0 to 9 (default 0). Higher is better but
7823 slower.
7824 log_level
7826 Set the log level from -1 to 3 (default 0). -1: none, 0: error, 1:
7827 warning, 2: info, 3: debug.
7828 xavs2-params
7830 Set xavs2 options using a list of key=value couples separated by
7831 ":".
7832 For example to specify libxavs2 encoding options with
7834 -xavs2-params:
7835 ffmpeg -i input -c:v libxavs2 -xavs2-params RdoqLevel=0 output.avs2
7837 libxvid
7839 Xvid MPEG-4 Part 2 encoder wrapper.
7840 This encoder requires the presence of the libxvidcore headers and
7842 library during configuration. You need to explicitly configure the
7843 build with "--enable-libxvid --enable-gpl".
7844 The native "mpeg4" encoder supports the MPEG-4 Part 2 format, so users
7846 can encode to this format without this library.
7847 Options
7849 The following options are supported by the libxvid wrapper. Some of the
7851 following options are listed but are not documented, and correspond to
7852 shared codec options. See the Codec Options chapter for their
7853 documentation. The other shared options which are not listed have no
7854 effect for the libxvid encoder.
7855 b
7857 g
7858 qmin
7859 qmax
7860 mpeg_quant
7861 threads
7862 bf
7863 b_qfactor
7864 b_qoffset
7865 flags
7866 Set specific encoding flags. Possible values:
7867 mv4 Use four motion vector by macroblock.
7869 aic Enable high quality AC prediction.
7871 gray
7873 Only encode grayscale.
7874 gmc Enable the use of global motion compensation (GMC).
7876 qpel
7878 Enable quarter-pixel motion compensation.
7879 cgop
7881 Enable closed GOP.
7882 global_header
7884 Place global headers in extradata instead of every keyframe.
7885 trellis
7887 me_method
7888 Set motion estimation method. Possible values in decreasing order
7889 of speed and increasing order of quality:
7890 zero
7892 Use no motion estimation (default).
7893 phods
7895 x1
7896 log Enable advanced diamond zonal search for 16x16 blocks and half-
7897 pixel refinement for 16x16 blocks. x1 and log are aliases for
7898 phods.
7899 epzs
7901 Enable all of the things described above, plus advanced diamond
7902 zonal search for 8x8 blocks, half-pixel refinement for 8x8
7903 blocks, and motion estimation on chroma planes.
7904 full
7906 Enable all of the things described above, plus extended 16x16
7907 and 8x8 blocks search.
7908 mbd Set macroblock decision algorithm. Possible values in the
7910 increasing order of quality:
7911 simple
7913 Use macroblock comparing function algorithm (default).
7914 bits
7916 Enable rate distortion-based half pixel and quarter pixel
7917 refinement for 16x16 blocks.
7918 rd Enable all of the things described above, plus rate distortion-
7920 based half pixel and quarter pixel refinement for 8x8 blocks,
7921 and rate distortion-based search using square pattern.
7922 lumi_aq
7924 Enable lumi masking adaptive quantization when set to 1. Default is
7925 0 (disabled).
7926 variance_aq
7928 Enable variance adaptive quantization when set to 1. Default is 0
7929 (disabled).
7930 When combined with lumi_aq, the resulting quality will not be
7932 better than any of the two specified individually. In other words,
7933 the resulting quality will be the worse one of the two effects.
7934 ssim
7936 Set structural similarity (SSIM) displaying method. Possible
7937 values:
7938 off Disable displaying of SSIM information.
7940 avg Output average SSIM at the end of encoding to stdout. The
7942 format of showing the average SSIM is:
7943 Average SSIM: %f
7945 For users who are not familiar with C, %f means a float number,
7947 or a decimal (e.g. 0.939232).
7948 frame
7950 Output both per-frame SSIM data during encoding and average
7951 SSIM at the end of encoding to stdout. The format of per-frame
7952 information is:
7953 SSIM: avg: %1.3f min: %1.3f max: %1.3f
7955 For users who are not familiar with C, %1.3f means a float
7957 number rounded to 3 digits after the dot (e.g. 0.932).
7958 ssim_acc
7960 Set SSIM accuracy. Valid options are integers within the range of
7961 0-4, while 0 gives the most accurate result and 4 computes the
7962 fastest.
7963 MediaFoundation
7965 This provides wrappers to encoders (both audio and video) in the
7966 MediaFoundation framework. It can access both SW and HW encoders.
7967 Video encoders can take input in either of nv12 or yuv420p form (some
7968 encoders support both, some support only either - in practice, nv12 is
7969 the safer choice, especially among HW encoders).
7970 mpeg2
7972 MPEG-2 video encoder.
7973 Options
7975 profile
7977 Select the mpeg2 profile to encode:
7978 422
7980 high
7981 ss Spatially Scalable
7982 snr SNR Scalable
7984 main
7986 simple
7987 level
7988 Select the mpeg2 level to encode:
7989 high
7991 high1440
7992 main
7993 low
7994 seq_disp_ext integer
7995 Specifies if the encoder should write a sequence_display_extension
7996 to the output.
7997 -1
7999 auto
8000 Decide automatically to write it or not (this is the default)
8001 by checking if the data to be written is different from the
8002 default or unspecified values.
8003 0
8005 never
8006 Never write it.
8007 1
8009 always
8010 Always write it.
8011 video_format integer
8013 Specifies the video_format written into the sequence display
8014 extension indicating the source of the video pictures. The default
8015 is unspecified, can be component, pal, ntsc, secam or mac. For
8016 maximum compatibility, use component.
8017 a53cc boolean
8019 Import closed captions (which must be ATSC compatible format) into
8020 output. Default is 1 (on).
8021 png
8023 PNG image encoder.
8024 Private options
8026 dpi integer
8028 Set physical density of pixels, in dots per inch, unset by default
8029 dpm integer
8031 Set physical density of pixels, in dots per meter, unset by default
8032 ProRes
8034 Apple ProRes encoder.
8035 FFmpeg contains 2 ProRes encoders, the prores-aw and prores-ks encoder.
8037 The used encoder can be chosen with the "-vcodec" option.
8038 Private Options for prores-ks
8040 profile integer
8042 Select the ProRes profile to encode
8043 proxy
8045 lt
8046 standard
8047 hq
8048 4444
8049 4444xq
8050 quant_mat integer
8051 Select quantization matrix.
8052 auto
8054 default
8055 proxy
8056 lt
8057 standard
8058 hq
8059 If set to auto, the matrix matching the profile will be picked. If
8061 not set, the matrix providing the highest quality, default, will be
8062 picked.
8063 bits_per_mb integer
8065 How many bits to allot for coding one macroblock. Different
8066 profiles use between 200 and 2400 bits per macroblock, the maximum
8067 is 8000.
8068 mbs_per_slice integer
8070 Number of macroblocks in each slice (1-8); the default value (8)
8071 should be good in almost all situations.
8072 vendor string
8074 Override the 4-byte vendor ID. A custom vendor ID like apl0 would
8075 claim the stream was produced by the Apple encoder.
8076 alpha_bits integer
8078 Specify number of bits for alpha component. Possible values are 0,
8079 8 and 16. Use 0 to disable alpha plane coding.
8080 Speed considerations
8082 In the default mode of operation the encoder has to honor frame
8084 constraints (i.e. not produce frames with size bigger than requested)
8085 while still making output picture as good as possible. A frame
8086 containing a lot of small details is harder to compress and the encoder
8087 would spend more time searching for appropriate quantizers for each
8088 slice.
8089 Setting a higher bits_per_mb limit will improve the speed.
8091 For the fastest encoding speed set the qscale parameter (4 is the
8093 recommended value) and do not set a size constraint.
8094 QSV encoders
8096 The family of Intel QuickSync Video encoders (MPEG-2, H.264, HEVC,
8097 JPEG/MJPEG and VP9)
8098 The ratecontrol method is selected as follows:
8100 • When global_quality is specified, a quality-based mode is used.
8102 Specifically this means either
8103 - CQP - constant quantizer scale, when the qscale codec flag is
8105 also set (the -qscale ffmpeg option).
8106 - LA_ICQ - intelligent constant quality with lookahead, when the
8108 look_ahead option is also set.
8109 - ICQ -- intelligent constant quality otherwise. For the ICQ
8111 modes, global quality range is 1 to 51, with 1 being the best
8112 quality.
8113 • Otherwise, a bitrate-based mode is used. For all of those, you
8115 should specify at least the desired average bitrate with the b
8116 option.
8117 - LA - VBR with lookahead, when the look_ahead option is
8119 specified.
8120 - VCM - video conferencing mode, when the vcm option is set.
8122 - CBR - constant bitrate, when maxrate is specified and equal to
8124 the average bitrate.
8125 - VBR - variable bitrate, when maxrate is specified, but is
8127 higher than the average bitrate.
8128 - AVBR - average VBR mode, when maxrate is not specified. This
8130 mode is further configured by the avbr_accuracy and
8131 avbr_convergence options.
8132 Note that depending on your system, a different mode than the one you
8134 specified may be selected by the encoder. Set the verbosity level to
8135 verbose or higher to see the actual settings used by the QSV runtime.
8136 Additional libavcodec global options are mapped to MSDK options as
8138 follows:
8139 • g/gop_size -> GopPicSize
8141 • bf/max_b_frames+1 -> GopRefDist
8143 • rc_init_occupancy/rc_initial_buffer_occupancy -> InitialDelayInKB
8145 • slices -> NumSlice
8147 • refs -> NumRefFrame
8149 • b_strategy/b_frame_strategy -> BRefType
8151 • cgop/CLOSED_GOP codec flag -> GopOptFlag
8153 • For the CQP mode, the i_qfactor/i_qoffset and b_qfactor/b_qoffset
8155 set the difference between QPP and QPI, and QPP and QPB
8156 respectively.
8157 • Setting the coder option to the value vlc will make the H.264
8159 encoder use CAVLC instead of CABAC.
8160 snow
8162 Options
8163 iterative_dia_size
8165 dia size for the iterative motion estimation
8166 VAAPI encoders
8168 Wrappers for hardware encoders accessible via VAAPI.
8169 These encoders only accept input in VAAPI hardware surfaces. If you
8171 have input in software frames, use the hwupload filter to upload them
8172 to the GPU.
8173 The following standard libavcodec options are used:
8175 • g / gop_size
8177 • bf / max_b_frames
8179 • profile
8181 If not set, this will be determined automatically from the format
8183 of the input frames and the profiles supported by the driver.
8184 • level
8186 • b / bit_rate
8188 • maxrate / rc_max_rate
8190 • bufsize / rc_buffer_size
8192 • rc_init_occupancy / rc_initial_buffer_occupancy
8194 • compression_level
8196 Speed / quality tradeoff: higher values are faster / worse quality.
8198 • q / global_quality
8200 Size / quality tradeoff: higher values are smaller / worse quality.
8202 • qmin
8204 • qmax
8206 • i_qfactor / i_quant_factor
8208 • i_qoffset / i_quant_offset
8210 • b_qfactor / b_quant_factor
8212 • b_qoffset / b_quant_offset
8214 • slices
8216 All encoders support the following options:
8218 low_power
8220 Some drivers/platforms offer a second encoder for some codecs
8221 intended to use less power than the default encoder; setting this
8222 option will attempt to use that encoder. Note that it may support
8223 a reduced feature set, so some other options may not be available
8224 in this mode.
8225 idr_interval
8227 Set the number of normal intra frames between full-refresh (IDR)
8228 frames in open-GOP mode. The intra frames are still IRAPs, but
8229 will not include global headers and may have non-decodable leading
8230 pictures.
8231 b_depth
8233 Set the B-frame reference depth. When set to one (the default),
8234 all B-frames will refer only to P- or I-frames. When set to
8235 greater values multiple layers of B-frames will be present, frames
8236 in each layer only referring to frames in higher layers.
8237 rc_mode
8239 Set the rate control mode to use. A given driver may only support
8240 a subset of modes.
8241 Possible modes:
8243 auto
8245 Choose the mode automatically based on driver support and the
8246 other options. This is the default.
8247 CQP Constant-quality.
8249 CBR Constant-bitrate.
8251 VBR Variable-bitrate.
8253 ICQ Intelligent constant-quality.
8255 QVBR
8257 Quality-defined variable-bitrate.
8258 AVBR
8260 Average variable bitrate.
8261 Each encoder also has its own specific options:
8263 h264_vaapi
8265 profile sets the value of profile_idc and the
8266 constraint_set*_flags. level sets the value of level_idc.
8267 coder
8269 Set entropy encoder (default is cabac). Possible values:
8270 ac
8272 cabac
8273 Use CABAC.
8274 vlc
8276 cavlc
8277 Use CAVLC.
8278 aud Include access unit delimiters in the stream (not included by
8280 default).
8281 sei Set SEI message types to include. Some combination of the
8283 following values:
8284 identifier
8286 Include a user_data_unregistered message containing
8287 information about the encoder.
8288 timing
8290 Include picture timing parameters (buffering_period and
8291 pic_timing messages).
8292 recovery_point
8294 Include recovery points where appropriate (recovery_point
8295 messages).
8296 hevc_vaapi
8298 profile and level set the values of general_profile_idc and
8299 general_level_idc respectively.
8300 aud Include access unit delimiters in the stream (not included by
8302 default).
8303 tier
8305 Set general_tier_flag. This may affect the level chosen for
8306 the stream if it is not explicitly specified.
8307 sei Set SEI message types to include. Some combination of the
8309 following values:
8310 hdr Include HDR metadata if the input frames have it
8312 (mastering_display_colour_volume and content_light_level
8313 messages).
8314 tiles
8316 Set the number of tiles to encode the input video with, as
8317 columns x rows. Larger numbers allow greater parallelism in
8318 both encoding and decoding, but may decrease coding efficiency.
8319 mjpeg_vaapi
8321 Only baseline DCT encoding is supported. The encoder always uses
8322 the standard quantisation and huffman tables - global_quality
8323 scales the standard quantisation table (range 1-100).
8324 For YUV, 4:2:0, 4:2:2 and 4:4:4 subsampling modes are supported.
8326 RGB is also supported, and will create an RGB JPEG.
8327 jfif
8329 Include JFIF header in each frame (not included by default).
8330 huffman
8332 Include standard huffman tables (on by default). Turning this
8333 off will save a few hundred bytes in each output frame, but may
8334 lose compatibility with some JPEG decoders which don't fully
8335 handle MJPEG.
8336 mpeg2_vaapi
8338 profile and level set the value of profile_and_level_indication.
8339 vp8_vaapi
8341 B-frames are not supported.
8342 global_quality sets the q_idx used for non-key frames (range
8344 0-127).
8345 loop_filter_level
8347 loop_filter_sharpness
8348 Manually set the loop filter parameters.
8349 vp9_vaapi
8351 global_quality sets the q_idx used for P-frames (range 0-255).
8352 loop_filter_level
8354 loop_filter_sharpness
8355 Manually set the loop filter parameters.
8356 B-frames are supported, but the output stream is always in encode
8358 order rather than display order. If B-frames are enabled, it may
8359 be necessary to use the vp9_raw_reorder bitstream filter to modify
8360 the output stream to display frames in the correct order.
8361 Only normal frames are produced - the vp9_superframe bitstream
8363 filter may be required to produce a stream usable with all
8364 decoders.
8365 vc2
8367 SMPTE VC-2 (previously BBC Dirac Pro). This codec was primarily aimed
8368 at professional broadcasting but since it supports yuv420, yuv422 and
8369 yuv444 at 8 (limited range or full range), 10 or 12 bits, this makes it
8370 suitable for other tasks which require low overhead and low compression
8371 (like screen recording).
8372 Options
8374 b Sets target video bitrate. Usually that's around 1:6 of the
8376 uncompressed video bitrate (e.g. for 1920x1080 50fps yuv422p10
8377 that's around 400Mbps). Higher values (close to the uncompressed
8378 bitrate) turn on lossless compression mode.
8379 field_order
8381 Enables field coding when set (e.g. to tt - top field first) for
8382 interlaced inputs. Should increase compression with interlaced
8383 content as it splits the fields and encodes each separately.
8384 wavelet_depth
8386 Sets the total amount of wavelet transforms to apply, between 1 and
8387 5 (default). Lower values reduce compression and quality. Less
8388 capable decoders may not be able to handle values of wavelet_depth
8389 over 3.
8390 wavelet_type
8392 Sets the transform type. Currently only 5_3 (LeGall) and 9_7
8393 (Deslauriers-Dubuc) are implemented, with 9_7 being the one with
8394 better compression and thus is the default.
8395 slice_width
8397 slice_height
8398 Sets the slice size for each slice. Larger values result in better
8399 compression. For compatibility with other more limited decoders
8400 use slice_width of 32 and slice_height of 8.
8401 tolerance
8403 Sets the undershoot tolerance of the rate control system in
8404 percent. This is to prevent an expensive search from being run.
8405 qm Sets the quantization matrix preset to use by default or when
8407 wavelet_depth is set to 5
8408 - default Uses the default quantization matrix from the
8410 specifications, extended with values for the fifth level. This
8411 provides a good balance between keeping detail and omitting
8412 artifacts.
8413 - flat Use a completely zeroed out quantization matrix. This
8415 increases PSNR but might reduce perception. Use in bogus
8416 benchmarks.
8417 - color Reduces detail but attempts to preserve color at
8419 extremely low bitrates.
8420
8422 dvdsub
8423 This codec encodes the bitmap subtitle format that is used in DVDs.
8424 Typically they are stored in VOBSUB file pairs (*.idx + *.sub), and
8425 they can also be used in Matroska files.
8426 Options
8428 palette
8430 Specify the global palette used by the bitmaps.
8431 The format for this option is a string containing 16 24-bits
8433 hexadecimal numbers (without 0x prefix) separated by commas, for
8434 example "0d00ee, ee450d, 101010, eaeaea, 0ce60b, ec14ed, ebff0b,
8435 0d617a, 7b7b7b, d1d1d1, 7b2a0e, 0d950c, 0f007b, cf0dec, cfa80c,
8436 7c127b".
8437 even_rows_fix
8439 When set to 1, enable a work-around that makes the number of pixel
8440 rows even in all subtitles. This fixes a problem with some players
8441 that cut off the bottom row if the number is odd. The work-around
8442 just adds a fully transparent row if needed. The overhead is low,
8443 typically one byte per subtitle on average.
8444 By default, this work-around is disabled.
8446
8448 When you configure your FFmpeg build, all the supported bitstream
8449 filters are enabled by default. You can list all available ones using
8450 the configure option "--list-bsfs".
8451 You can disable all the bitstream filters using the configure option
8453 "--disable-bsfs", and selectively enable any bitstream filter using the
8454 option "--enable-bsf=BSF", or you can disable a particular bitstream
8455 filter using the option "--disable-bsf=BSF".
8456 The option "-bsfs" of the ff* tools will display the list of all the
8458 supported bitstream filters included in your build.
8459 The ff* tools have a -bsf option applied per stream, taking a comma-
8461 separated list of filters, whose parameters follow the filter name
8462 after a '='.
8463 ffmpeg -i INPUT -c:v copy -bsf:v filter1[=opt1=str1:opt2=str2][,filter2] OUTPUT
8465 Below is a description of the currently available bitstream filters,
8467 with their parameters, if any.
8468 aac_adtstoasc
8470 Convert MPEG-2/4 AAC ADTS to an MPEG-4 Audio Specific Configuration
8471 bitstream.
8472 This filter creates an MPEG-4 AudioSpecificConfig from an MPEG-2/4 ADTS
8474 header and removes the ADTS header.
8475 This filter is required for example when copying an AAC stream from a
8477 raw ADTS AAC or an MPEG-TS container to MP4A-LATM, to an FLV file, or
8478 to MOV/MP4 files and related formats such as 3GP or M4A. Please note
8479 that it is auto-inserted for MP4A-LATM and MOV/MP4 and related formats.
8480 av1_metadata
8482 Modify metadata embedded in an AV1 stream.
8483 td Insert or remove temporal delimiter OBUs in all temporal units of
8485 the stream.
8486 insert
8488 Insert a TD at the beginning of every TU which does not already
8489 have one.
8490 remove
8492 Remove the TD from the beginning of every TU which has one.
8493 color_primaries
8495 transfer_characteristics
8496 matrix_coefficients
8497 Set the color description fields in the stream (see AV1 section
8498 6.4.2).
8499 color_range
8501 Set the color range in the stream (see AV1 section 6.4.2; note that
8502 this cannot be set for streams using BT.709 primaries, sRGB
8503 transfer characteristic and identity (RGB) matrix coefficients).
8504 tv Limited range.
8506 pc Full range.
8508 chroma_sample_position
8510 Set the chroma sample location in the stream (see AV1 section
8511 6.4.2). This can only be set for 4:2:0 streams.
8512 vertical
8514 Left position (matching the default in MPEG-2 and H.264).
8515 colocated
8517 Top-left position.
8518 tick_rate
8520 Set the tick rate (time_scale / num_units_in_display_tick) in the
8521 timing info in the sequence header.
8522 num_ticks_per_picture
8524 Set the number of ticks in each picture, to indicate that the
8525 stream has a fixed framerate. Ignored if tick_rate is not also
8526 set.
8527 delete_padding
8529 Deletes Padding OBUs.
8530 chomp
8532 Remove zero padding at the end of a packet.
8533 dca_core
8535 Extract the core from a DCA/DTS stream, dropping extensions such as
8536 DTS-HD.
8537 dump_extra
8539 Add extradata to the beginning of the filtered packets except when said
8540 packets already exactly begin with the extradata that is intended to be
8541 added.
8542 freq
8544 The additional argument specifies which packets should be filtered.
8545 It accepts the values:
8546 k
8548 keyframe
8549 add extradata to all key packets
8550 e
8552 all add extradata to all packets
8553 If not specified it is assumed k.
8555 For example the following ffmpeg command forces a global header (thus
8557 disabling individual packet headers) in the H.264 packets generated by
8558 the "libx264" encoder, but corrects them by adding the header stored in
8559 extradata to the key packets:
8560 ffmpeg -i INPUT -map 0 -flags:v +global_header -c:v libx264 -bsf:v dump_extra out.ts
8562 eac3_core
8564 Extract the core from a E-AC-3 stream, dropping extra channels.
8565 extract_extradata
8567 Extract the in-band extradata.
8568 Certain codecs allow the long-term headers (e.g. MPEG-2 sequence
8570 headers, or H.264/HEVC (VPS/)SPS/PPS) to be transmitted either "in-
8571 band" (i.e. as a part of the bitstream containing the coded frames) or
8572 "out of band" (e.g. on the container level). This latter form is called
8573 "extradata" in FFmpeg terminology.
8574 This bitstream filter detects the in-band headers and makes them
8576 available as extradata.
8577 remove
8579 When this option is enabled, the long-term headers are removed from
8580 the bitstream after extraction.
8581 filter_units
8583 Remove units with types in or not in a given set from the stream.
8584 pass_types
8586 List of unit types or ranges of unit types to pass through while
8587 removing all others. This is specified as a '|'-separated list of
8588 unit type values or ranges of values with '-'.
8589 remove_types
8591 Identical to pass_types, except the units in the given set removed
8592 and all others passed through.
8593 Extradata is unchanged by this transformation, but note that if the
8595 stream contains inline parameter sets then the output may be unusable
8596 if they are removed.
8597 For example, to remove all non-VCL NAL units from an H.264 stream:
8599 ffmpeg -i INPUT -c:v copy -bsf:v 'filter_units=pass_types=1-5' OUTPUT
8601 To remove all AUDs, SEI and filler from an H.265 stream:
8603 ffmpeg -i INPUT -c:v copy -bsf:v 'filter_units=remove_types=35|38-40' OUTPUT
8605 hapqa_extract
8607 Extract Rgb or Alpha part of an HAPQA file, without recompression, in
8608 order to create an HAPQ or an HAPAlphaOnly file.
8609 texture
8611 Specifies the texture to keep.
8612 color
8614 alpha
8615 Convert HAPQA to HAPQ
8617 ffmpeg -i hapqa_inputfile.mov -c copy -bsf:v hapqa_extract=texture=color -tag:v HapY -metadata:s:v:0 encoder="HAPQ" hapq_file.mov
8619 Convert HAPQA to HAPAlphaOnly
8621 ffmpeg -i hapqa_inputfile.mov -c copy -bsf:v hapqa_extract=texture=alpha -tag:v HapA -metadata:s:v:0 encoder="HAPAlpha Only" hapalphaonly_file.mov
8623 h264_metadata
8625 Modify metadata embedded in an H.264 stream.
8626 aud Insert or remove AUD NAL units in all access units of the stream.
8628 insert
8630 remove
8631 sample_aspect_ratio
8632 Set the sample aspect ratio of the stream in the VUI parameters.
8633 overscan_appropriate_flag
8635 Set whether the stream is suitable for display using overscan or
8636 not (see H.264 section E.2.1).
8637 video_format
8639 video_full_range_flag
8640 Set the video format in the stream (see H.264 section E.2.1 and
8641 table E-2).
8642 colour_primaries
8644 transfer_characteristics
8645 matrix_coefficients
8646 Set the colour description in the stream (see H.264 section E.2.1
8647 and tables E-3, E-4 and E-5).
8648 chroma_sample_loc_type
8650 Set the chroma sample location in the stream (see H.264 section
8651 E.2.1 and figure E-1).
8652 tick_rate
8654 Set the tick rate (time_scale / num_units_in_tick) in the VUI
8655 parameters. This is the smallest time unit representable in the
8656 stream, and in many cases represents the field rate of the stream
8657 (double the frame rate).
8658 fixed_frame_rate_flag
8660 Set whether the stream has fixed framerate - typically this
8661 indicates that the framerate is exactly half the tick rate, but the
8662 exact meaning is dependent on interlacing and the picture structure
8663 (see H.264 section E.2.1 and table E-6).
8664 zero_new_constraint_set_flags
8666 Zero constraint_set4_flag and constraint_set5_flag in the SPS.
8667 These bits were reserved in a previous version of the H.264 spec,
8668 and thus some hardware decoders require these to be zero. The
8669 result of zeroing this is still a valid bitstream.
8670 crop_left
8672 crop_right
8673 crop_top
8674 crop_bottom
8675 Set the frame cropping offsets in the SPS. These values will
8676 replace the current ones if the stream is already cropped.
8677 These fields are set in pixels. Note that some sizes may not be
8679 representable if the chroma is subsampled or the stream is
8680 interlaced (see H.264 section 7.4.2.1.1).
8681 sei_user_data
8683 Insert a string as SEI unregistered user data. The argument must
8684 be of the form UUID+string, where the UUID is as hex digits
8685 possibly separated by hyphens, and the string can be anything.
8686 For example, 086f3693-b7b3-4f2c-9653-21492feee5b8+hello will insert
8688 the string ``hello'' associated with the given UUID.
8689 delete_filler
8691 Deletes both filler NAL units and filler SEI messages.
8692 level
8694 Set the level in the SPS. Refer to H.264 section A.3 and tables
8695 A-1 to A-5.
8696 The argument must be the name of a level (for example, 4.2), a
8698 level_idc value (for example, 42), or the special name auto
8699 indicating that the filter should attempt to guess the level from
8700 the input stream properties.
8701 h264_mp4toannexb
8703 Convert an H.264 bitstream from length prefixed mode to start code
8704 prefixed mode (as defined in the Annex B of the ITU-T H.264
8705 specification).
8706 This is required by some streaming formats, typically the MPEG-2
8708 transport stream format (muxer "mpegts").
8709 For example to remux an MP4 file containing an H.264 stream to mpegts
8711 format with ffmpeg, you can use the command:
8712 ffmpeg -i INPUT.mp4 -codec copy -bsf:v h264_mp4toannexb OUTPUT.ts
8714 Please note that this filter is auto-inserted for MPEG-TS (muxer
8716 "mpegts") and raw H.264 (muxer "h264") output formats.
8717 h264_redundant_pps
8719 This applies a specific fixup to some Blu-ray streams which contain
8720 redundant PPSs modifying irrelevant parameters of the stream which
8721 confuse other transformations which require correct extradata.
8722 A new single global PPS is created, and all of the redundant PPSs
8724 within the stream are removed.
8725 hevc_metadata
8727 Modify metadata embedded in an HEVC stream.
8728 aud Insert or remove AUD NAL units in all access units of the stream.
8730 insert
8732 remove
8733 sample_aspect_ratio
8734 Set the sample aspect ratio in the stream in the VUI parameters.
8735 video_format
8737 video_full_range_flag
8738 Set the video format in the stream (see H.265 section E.3.1 and
8739 table E.2).
8740 colour_primaries
8742 transfer_characteristics
8743 matrix_coefficients
8744 Set the colour description in the stream (see H.265 section E.3.1
8745 and tables E.3, E.4 and E.5).
8746 chroma_sample_loc_type
8748 Set the chroma sample location in the stream (see H.265 section
8749 E.3.1 and figure E.1).
8750 tick_rate
8752 Set the tick rate in the VPS and VUI parameters (time_scale /
8753 num_units_in_tick). Combined with num_ticks_poc_diff_one, this can
8754 set a constant framerate in the stream. Note that it is likely to
8755 be overridden by container parameters when the stream is in a
8756 container.
8757 num_ticks_poc_diff_one
8759 Set poc_proportional_to_timing_flag in VPS and VUI and use this
8760 value to set num_ticks_poc_diff_one_minus1 (see H.265 sections
8761 7.4.3.1 and E.3.1). Ignored if tick_rate is not also set.
8762 crop_left
8764 crop_right
8765 crop_top
8766 crop_bottom
8767 Set the conformance window cropping offsets in the SPS. These
8768 values will replace the current ones if the stream is already
8769 cropped.
8770 These fields are set in pixels. Note that some sizes may not be
8772 representable if the chroma is subsampled (H.265 section
8773 7.4.3.2.1).
8774 level
8776 Set the level in the VPS and SPS. See H.265 section A.4 and tables
8777 A.6 and A.7.
8778 The argument must be the name of a level (for example, 5.1), a
8780 general_level_idc value (for example, 153 for level 5.1), or the
8781 special name auto indicating that the filter should attempt to
8782 guess the level from the input stream properties.
8783 hevc_mp4toannexb
8785 Convert an HEVC/H.265 bitstream from length prefixed mode to start code
8786 prefixed mode (as defined in the Annex B of the ITU-T H.265
8787 specification).
8788 This is required by some streaming formats, typically the MPEG-2
8790 transport stream format (muxer "mpegts").
8791 For example to remux an MP4 file containing an HEVC stream to mpegts
8793 format with ffmpeg, you can use the command:
8794 ffmpeg -i INPUT.mp4 -codec copy -bsf:v hevc_mp4toannexb OUTPUT.ts
8796 Please note that this filter is auto-inserted for MPEG-TS (muxer
8798 "mpegts") and raw HEVC/H.265 (muxer "h265" or "hevc") output formats.
8799 imxdump
8801 Modifies the bitstream to fit in MOV and to be usable by the Final Cut
8802 Pro decoder. This filter only applies to the mpeg2video codec, and is
8803 likely not needed for Final Cut Pro 7 and newer with the appropriate
8804 -tag:v.
8805 For example, to remux 30 MB/sec NTSC IMX to MOV:
8807 ffmpeg -i input.mxf -c copy -bsf:v imxdump -tag:v mx3n output.mov
8809 mjpeg2jpeg
8811 Convert MJPEG/AVI1 packets to full JPEG/JFIF packets.
8812 MJPEG is a video codec wherein each video frame is essentially a JPEG
8814 image. The individual frames can be extracted without loss, e.g. by
8815 ffmpeg -i ../some_mjpeg.avi -c:v copy frames_%d.jpg
8817 Unfortunately, these chunks are incomplete JPEG images, because they
8819 lack the DHT segment required for decoding. Quoting from
8820 <http://www.digitalpreservation.gov/formats/fdd/fdd000063.shtml>:
8821 Avery Lee, writing in the rec.video.desktop newsgroup in 2001,
8823 commented that "MJPEG, or at least the MJPEG in AVIs having the MJPG
8824 fourcc, is restricted JPEG with a fixed -- and *omitted* -- Huffman
8825 table. The JPEG must be YCbCr colorspace, it must be 4:2:2, and it must
8826 use basic Huffman encoding, not arithmetic or progressive. . . . You
8827 can indeed extract the MJPEG frames and decode them with a regular JPEG
8828 decoder, but you have to prepend the DHT segment to them, or else the
8829 decoder won't have any idea how to decompress the data. The exact table
8830 necessary is given in the OpenDML spec."
8831 This bitstream filter patches the header of frames extracted from an
8833 MJPEG stream (carrying the AVI1 header ID and lacking a DHT segment) to
8834 produce fully qualified JPEG images.
8835 ffmpeg -i mjpeg-movie.avi -c:v copy -bsf:v mjpeg2jpeg frame_%d.jpg
8837 exiftran -i -9 frame*.jpg
8838 ffmpeg -i frame_%d.jpg -c:v copy rotated.avi
8839 mjpegadump
8841 Add an MJPEG A header to the bitstream, to enable decoding by
8842 Quicktime.
8843 mov2textsub
8845 Extract a representable text file from MOV subtitles, stripping the
8846 metadata header from each subtitle packet.
8847 See also the text2movsub filter.
8849 mp3decomp
8851 Decompress non-standard compressed MP3 audio headers.
8852 mpeg2_metadata
8854 Modify metadata embedded in an MPEG-2 stream.
8855 display_aspect_ratio
8857 Set the display aspect ratio in the stream.
8858 The following fixed values are supported:
8860 4/3
8862 16/9
8863 221/100
8864 Any other value will result in square pixels being signalled
8866 instead (see H.262 section 6.3.3 and table 6-3).
8867 frame_rate
8869 Set the frame rate in the stream. This is constructed from a table
8870 of known values combined with a small multiplier and divisor - if
8871 the supplied value is not exactly representable, the nearest
8872 representable value will be used instead (see H.262 section 6.3.3
8873 and table 6-4).
8874 video_format
8876 Set the video format in the stream (see H.262 section 6.3.6 and
8877 table 6-6).
8878 colour_primaries
8880 transfer_characteristics
8881 matrix_coefficients
8882 Set the colour description in the stream (see H.262 section 6.3.6
8883 and tables 6-7, 6-8 and 6-9).
8884 mpeg4_unpack_bframes
8886 Unpack DivX-style packed B-frames.
8887 DivX-style packed B-frames are not valid MPEG-4 and were only a
8889 workaround for the broken Video for Windows subsystem. They use more
8890 space, can cause minor AV sync issues, require more CPU power to decode
8891 (unless the player has some decoded picture queue to compensate the
8892 2,0,2,0 frame per packet style) and cause trouble if copied into a
8893 standard container like mp4 or mpeg-ps/ts, because MPEG-4 decoders may
8894 not be able to decode them, since they are not valid MPEG-4.
8895 For example to fix an AVI file containing an MPEG-4 stream with DivX-
8897 style packed B-frames using ffmpeg, you can use the command:
8898 ffmpeg -i INPUT.avi -codec copy -bsf:v mpeg4_unpack_bframes OUTPUT.avi
8900 noise
8902 Damages the contents of packets or simply drops them without damaging
8903 the container. Can be used for fuzzing or testing error
8904 resilience/concealment.
8905 Parameters:
8907 amount
8909 Accepts an expression whose evaluation per-packet determines how
8910 often bytes in that packet will be modified. A value below 0 will
8911 result in a variable frequency. Default is 0 which results in no
8912 modification. However, if neither amount nor drop is specified,
8913 amount will be set to -1. See below for accepted variables.
8914 drop
8916 Accepts an expression evaluated per-packet whose value determines
8917 whether that packet is dropped. Evaluation to a positive value
8918 results in the packet being dropped. Evaluation to a negative value
8919 results in a variable chance of it being dropped, roughly inverse
8920 in proportion to the magnitude of the value. Default is 0 which
8921 results in no drops. See below for accepted variables.
8922 dropamount
8924 Accepts a non-negative integer, which assigns a variable chance of
8925 it being dropped, roughly inverse in proportion to the value.
8926 Default is 0 which results in no drops. This option is kept for
8927 backwards compatibility and is equivalent to setting drop to a
8928 negative value with the same magnitude i.e. "dropamount=4" is the
8929 same as "drop=-4". Ignored if drop is also specified.
8930 Both "amount" and "drop" accept expressions containing the following
8932 variables:
8933 n The index of the packet, starting from zero.
8935 tb The timebase for packet timestamps.
8937 pts Packet presentation timestamp.
8939 dts Packet decoding timestamp.
8941 nopts
8943 Constant representing AV_NOPTS_VALUE.
8944 startpts
8946 First non-AV_NOPTS_VALUE PTS seen in the stream.
8947 startdts
8949 First non-AV_NOPTS_VALUE DTS seen in the stream.
8950 duration
8952 d Packet duration, in timebase units.
8953 pos Packet position in input; may be -1 when unknown or not set.
8955 size
8957 Packet size, in bytes.
8958 key Whether packet is marked as a keyframe.
8960 state
8962 A pseudo random integer, primarily derived from the content of
8963 packet payload.
8964 Examples
8966 Apply modification to every byte but don't drop any packets.
8968 ffmpeg -i INPUT -c copy -bsf noise=1 output.mkv
8970 Drop every video packet not marked as a keyframe after timestamp 30s
8972 but do not modify any of the remaining packets.
8973 ffmpeg -i INPUT -c copy -bsf:v noise=drop='gt(t\,30)*not(key)' output.mkv
8975 Drop one second of audio every 10 seconds and add some random noise to
8977 the rest.
8978 ffmpeg -i INPUT -c copy -bsf:a noise=amount=-1:drop='between(mod(t\,10)\,9\,10)' output.mkv
8980 null
8982 This bitstream filter passes the packets through unchanged.
8983 pcm_rechunk
8985 Repacketize PCM audio to a fixed number of samples per packet or a
8986 fixed packet rate per second. This is similar to the asetnsamples audio
8987 filter but works on audio packets instead of audio frames.
8988 nb_out_samples, n
8990 Set the number of samples per each output audio packet. The number
8991 is intended as the number of samples per each channel. Default
8992 value is 1024.
8993 pad, p
8995 If set to 1, the filter will pad the last audio packet with
8996 silence, so that it will contain the same number of samples (or
8997 roughly the same number of samples, see frame_rate) as the previous
8998 ones. Default value is 1.
8999 frame_rate, r
9001 This option makes the filter output a fixed number of packets per
9002 second instead of a fixed number of samples per packet. If the
9003 audio sample rate is not divisible by the frame rate then the
9004 number of samples will not be constant but will vary slightly so
9005 that each packet will start as close to the frame boundary as
9006 possible. Using this option has precedence over nb_out_samples.
9007 You can generate the well known 1602-1601-1602-1601-1602 pattern of
9009 48kHz audio for NTSC frame rate using the frame_rate option.
9010 ffmpeg -f lavfi -i sine=r=48000:d=1 -c pcm_s16le -bsf pcm_rechunk=r=30000/1001 -f framecrc -
9012 prores_metadata
9014 Modify color property metadata embedded in prores stream.
9015 color_primaries
9017 Set the color primaries. Available values are:
9018 auto
9020 Keep the same color primaries property (default).
9021 unknown
9023 bt709
9024 bt470bg
9025 BT601 625
9026 smpte170m
9028 BT601 525
9029 bt2020
9031 smpte431
9032 DCI P3
9033 smpte432
9035 P3 D65
9036 transfer_characteristics
9038 Set the color transfer. Available values are:
9039 auto
9041 Keep the same transfer characteristics property (default).
9042 unknown
9044 bt709
9045 BT 601, BT 709, BT 2020
9046 smpte2084
9048 SMPTE ST 2084
9049 arib-std-b67
9051 ARIB STD-B67
9052 matrix_coefficients
9054 Set the matrix coefficient. Available values are:
9055 auto
9057 Keep the same colorspace property (default).
9058 unknown
9060 bt709
9061 smpte170m
9062 BT 601
9063 bt2020nc
9065 Set Rec709 colorspace for each frame of the file
9067 ffmpeg -i INPUT -c copy -bsf:v prores_metadata=color_primaries=bt709:color_trc=bt709:colorspace=bt709 output.mov
9069 Set Hybrid Log-Gamma parameters for each frame of the file
9071 ffmpeg -i INPUT -c copy -bsf:v prores_metadata=color_primaries=bt2020:color_trc=arib-std-b67:colorspace=bt2020nc output.mov
9073 remove_extra
9075 Remove extradata from packets.
9076 It accepts the following parameter:
9078 freq
9080 Set which frame types to remove extradata from.
9081 k Remove extradata from non-keyframes only.
9083 keyframe
9085 Remove extradata from keyframes only.
9086 e, all
9088 Remove extradata from all frames.
9089 setts
9091 Set PTS and DTS in packets.
9092 It accepts the following parameters:
9094 ts
9096 pts
9097 dts Set expressions for PTS, DTS or both.
9098 The expressions are evaluated through the eval API and can contain the
9100 following constants:
9101 N The count of the input packet. Starting from 0.
9103 TS The demux timestamp in input in case of "ts" or "dts" option or
9105 presentation timestamp in case of "pts" option.
9106 POS The original position in the file of the packet, or undefined if
9108 undefined for the current packet
9109 DTS The demux timestamp in input.
9111 PTS The presentation timestamp in input.
9113 STARTDTS
9115 The DTS of the first packet.
9116 STARTPTS
9118 The PTS of the first packet.
9119 PREV_INDTS
9121 The previous input DTS.
9122 PREV_INPTS
9124 The previous input PTS.
9125 PREV_OUTDTS
9127 The previous output DTS.
9128 PREV_OUTPTS
9130 The previous output PTS.
9131 TB The timebase of stream packet belongs.
9133 SR The sample rate of stream packet belongs.
9135 NOPTS
9137 The AV_NOPTS_VALUE constant.
9138 text2movsub
9140 Convert text subtitles to MOV subtitles (as used by the "mov_text"
9141 codec) with metadata headers.
9142 See also the mov2textsub filter.
9144 trace_headers
9146 Log trace output containing all syntax elements in the coded stream
9147 headers (everything above the level of individual coded blocks). This
9148 can be useful for debugging low-level stream issues.
9149 Supports AV1, H.264, H.265, (M)JPEG, MPEG-2 and VP9, but depending on
9151 the build only a subset of these may be available.
9152 truehd_core
9154 Extract the core from a TrueHD stream, dropping ATMOS data.
9155 vp9_metadata
9157 Modify metadata embedded in a VP9 stream.
9158 color_space
9160 Set the color space value in the frame header. Note that any frame
9161 set to RGB will be implicitly set to PC range and that RGB is
9162 incompatible with profiles 0 and 2.
9163 unknown
9165 bt601
9166 bt709
9167 smpte170
9168 smpte240
9169 bt2020
9170 rgb
9171 color_range
9172 Set the color range value in the frame header. Note that any value
9173 imposed by the color space will take precedence over this value.
9174 tv
9176 pc
9177 vp9_superframe
9179 Merge VP9 invisible (alt-ref) frames back into VP9 superframes. This
9180 fixes merging of split/segmented VP9 streams where the alt-ref frame
9181 was split from its visible counterpart.
9182 vp9_superframe_split
9184 Split VP9 superframes into single frames.
9185 vp9_raw_reorder
9187 Given a VP9 stream with correct timestamps but possibly out of order,
9188 insert additional show-existing-frame packets to correct the ordering.
9189
9191 The libavformat library provides some generic global options, which can
9192 be set on all the muxers and demuxers. In addition each muxer or
9193 demuxer may support so-called private options, which are specific for
9194 that component.
9195 Options may be set by specifying -option value in the FFmpeg tools, or
9197 by setting the value explicitly in the "AVFormatContext" options or
9198 using the libavutil/opt.h API for programmatic use.
9199 The list of supported options follows:
9201 avioflags flags (input/output)
9203 Possible values:
9204 direct
9206 Reduce buffering.
9207 probesize integer (input)
9209 Set probing size in bytes, i.e. the size of the data to analyze to
9210 get stream information. A higher value will enable detecting more
9211 information in case it is dispersed into the stream, but will
9212 increase latency. Must be an integer not lesser than 32. It is
9213 5000000 by default.
9214 max_probe_packets integer (input)
9216 Set the maximum number of buffered packets when probing a codec.
9217 Default is 2500 packets.
9218 packetsize integer (output)
9220 Set packet size.
9221 fflags flags
9223 Set format flags. Some are implemented for a limited number of
9224 formats.
9225 Possible values for input files:
9227 discardcorrupt
9229 Discard corrupted packets.
9230 fastseek
9232 Enable fast, but inaccurate seeks for some formats.
9233 genpts
9235 Generate missing PTS if DTS is present.
9236 igndts
9238 Ignore DTS if PTS is set. Inert when nofillin is set.
9239 ignidx
9241 Ignore index.
9242 nobuffer
9244 Reduce the latency introduced by buffering during initial input
9245 streams analysis.
9246 nofillin
9248 Do not fill in missing values in packet fields that can be
9249 exactly calculated.
9250 noparse
9252 Disable AVParsers, this needs "+nofillin" too.
9253 sortdts
9255 Try to interleave output packets by DTS. At present, available
9256 only for AVIs with an index.
9257 Possible values for output files:
9259 autobsf
9261 Automatically apply bitstream filters as required by the output
9262 format. Enabled by default.
9263 bitexact
9265 Only write platform-, build- and time-independent data. This
9266 ensures that file and data checksums are reproducible and match
9267 between platforms. Its primary use is for regression testing.
9268 flush_packets
9270 Write out packets immediately.
9271 shortest
9273 Stop muxing at the end of the shortest stream. It may be
9274 needed to increase max_interleave_delta to avoid flushing the
9275 longer streams before EOF.
9276 seek2any integer (input)
9278 Allow seeking to non-keyframes on demuxer level when supported if
9279 set to 1. Default is 0.
9280 analyzeduration integer (input)
9282 Specify how many microseconds are analyzed to probe the input. A
9283 higher value will enable detecting more accurate information, but
9284 will increase latency. It defaults to 5,000,000 microseconds = 5
9285 seconds.
9286 cryptokey hexadecimal string (input)
9288 Set decryption key.
9289 indexmem integer (input)
9291 Set max memory used for timestamp index (per stream).
9292 rtbufsize integer (input)
9294 Set max memory used for buffering real-time frames.
9295 fdebug flags (input/output)
9297 Print specific debug info.
9298 Possible values:
9300 ts
9302 max_delay integer (input/output)
9303 Set maximum muxing or demuxing delay in microseconds.
9304 fpsprobesize integer (input)
9306 Set number of frames used to probe fps.
9307 audio_preload integer (output)
9309 Set microseconds by which audio packets should be interleaved
9310 earlier.
9311 chunk_duration integer (output)
9313 Set microseconds for each chunk.
9314 chunk_size integer (output)
9316 Set size in bytes for each chunk.
9317 err_detect, f_err_detect flags (input)
9319 Set error detection flags. "f_err_detect" is deprecated and should
9320 be used only via the ffmpeg tool.
9321 Possible values:
9323 crccheck
9325 Verify embedded CRCs.
9326 bitstream
9328 Detect bitstream specification deviations.
9329 buffer
9331 Detect improper bitstream length.
9332 explode
9334 Abort decoding on minor error detection.
9335 careful
9337 Consider things that violate the spec and have not been seen in
9338 the wild as errors.
9339 compliant
9341 Consider all spec non compliancies as errors.
9342 aggressive
9344 Consider things that a sane encoder should not do as an error.
9345 max_interleave_delta integer (output)
9347 Set maximum buffering duration for interleaving. The duration is
9348 expressed in microseconds, and defaults to 10000000 (10 seconds).
9349 To ensure all the streams are interleaved correctly, libavformat
9351 will wait until it has at least one packet for each stream before
9352 actually writing any packets to the output file. When some streams
9353 are "sparse" (i.e. there are large gaps between successive
9354 packets), this can result in excessive buffering.
9355 This field specifies the maximum difference between the timestamps
9357 of the first and the last packet in the muxing queue, above which
9358 libavformat will output a packet regardless of whether it has
9359 queued a packet for all the streams.
9360 If set to 0, libavformat will continue buffering packets until it
9362 has a packet for each stream, regardless of the maximum timestamp
9363 difference between the buffered packets.
9364 use_wallclock_as_timestamps integer (input)
9366 Use wallclock as timestamps if set to 1. Default is 0.
9367 avoid_negative_ts integer (output)
9369 Possible values:
9370 make_non_negative
9372 Shift timestamps to make them non-negative. Also note that
9373 this affects only leading negative timestamps, and not non-
9374 monotonic negative timestamps.
9375 make_zero
9377 Shift timestamps so that the first timestamp is 0.
9378 auto (default)
9380 Enables shifting when required by the target format.
9381 disabled
9383 Disables shifting of timestamp.
9384 When shifting is enabled, all output timestamps are shifted by the
9386 same amount. Audio, video, and subtitles desynching and relative
9387 timestamp differences are preserved compared to how they would have
9388 been without shifting.
9389 skip_initial_bytes integer (input)
9391 Set number of bytes to skip before reading header and frames if set
9392 to 1. Default is 0.
9393 correct_ts_overflow integer (input)
9395 Correct single timestamp overflows if set to 1. Default is 1.
9396 flush_packets integer (output)
9398 Flush the underlying I/O stream after each packet. Default is -1
9399 (auto), which means that the underlying protocol will decide, 1
9400 enables it, and has the effect of reducing the latency, 0 disables
9401 it and may increase IO throughput in some cases.
9402 output_ts_offset offset (output)
9404 Set the output time offset.
9405 offset must be a time duration specification, see the Time duration
9407 section in the ffmpeg-utils(1) manual.
9408 The offset is added by the muxer to the output timestamps.
9410 Specifying a positive offset means that the corresponding streams
9412 are delayed bt the time duration specified in offset. Default value
9413 is 0 (meaning that no offset is applied).
9414 format_whitelist list (input)
9416 "," separated list of allowed demuxers. By default all are allowed.
9417 dump_separator string (input)
9419 Separator used to separate the fields printed on the command line
9420 about the Stream parameters. For example, to separate the fields
9421 with newlines and indentation:
9422 ffprobe -dump_separator "
9424 " -i ~/videos/matrixbench_mpeg2.mpg
9425 max_streams integer (input)
9427 Specifies the maximum number of streams. This can be used to reject
9428 files that would require too many resources due to a large number
9429 of streams.
9430 skip_estimate_duration_from_pts bool (input)
9432 Skip estimation of input duration when calculated using PTS. At
9433 present, applicable for MPEG-PS and MPEG-TS.
9434 strict, f_strict integer (input/output)
9436 Specify how strictly to follow the standards. "f_strict" is
9437 deprecated and should be used only via the ffmpeg tool.
9438 Possible values:
9440 very
9442 strictly conform to an older more strict version of the spec or
9443 reference software
9444 strict
9446 strictly conform to all the things in the spec no matter what
9447 consequences
9448 normal
9450 unofficial
9451 allow unofficial extensions
9452 experimental
9454 allow non standardized experimental things, experimental
9455 (unfinished/work in progress/not well tested) decoders and
9456 encoders. Note: experimental decoders can pose a security
9457 risk, do not use this for decoding untrusted input.
9458 Format stream specifiers
9460 Format stream specifiers allow selection of one or more streams that
9461 match specific properties.
9462 The exact semantics of stream specifiers is defined by the
9464 "avformat_match_stream_specifier()" function declared in the
9465 libavformat/avformat.h header and documented in the Stream specifiers
9466 section in the ffmpeg(1) manual.
9467
9469 Demuxers are configured elements in FFmpeg that can read the multimedia
9470 streams from a particular type of file.
9471 When you configure your FFmpeg build, all the supported demuxers are
9473 enabled by default. You can list all available ones using the configure
9474 option "--list-demuxers".
9475 You can disable all the demuxers using the configure option
9477 "--disable-demuxers", and selectively enable a single demuxer with the
9478 option "--enable-demuxer=DEMUXER", or disable it with the option
9479 "--disable-demuxer=DEMUXER".
9480 The option "-demuxers" of the ff* tools will display the list of
9482 enabled demuxers. Use "-formats" to view a combined list of enabled
9483 demuxers and muxers.
9484 The description of some of the currently available demuxers follows.
9486 aa
9488 Audible Format 2, 3, and 4 demuxer.
9489 This demuxer is used to demux Audible Format 2, 3, and 4 (.aa) files.
9491 aac
9493 Raw Audio Data Transport Stream AAC demuxer.
9494 This demuxer is used to demux an ADTS input containing a single AAC
9496 stream alongwith any ID3v1/2 or APE tags in it.
9497 apng
9499 Animated Portable Network Graphics demuxer.
9500 This demuxer is used to demux APNG files. All headers, but the PNG
9502 signature, up to (but not including) the first fcTL chunk are
9503 transmitted as extradata. Frames are then split as being all the
9504 chunks between two fcTL ones, or between the last fcTL and IEND chunks.
9505 -ignore_loop bool
9507 Ignore the loop variable in the file if set. Default is enabled.
9508 -max_fps int
9510 Maximum framerate in frames per second. Default of 0 imposes no
9511 limit.
9512 -default_fps int
9514 Default framerate in frames per second when none is specified in
9515 the file (0 meaning as fast as possible). Default is 15.
9516 asf
9518 Advanced Systems Format demuxer.
9519 This demuxer is used to demux ASF files and MMS network streams.
9521 -no_resync_search bool
9523 Do not try to resynchronize by looking for a certain optional start
9524 code.
9525 concat
9527 Virtual concatenation script demuxer.
9528 This demuxer reads a list of files and other directives from a text
9530 file and demuxes them one after the other, as if all their packets had
9531 been muxed together.
9532 The timestamps in the files are adjusted so that the first file starts
9534 at 0 and each next file starts where the previous one finishes. Note
9535 that it is done globally and may cause gaps if all streams do not have
9536 exactly the same length.
9537 All files must have the same streams (same codecs, same time base,
9539 etc.).
9540 The duration of each file is used to adjust the timestamps of the next
9542 file: if the duration is incorrect (because it was computed using the
9543 bit-rate or because the file is truncated, for example), it can cause
9544 artifacts. The "duration" directive can be used to override the
9545 duration stored in each file.
9546 Syntax
9548 The script is a text file in extended-ASCII, with one directive per
9550 line. Empty lines, leading spaces and lines starting with '#' are
9551 ignored. The following directive is recognized:
9552 "file path"
9554 Path to a file to read; special characters and spaces must be
9555 escaped with backslash or single quotes.
9556 All subsequent file-related directives apply to that file.
9558 "ffconcat version 1.0"
9560 Identify the script type and version.
9561 To make FFmpeg recognize the format automatically, this directive
9563 must appear exactly as is (no extra space or byte-order-mark) on
9564 the very first line of the script.
9565 "duration dur"
9567 Duration of the file. This information can be specified from the
9568 file; specifying it here may be more efficient or help if the
9569 information from the file is not available or accurate.
9570 If the duration is set for all files, then it is possible to seek
9572 in the whole concatenated video.
9573 "inpoint timestamp"
9575 In point of the file. When the demuxer opens the file it instantly
9576 seeks to the specified timestamp. Seeking is done so that all
9577 streams can be presented successfully at In point.
9578 This directive works best with intra frame codecs, because for non-
9580 intra frame ones you will usually get extra packets before the
9581 actual In point and the decoded content will most likely contain
9582 frames before In point too.
9583 For each file, packets before the file In point will have
9585 timestamps less than the calculated start timestamp of the file
9586 (negative in case of the first file), and the duration of the files
9587 (if not specified by the "duration" directive) will be reduced
9588 based on their specified In point.
9589 Because of potential packets before the specified In point, packet
9591 timestamps may overlap between two concatenated files.
9592 "outpoint timestamp"
9594 Out point of the file. When the demuxer reaches the specified
9595 decoding timestamp in any of the streams, it handles it as an end
9596 of file condition and skips the current and all the remaining
9597 packets from all streams.
9598 Out point is exclusive, which means that the demuxer will not
9600 output packets with a decoding timestamp greater or equal to Out
9601 point.
9602 This directive works best with intra frame codecs and formats where
9604 all streams are tightly interleaved. For non-intra frame codecs you
9605 will usually get additional packets with presentation timestamp
9606 after Out point therefore the decoded content will most likely
9607 contain frames after Out point too. If your streams are not tightly
9608 interleaved you may not get all the packets from all streams before
9609 Out point and you may only will be able to decode the earliest
9610 stream until Out point.
9611 The duration of the files (if not specified by the "duration"
9613 directive) will be reduced based on their specified Out point.
9614 "file_packet_metadata key=value"
9616 Metadata of the packets of the file. The specified metadata will be
9617 set for each file packet. You can specify this directive multiple
9618 times to add multiple metadata entries. This directive is
9619 deprecated, use "file_packet_meta" instead.
9620 "file_packet_meta key value"
9622 Metadata of the packets of the file. The specified metadata will be
9623 set for each file packet. You can specify this directive multiple
9624 times to add multiple metadata entries.
9625 "option key value"
9627 Option to access, open and probe the file. Can be present multiple
9628 times.
9629 "stream"
9631 Introduce a stream in the virtual file. All subsequent stream-
9632 related directives apply to the last introduced stream. Some
9633 streams properties must be set in order to allow identifying the
9634 matching streams in the subfiles. If no streams are defined in the
9635 script, the streams from the first file are copied.
9636 "exact_stream_id id"
9638 Set the id of the stream. If this directive is given, the string
9639 with the corresponding id in the subfiles will be used. This is
9640 especially useful for MPEG-PS (VOB) files, where the order of the
9641 streams is not reliable.
9642 "stream_meta key value"
9644 Metadata for the stream. Can be present multiple times.
9645 "stream_codec value"
9647 Codec for the stream.
9648 "stream_extradata hex_string"
9650 Extradata for the string, encoded in hexadecimal.
9651 "chapter id start end"
9653 Add a chapter. id is an unique identifier, possibly small and
9654 consecutive.
9655 Options
9657 This demuxer accepts the following option:
9659 safe
9661 If set to 1, reject unsafe file paths and directives. A file path
9662 is considered safe if it does not contain a protocol specification
9663 and is relative and all components only contain characters from the
9664 portable character set (letters, digits, period, underscore and
9665 hyphen) and have no period at the beginning of a component.
9666 If set to 0, any file name is accepted.
9668 The default is 1.
9670 auto_convert
9672 If set to 1, try to perform automatic conversions on packet data to
9673 make the streams concatenable. The default is 1.
9674 Currently, the only conversion is adding the h264_mp4toannexb
9676 bitstream filter to H.264 streams in MP4 format. This is necessary
9677 in particular if there are resolution changes.
9678 segment_time_metadata
9680 If set to 1, every packet will contain the lavf.concat.start_time
9681 and the lavf.concat.duration packet metadata values which are the
9682 start_time and the duration of the respective file segments in the
9683 concatenated output expressed in microseconds. The duration
9684 metadata is only set if it is known based on the concat file. The
9685 default is 0.
9686 Examples
9688 • Use absolute filenames and include some comments:
9690 # my first filename
9692 file /mnt/share/file-1.wav
9693 # my second filename including whitespace
9694 file '/mnt/share/file 2.wav'
9695 # my third filename including whitespace plus single quote
9696 file '/mnt/share/file 3'\''.wav'
9697 • Allow for input format auto-probing, use safe filenames and set the
9699 duration of the first file:
9700 ffconcat version 1.0
9702 file file-1.wav
9704 duration 20.0
9705 file subdir/file-2.wav
9707 dash
9709 Dynamic Adaptive Streaming over HTTP demuxer.
9710 This demuxer presents all AVStreams found in the manifest. By setting
9712 the discard flags on AVStreams the caller can decide which streams to
9713 actually receive. Each stream mirrors the "id" and "bandwidth"
9714 properties from the "<Representation>" as metadata keys named "id" and
9715 "variant_bitrate" respectively.
9716 imf
9718 Interoperable Master Format demuxer.
9719 This demuxer presents audio and video streams found in an IMF
9721 Composition.
9722 flv, live_flv, kux
9724 Adobe Flash Video Format demuxer.
9725 This demuxer is used to demux FLV files and RTMP network streams. In
9727 case of live network streams, if you force format, you may use live_flv
9728 option instead of flv to survive timestamp discontinuities. KUX is a
9729 flv variant used on the Youku platform.
9730 ffmpeg -f flv -i myfile.flv ...
9732 ffmpeg -f live_flv -i rtmp://<any.server>/anything/key ....
9733 -flv_metadata bool
9735 Allocate the streams according to the onMetaData array content.
9736 -flv_ignore_prevtag bool
9738 Ignore the size of previous tag value.
9739 -flv_full_metadata bool
9741 Output all context of the onMetadata.
9742 gif
9744 Animated GIF demuxer.
9745 It accepts the following options:
9747 min_delay
9749 Set the minimum valid delay between frames in hundredths of
9750 seconds. Range is 0 to 6000. Default value is 2.
9751 max_gif_delay
9753 Set the maximum valid delay between frames in hundredth of seconds.
9754 Range is 0 to 65535. Default value is 65535 (nearly eleven
9755 minutes), the maximum value allowed by the specification.
9756 default_delay
9758 Set the default delay between frames in hundredths of seconds.
9759 Range is 0 to 6000. Default value is 10.
9760 ignore_loop
9762 GIF files can contain information to loop a certain number of times
9763 (or infinitely). If ignore_loop is set to 1, then the loop setting
9764 from the input will be ignored and looping will not occur. If set
9765 to 0, then looping will occur and will cycle the number of times
9766 according to the GIF. Default value is 1.
9767 For example, with the overlay filter, place an infinitely looping GIF
9769 over another video:
9770 ffmpeg -i input.mp4 -ignore_loop 0 -i input.gif -filter_complex overlay=shortest=1 out.mkv
9772 Note that in the above example the shortest option for overlay filter
9774 is used to end the output video at the length of the shortest input
9775 file, which in this case is input.mp4 as the GIF in this example loops
9776 infinitely.
9777 hls
9779 HLS demuxer
9780 Apple HTTP Live Streaming demuxer.
9782 This demuxer presents all AVStreams from all variant streams. The id
9784 field is set to the bitrate variant index number. By setting the
9785 discard flags on AVStreams (by pressing 'a' or 'v' in ffplay), the
9786 caller can decide which variant streams to actually receive. The total
9787 bitrate of the variant that the stream belongs to is available in a
9788 metadata key named "variant_bitrate".
9789 It accepts the following options:
9791 live_start_index
9793 segment index to start live streams at (negative values are from
9794 the end).
9795 allowed_extensions
9797 ',' separated list of file extensions that hls is allowed to
9798 access.
9799 max_reload
9801 Maximum number of times a insufficient list is attempted to be
9802 reloaded. Default value is 1000.
9803 m3u8_hold_counters
9805 The maximum number of times to load m3u8 when it refreshes without
9806 new segments. Default value is 1000.
9807 http_persistent
9809 Use persistent HTTP connections. Applicable only for HTTP streams.
9810 Enabled by default.
9811 http_multiple
9813 Use multiple HTTP connections for downloading HTTP segments.
9814 Enabled by default for HTTP/1.1 servers.
9815 http_seekable
9817 Use HTTP partial requests for downloading HTTP segments. 0 =
9818 disable, 1 = enable, -1 = auto, Default is auto.
9819 seg_format_options
9821 Set options for the demuxer of media segments using a list of
9822 key=value pairs separated by ":".
9823 image2
9825 Image file demuxer.
9826 This demuxer reads from a list of image files specified by a pattern.
9828 The syntax and meaning of the pattern is specified by the option
9829 pattern_type.
9830 The pattern may contain a suffix which is used to automatically
9832 determine the format of the images contained in the files.
9833 The size, the pixel format, and the format of each image must be the
9835 same for all the files in the sequence.
9836 This demuxer accepts the following options:
9838 framerate
9840 Set the frame rate for the video stream. It defaults to 25.
9841 loop
9843 If set to 1, loop over the input. Default value is 0.
9844 pattern_type
9846 Select the pattern type used to interpret the provided filename.
9847 pattern_type accepts one of the following values.
9849 none
9851 Disable pattern matching, therefore the video will only contain
9852 the specified image. You should use this option if you do not
9853 want to create sequences from multiple images and your
9854 filenames may contain special pattern characters.
9855 sequence
9857 Select a sequence pattern type, used to specify a sequence of
9858 files indexed by sequential numbers.
9859 A sequence pattern may contain the string "%d" or "%0Nd", which
9861 specifies the position of the characters representing a
9862 sequential number in each filename matched by the pattern. If
9863 the form "%d0Nd" is used, the string representing the number in
9864 each filename is 0-padded and N is the total number of 0-padded
9865 digits representing the number. The literal character '%' can
9866 be specified in the pattern with the string "%%".
9867 If the sequence pattern contains "%d" or "%0Nd", the first
9869 filename of the file list specified by the pattern must contain
9870 a number inclusively contained between start_number and
9871 start_number+start_number_range-1, and all the following
9872 numbers must be sequential.
9873 For example the pattern "img-%03d.bmp" will match a sequence of
9875 filenames of the form img-001.bmp, img-002.bmp, ...,
9876 img-010.bmp, etc.; the pattern "i%%m%%g-%d.jpg" will match a
9877 sequence of filenames of the form i%m%g-1.jpg, i%m%g-2.jpg,
9878 ..., i%m%g-10.jpg, etc.
9879 Note that the pattern must not necessarily contain "%d" or
9881 "%0Nd", for example to convert a single image file img.jpeg you
9882 can employ the command:
9883 ffmpeg -i img.jpeg img.png
9885 glob
9887 Select a glob wildcard pattern type.
9888 The pattern is interpreted like a "glob()" pattern. This is
9890 only selectable if libavformat was compiled with globbing
9891 support.
9892 glob_sequence (deprecated, will be removed)
9894 Select a mixed glob wildcard/sequence pattern.
9895 If your version of libavformat was compiled with globbing
9897 support, and the provided pattern contains at least one glob
9898 meta character among "%*?[]{}" that is preceded by an unescaped
9899 "%", the pattern is interpreted like a "glob()" pattern,
9900 otherwise it is interpreted like a sequence pattern.
9901 All glob special characters "%*?[]{}" must be prefixed with
9903 "%". To escape a literal "%" you shall use "%%".
9904 For example the pattern "foo-%*.jpeg" will match all the
9906 filenames prefixed by "foo-" and terminating with ".jpeg", and
9907 "foo-%?%?%?.jpeg" will match all the filenames prefixed with
9908 "foo-", followed by a sequence of three characters, and
9909 terminating with ".jpeg".
9910 This pattern type is deprecated in favor of glob and sequence.
9912 Default value is glob_sequence.
9914 pixel_format
9916 Set the pixel format of the images to read. If not specified the
9917 pixel format is guessed from the first image file in the sequence.
9918 start_number
9920 Set the index of the file matched by the image file pattern to
9921 start to read from. Default value is 0.
9922 start_number_range
9924 Set the index interval range to check when looking for the first
9925 image file in the sequence, starting from start_number. Default
9926 value is 5.
9927 ts_from_file
9929 If set to 1, will set frame timestamp to modification time of image
9930 file. Note that monotonity of timestamps is not provided: images go
9931 in the same order as without this option. Default value is 0. If
9932 set to 2, will set frame timestamp to the modification time of the
9933 image file in nanosecond precision.
9934 video_size
9936 Set the video size of the images to read. If not specified the
9937 video size is guessed from the first image file in the sequence.
9938 export_path_metadata
9940 If set to 1, will add two extra fields to the metadata found in
9941 input, making them also available for other filters (see drawtext
9942 filter for examples). Default value is 0. The extra fields are
9943 described below:
9944 lavf.image2dec.source_path
9946 Corresponds to the full path to the input file being read.
9947 lavf.image2dec.source_basename
9949 Corresponds to the name of the file being read.
9950 Examples
9952 • Use ffmpeg for creating a video from the images in the file
9954 sequence img-001.jpeg, img-002.jpeg, ..., assuming an input frame
9955 rate of 10 frames per second:
9956 ffmpeg -framerate 10 -i 'img-%03d.jpeg' out.mkv
9958 • As above, but start by reading from a file with index 100 in the
9960 sequence:
9961 ffmpeg -framerate 10 -start_number 100 -i 'img-%03d.jpeg' out.mkv
9963 • Read images matching the "*.png" glob pattern , that is all the
9965 files terminating with the ".png" suffix:
9966 ffmpeg -framerate 10 -pattern_type glob -i "*.png" out.mkv
9968 libgme
9970 The Game Music Emu library is a collection of video game music file
9971 emulators.
9972 See <https://bitbucket.org/mpyne/game-music-emu/overview> for more
9974 information.
9975 It accepts the following options:
9977 track_index
9979 Set the index of which track to demux. The demuxer can only export
9980 one track. Track indexes start at 0. Default is to pick the first
9981 track. Number of tracks is exported as tracks metadata entry.
9982 sample_rate
9984 Set the sampling rate of the exported track. Range is 1000 to
9985 999999. Default is 44100.
9986 max_size (bytes)
9988 The demuxer buffers the entire file into memory. Adjust this value
9989 to set the maximum buffer size, which in turn, acts as a ceiling
9990 for the size of files that can be read. Default is 50 MiB.
9991 libmodplug
9993 ModPlug based module demuxer
9994 See <https://github.com/Konstanty/libmodplug>
9996 It will export one 2-channel 16-bit 44.1 kHz audio stream. Optionally,
9998 a "pal8" 16-color video stream can be exported with or without printed
9999 metadata.
10000 It accepts the following options:
10002 noise_reduction
10004 Apply a simple low-pass filter. Can be 1 (on) or 0 (off). Default
10005 is 0.
10006 reverb_depth
10008 Set amount of reverb. Range 0-100. Default is 0.
10009 reverb_delay
10011 Set delay in ms, clamped to 40-250 ms. Default is 0.
10012 bass_amount
10014 Apply bass expansion a.k.a. XBass or megabass. Range is 0 (quiet)
10015 to 100 (loud). Default is 0.
10016 bass_range
10018 Set cutoff i.e. upper-bound for bass frequencies. Range is 10-100
10019 Hz. Default is 0.
10020 surround_depth
10022 Apply a Dolby Pro-Logic surround effect. Range is 0 (quiet) to 100
10023 (heavy). Default is 0.
10024 surround_delay
10026 Set surround delay in ms, clamped to 5-40 ms. Default is 0.
10027 max_size
10029 The demuxer buffers the entire file into memory. Adjust this value
10030 to set the maximum buffer size, which in turn, acts as a ceiling
10031 for the size of files that can be read. Range is 0 to 100 MiB. 0
10032 removes buffer size limit (not recommended). Default is 5 MiB.
10033 video_stream_expr
10035 String which is evaluated using the eval API to assign colors to
10036 the generated video stream. Variables which can be used are "x",
10037 "y", "w", "h", "t", "speed", "tempo", "order", "pattern" and "row".
10038 video_stream
10040 Generate video stream. Can be 1 (on) or 0 (off). Default is 0.
10041 video_stream_w
10043 Set video frame width in 'chars' where one char indicates 8 pixels.
10044 Range is 20-512. Default is 30.
10045 video_stream_h
10047 Set video frame height in 'chars' where one char indicates 8
10048 pixels. Range is 20-512. Default is 30.
10049 video_stream_ptxt
10051 Print metadata on video stream. Includes "speed", "tempo", "order",
10052 "pattern", "row" and "ts" (time in ms). Can be 1 (on) or 0 (off).
10053 Default is 1.
10054 libopenmpt
10056 libopenmpt based module demuxer
10057 See <https://lib.openmpt.org/libopenmpt/> for more information.
10059 Some files have multiple subsongs (tracks) this can be set with the
10061 subsong option.
10062 It accepts the following options:
10064 subsong
10066 Set the subsong index. This can be either 'all', 'auto', or the
10067 index of the subsong. Subsong indexes start at 0. The default is
10068 'auto'.
10069 The default value is to let libopenmpt choose.
10071 layout
10073 Set the channel layout. Valid values are 1, 2, and 4 channel
10074 layouts. The default value is STEREO.
10075 sample_rate
10077 Set the sample rate for libopenmpt to output. Range is from 1000
10078 to INT_MAX. The value default is 48000.
10079 mov/mp4/3gp
10081 Demuxer for Quicktime File Format & ISO/IEC Base Media File Format
10082 (ISO/IEC 14496-12 or MPEG-4 Part 12, ISO/IEC 15444-12 or JPEG 2000 Part
10083 12).
10084 Registered extensions: mov, mp4, m4a, 3gp, 3g2, mj2, psp, m4b, ism,
10086 ismv, isma, f4v
10087 Options
10089 This demuxer accepts the following options:
10091 enable_drefs
10093 Enable loading of external tracks, disabled by default. Enabling
10094 this can theoretically leak information in some use cases.
10095 use_absolute_path
10097 Allows loading of external tracks via absolute paths, disabled by
10098 default. Enabling this poses a security risk. It should only be
10099 enabled if the source is known to be non-malicious.
10100 seek_streams_individually
10102 When seeking, identify the closest point in each stream
10103 individually and demux packets in that stream from identified
10104 point. This can lead to a different sequence of packets compared to
10105 demuxing linearly from the beginning. Default is true.
10106 ignore_editlist
10108 Ignore any edit list atoms. The demuxer, by default, modifies the
10109 stream index to reflect the timeline described by the edit list.
10110 Default is false.
10111 advanced_editlist
10113 Modify the stream index to reflect the timeline described by the
10114 edit list. "ignore_editlist" must be set to false for this option
10115 to be effective. If both "ignore_editlist" and this option are set
10116 to false, then only the start of the stream index is modified to
10117 reflect initial dwell time or starting timestamp described by the
10118 edit list. Default is true.
10119 ignore_chapters
10121 Don't parse chapters. This includes GoPro 'HiLight' tags/moments.
10122 Note that chapters are only parsed when input is seekable. Default
10123 is false.
10124 use_mfra_for
10126 For seekable fragmented input, set fragment's starting timestamp
10127 from media fragment random access box, if present.
10128 Following options are available:
10130 auto
10132 Auto-detect whether to set mfra timestamps as PTS or DTS
10133 (default)
10134 dts Set mfra timestamps as DTS
10136 pts Set mfra timestamps as PTS
10138 0 Don't use mfra box to set timestamps
10140 use_tfdt
10142 For fragmented input, set fragment's starting timestamp to
10143 "baseMediaDecodeTime" from the "tfdt" box. Default is enabled,
10144 which will prefer to use the "tfdt" box to set DTS. Disable to use
10145 the "earliest_presentation_time" from the "sidx" box. In either
10146 case, the timestamp from the "mfra" box will be used if it's
10147 available and "use_mfra_for" is set to pts or dts.
10148 export_all
10150 Export unrecognized boxes within the udta box as metadata entries.
10151 The first four characters of the box type are set as the key.
10152 Default is false.
10153 export_xmp
10155 Export entire contents of XMP_ box and uuid box as a string with
10156 key "xmp". Note that if "export_all" is set and this option isn't,
10157 the contents of XMP_ box are still exported but with key "XMP_".
10158 Default is false.
10159 activation_bytes
10161 4-byte key required to decrypt Audible AAX and AAX+ files. See
10162 Audible AAX subsection below.
10163 audible_fixed_key
10165 Fixed key used for handling Audible AAX/AAX+ files. It has been
10166 pre-set so should not be necessary to specify.
10167 decryption_key
10169 16-byte key, in hex, to decrypt files encrypted using ISO Common
10170 Encryption (CENC/AES-128 CTR; ISO/IEC 23001-7).
10171 max_stts_delta
10173 Very high sample deltas written in a trak's stts box may
10174 occasionally be intended but usually they are written in error or
10175 used to store a negative value for dts correction when treated as
10176 signed 32-bit integers. This option lets the user set an upper
10177 limit, beyond which the delta is clamped to 1. Values greater than
10178 the limit if negative when cast to int32 are used to adjust onward
10179 dts.
10180 Unit is the track time scale. Range is 0 to UINT_MAX. Default is
10182 "UINT_MAX - 48000*10" which allows upto a 10 second dts correction
10183 for 48 kHz audio streams while accommodating 99.9% of "uint32"
10184 range.
10185 Audible AAX
10187 Audible AAX files are encrypted M4B files, and they can be decrypted by
10189 specifying a 4 byte activation secret.
10190 ffmpeg -activation_bytes 1CEB00DA -i test.aax -vn -c:a copy output.mp4
10192 mpegts
10194 MPEG-2 transport stream demuxer.
10195 This demuxer accepts the following options:
10197 resync_size
10199 Set size limit for looking up a new synchronization. Default value
10200 is 65536.
10201 skip_unknown_pmt
10203 Skip PMTs for programs not defined in the PAT. Default value is 0.
10204 fix_teletext_pts
10206 Override teletext packet PTS and DTS values with the timestamps
10207 calculated from the PCR of the first program which the teletext
10208 stream is part of and is not discarded. Default value is 1, set
10209 this option to 0 if you want your teletext packet PTS and DTS
10210 values untouched.
10211 ts_packetsize
10213 Output option carrying the raw packet size in bytes. Show the
10214 detected raw packet size, cannot be set by the user.
10215 scan_all_pmts
10217 Scan and combine all PMTs. The value is an integer with value from
10218 -1 to 1 (-1 means automatic setting, 1 means enabled, 0 means
10219 disabled). Default value is -1.
10220 merge_pmt_versions
10222 Re-use existing streams when a PMT's version is updated and
10223 elementary streams move to different PIDs. Default value is 0.
10224 mpjpeg
10226 MJPEG encapsulated in multi-part MIME demuxer.
10227 This demuxer allows reading of MJPEG, where each frame is represented
10229 as a part of multipart/x-mixed-replace stream.
10230 strict_mime_boundary
10232 Default implementation applies a relaxed standard to multi-part
10233 MIME boundary detection, to prevent regression with numerous
10234 existing endpoints not generating a proper MIME MJPEG stream.
10235 Turning this option on by setting it to 1 will result in a stricter
10236 check of the boundary value.
10237 rawvideo
10239 Raw video demuxer.
10240 This demuxer allows one to read raw video data. Since there is no
10242 header specifying the assumed video parameters, the user must specify
10243 them in order to be able to decode the data correctly.
10244 This demuxer accepts the following options:
10246 framerate
10248 Set input video frame rate. Default value is 25.
10249 pixel_format
10251 Set the input video pixel format. Default value is "yuv420p".
10252 video_size
10254 Set the input video size. This value must be specified explicitly.
10255 For example to read a rawvideo file input.raw with ffplay, assuming a
10257 pixel format of "rgb24", a video size of "320x240", and a frame rate of
10258 10 images per second, use the command:
10259 ffplay -f rawvideo -pixel_format rgb24 -video_size 320x240 -framerate 10 input.raw
10261 sbg
10263 SBaGen script demuxer.
10264 This demuxer reads the script language used by SBaGen
10266 <http://uazu.net/sbagen/> to generate binaural beats sessions. A SBG
10267 script looks like that:
10268 -SE
10270 a: 300-2.5/3 440+4.5/0
10271 b: 300-2.5/0 440+4.5/3
10272 off: -
10273 NOW == a
10274 +0:07:00 == b
10275 +0:14:00 == a
10276 +0:21:00 == b
10277 +0:30:00 off
10278 A SBG script can mix absolute and relative timestamps. If the script
10280 uses either only absolute timestamps (including the script start time)
10281 or only relative ones, then its layout is fixed, and the conversion is
10282 straightforward. On the other hand, if the script mixes both kind of
10283 timestamps, then the NOW reference for relative timestamps will be
10284 taken from the current time of day at the time the script is read, and
10285 the script layout will be frozen according to that reference. That
10286 means that if the script is directly played, the actual times will
10287 match the absolute timestamps up to the sound controller's clock
10288 accuracy, but if the user somehow pauses the playback or seeks, all
10289 times will be shifted accordingly.
10290 tedcaptions
10292 JSON captions used for <http://www.ted.com/>.
10293 TED does not provide links to the captions, but they can be guessed
10295 from the page. The file tools/bookmarklets.html from the FFmpeg source
10296 tree contains a bookmarklet to expose them.
10297 This demuxer accepts the following option:
10299 start_time
10301 Set the start time of the TED talk, in milliseconds. The default is
10302 15000 (15s). It is used to sync the captions with the downloadable
10303 videos, because they include a 15s intro.
10304 Example: convert the captions to a format most players understand:
10306 ffmpeg -i http://www.ted.com/talks/subtitles/id/1/lang/en talk1-en.srt
10308 vapoursynth
10310 Vapoursynth wrapper.
10311 Due to security concerns, Vapoursynth scripts will not be autodetected
10313 so the input format has to be forced. For ff* CLI tools, add "-f
10314 vapoursynth" before the input "-i yourscript.vpy".
10315 This demuxer accepts the following option:
10317 max_script_size
10319 The demuxer buffers the entire script into memory. Adjust this
10320 value to set the maximum buffer size, which in turn, acts as a
10321 ceiling for the size of scripts that can be read. Default is 1
10322 MiB.
10323
10325 Muxers are configured elements in FFmpeg which allow writing multimedia
10326 streams to a particular type of file.
10327 When you configure your FFmpeg build, all the supported muxers are
10329 enabled by default. You can list all available muxers using the
10330 configure option "--list-muxers".
10331 You can disable all the muxers with the configure option
10333 "--disable-muxers" and selectively enable / disable single muxers with
10334 the options "--enable-muxer=MUXER" / "--disable-muxer=MUXER".
10335 The option "-muxers" of the ff* tools will display the list of enabled
10337 muxers. Use "-formats" to view a combined list of enabled demuxers and
10338 muxers.
10339 A description of some of the currently available muxers follows.
10341 a64
10343 A64 muxer for Commodore 64 video. Accepts a single "a64_multi" or
10344 "a64_multi5" codec video stream.
10345 adts
10347 Audio Data Transport Stream muxer. It accepts a single AAC stream.
10348 Options
10350 It accepts the following options:
10352 write_id3v2 bool
10354 Enable to write ID3v2.4 tags at the start of the stream. Default is
10355 disabled.
10356 write_apetag bool
10358 Enable to write APE tags at the end of the stream. Default is
10359 disabled.
10360 write_mpeg2 bool
10362 Enable to set MPEG version bit in the ADTS frame header to 1 which
10363 indicates MPEG-2. Default is 0, which indicates MPEG-4.
10364 aiff
10366 Audio Interchange File Format muxer.
10367 Options
10369 It accepts the following options:
10371 write_id3v2
10373 Enable ID3v2 tags writing when set to 1. Default is 0 (disabled).
10374 id3v2_version
10376 Select ID3v2 version to write. Currently only version 3 and 4 (aka.
10377 ID3v2.3 and ID3v2.4) are supported. The default is version 4.
10378 alp
10380 Muxer for audio of High Voltage Software's Lego Racers game. It accepts
10381 a single ADPCM_IMA_ALP stream with no more than 2 channels nor a sample
10382 rate greater than 44100 Hz.
10383 Extensions: tun, pcm
10385 Options
10387 It accepts the following options:
10389 type type
10391 Set file type.
10392 tun Set file type as music. Must have a sample rate of 22050 Hz.
10394 pcm Set file type as sfx.
10396 auto
10398 Set file type as per output file extension. ".pcm" results in
10399 type "pcm" else type "tun" is set. (default)
10400 asf
10402 Advanced Systems Format muxer.
10403 Note that Windows Media Audio (wma) and Windows Media Video (wmv) use
10405 this muxer too.
10406 Options
10408 It accepts the following options:
10410 packet_size
10412 Set the muxer packet size. By tuning this setting you may reduce
10413 data fragmentation or muxer overhead depending on your source.
10414 Default value is 3200, minimum is 100, maximum is 64k.
10415 avi
10417 Audio Video Interleaved muxer.
10418 Options
10420 It accepts the following options:
10422 reserve_index_space
10424 Reserve the specified amount of bytes for the OpenDML master index
10425 of each stream within the file header. By default additional master
10426 indexes are embedded within the data packets if there is no space
10427 left in the first master index and are linked together as a chain
10428 of indexes. This index structure can cause problems for some use
10429 cases, e.g. third-party software strictly relying on the OpenDML
10430 index specification or when file seeking is slow. Reserving enough
10431 index space in the file header avoids these problems.
10432 The required index space depends on the output file size and should
10434 be about 16 bytes per gigabyte. When this option is omitted or set
10435 to zero the necessary index space is guessed.
10436 write_channel_mask
10438 Write the channel layout mask into the audio stream header.
10439 This option is enabled by default. Disabling the channel mask can
10441 be useful in specific scenarios, e.g. when merging multiple audio
10442 streams into one for compatibility with software that only supports
10443 a single audio stream in AVI (see the "amerge" section in the
10444 ffmpeg-filters manual).
10445 flipped_raw_rgb
10447 If set to true, store positive height for raw RGB bitmaps, which
10448 indicates bitmap is stored bottom-up. Note that this option does
10449 not flip the bitmap which has to be done manually beforehand, e.g.
10450 by using the vflip filter. Default is false and indicates bitmap
10451 is stored top down.
10452 chromaprint
10454 Chromaprint fingerprinter.
10455 This muxer feeds audio data to the Chromaprint library, which generates
10457 a fingerprint for the provided audio data. See
10458 <https://acoustid.org/chromaprint>
10459 It takes a single signed native-endian 16-bit raw audio stream of at
10461 most 2 channels.
10462 Options
10464 silence_threshold
10466 Threshold for detecting silence. Range is from -1 to 32767, where
10467 -1 disables silence detection. Silence detection can only be used
10468 with version 3 of the algorithm. Silence detection must be
10469 disabled for use with the AcoustID service. Default is -1.
10470 algorithm
10472 Version of algorithm to fingerprint with. Range is 0 to 4. Version
10473 3 enables silence detection. Default is 1.
10474 fp_format
10476 Format to output the fingerprint as. Accepts the following options:
10477 raw Binary raw fingerprint
10479 compressed
10481 Binary compressed fingerprint
10482 base64
10484 Base64 compressed fingerprint (default)
10485 crc
10487 CRC (Cyclic Redundancy Check) testing format.
10488 This muxer computes and prints the Adler-32 CRC of all the input audio
10490 and video frames. By default audio frames are converted to signed
10491 16-bit raw audio and video frames to raw video before computing the
10492 CRC.
10493 The output of the muxer consists of a single line of the form:
10495 CRC=0xCRC, where CRC is a hexadecimal number 0-padded to 8 digits
10496 containing the CRC for all the decoded input frames.
10497 See also the framecrc muxer.
10499 Examples
10501 For example to compute the CRC of the input, and store it in the file
10503 out.crc:
10504 ffmpeg -i INPUT -f crc out.crc
10506 You can print the CRC to stdout with the command:
10508 ffmpeg -i INPUT -f crc -
10510 You can select the output format of each frame with ffmpeg by
10512 specifying the audio and video codec and format. For example to compute
10513 the CRC of the input audio converted to PCM unsigned 8-bit and the
10514 input video converted to MPEG-2 video, use the command:
10515 ffmpeg -i INPUT -c:a pcm_u8 -c:v mpeg2video -f crc -
10517 dash
10519 Dynamic Adaptive Streaming over HTTP (DASH) muxer that creates segments
10520 and manifest files according to the MPEG-DASH standard ISO/IEC
10521 23009-1:2014.
10522 For more information see:
10524 • ISO DASH Specification:
10526 <http://standards.iso.org/ittf/PubliclyAvailableStandards/c065274_ISO_IEC_23009-1_2014.zip>
10527 • WebM DASH Specification:
10529 <https://sites.google.com/a/webmproject.org/wiki/adaptive-streaming/webm-dash-specification>
10530 It creates a MPD manifest file and segment files for each stream.
10532 The segment filename might contain pre-defined identifiers used with
10534 SegmentTemplate as defined in section 5.3.9.4.4 of the standard.
10535 Available identifiers are "$RepresentationID$", "$Number$",
10536 "$Bandwidth$" and "$Time$". In addition to the standard identifiers,
10537 an ffmpeg-specific "$ext$" identifier is also supported. When
10538 specified ffmpeg will replace $ext$ in the file name with muxing
10539 format's extensions such as mp4, webm etc.,
10540 ffmpeg -re -i <input> -map 0 -map 0 -c:a libfdk_aac -c:v libx264 \
10542 -b:v:0 800k -b:v:1 300k -s:v:1 320x170 -profile:v:1 baseline \
10543 -profile:v:0 main -bf 1 -keyint_min 120 -g 120 -sc_threshold 0 \
10544 -b_strategy 0 -ar:a:1 22050 -use_timeline 1 -use_template 1 \
10545 -window_size 5 -adaptation_sets "id=0,streams=v id=1,streams=a" \
10546 -f dash /path/to/out.mpd
10547 seg_duration duration
10549 Set the segment length in seconds (fractional value can be set).
10550 The value is treated as average segment duration when use_template
10551 is enabled and use_timeline is disabled and as minimum segment
10552 duration for all the other use cases.
10553 frag_duration duration
10555 Set the length in seconds of fragments within segments (fractional
10556 value can be set).
10557 frag_type type
10559 Set the type of interval for fragmentation.
10560 window_size size
10562 Set the maximum number of segments kept in the manifest.
10563 extra_window_size size
10565 Set the maximum number of segments kept outside of the manifest
10566 before removing from disk.
10567 remove_at_exit remove
10569 Enable (1) or disable (0) removal of all segments when finished.
10570 use_template template
10572 Enable (1) or disable (0) use of SegmentTemplate instead of
10573 SegmentList.
10574 use_timeline timeline
10576 Enable (1) or disable (0) use of SegmentTimeline in
10577 SegmentTemplate.
10578 single_file single_file
10580 Enable (1) or disable (0) storing all segments in one file,
10581 accessed using byte ranges.
10582 single_file_name file_name
10584 DASH-templated name to be used for baseURL. Implies single_file set
10585 to "1". In the template, "$ext$" is replaced with the file name
10586 extension specific for the segment format.
10587 init_seg_name init_name
10589 DASH-templated name to used for the initialization segment. Default
10590 is "init-stream$RepresentationID$.$ext$". "$ext$" is replaced with
10591 the file name extension specific for the segment format.
10592 media_seg_name segment_name
10594 DASH-templated name to used for the media segments. Default is
10595 "chunk-stream$RepresentationID$-$Number%05d$.$ext$". "$ext$" is
10596 replaced with the file name extension specific for the segment
10597 format.
10598 utc_timing_url utc_url
10600 URL of the page that will return the UTC timestamp in ISO format.
10601 Example: "https://time.akamai.com/?iso"
10602 method method
10604 Use the given HTTP method to create output files. Generally set to
10605 PUT or POST.
10606 http_user_agent user_agent
10608 Override User-Agent field in HTTP header. Applicable only for HTTP
10609 output.
10610 http_persistent http_persistent
10612 Use persistent HTTP connections. Applicable only for HTTP output.
10613 hls_playlist hls_playlist
10615 Generate HLS playlist files as well. The master playlist is
10616 generated with the filename hls_master_name. One media playlist
10617 file is generated for each stream with filenames media_0.m3u8,
10618 media_1.m3u8, etc.
10619 hls_master_name file_name
10621 HLS master playlist name. Default is "master.m3u8".
10622 streaming streaming
10624 Enable (1) or disable (0) chunk streaming mode of output. In chunk
10625 streaming mode, each frame will be a moof fragment which forms a
10626 chunk.
10627 adaptation_sets adaptation_sets
10629 Assign streams to AdaptationSets. Syntax is "id=x,streams=a,b,c
10630 id=y,streams=d,e" with x and y being the IDs of the adaptation sets
10631 and a,b,c,d and e are the indices of the mapped streams.
10632 To map all video (or audio) streams to an AdaptationSet, "v" (or
10634 "a") can be used as stream identifier instead of IDs.
10635 When no assignment is defined, this defaults to an AdaptationSet
10637 for each stream.
10638 Optional syntax is
10640 "id=x,seg_duration=x,frag_duration=x,frag_type=type,descriptor=descriptor_string,streams=a,b,c
10641 id=y,seg_duration=y,frag_type=type,streams=d,e" and so on,
10642 descriptor is useful to the scheme defined by ISO/IEC
10643 23009-1:2014/Amd.2:2015. For example, -adaptation_sets
10644 "id=0,descriptor=<SupplementalProperty
10645 schemeIdUri=\"urn:mpeg:dash:srd:2014\"
10646 value=\"0,0,0,1,1,2,2\"/>,streams=v". Please note that descriptor
10647 string should be a self-closing xml tag. seg_duration,
10648 frag_duration and frag_type override the global option values for
10649 each adaptation set. For example, -adaptation_sets
10650 "id=0,seg_duration=2,frag_duration=1,frag_type=duration,streams=v
10651 id=1,seg_duration=2,frag_type=none,streams=a" type_id marks an
10652 adaptation set as containing streams meant to be used for Trick
10653 Mode for the referenced adaptation set. For example,
10654 -adaptation_sets "id=0,seg_duration=2,frag_type=none,streams=0
10655 id=1,seg_duration=10,frag_type=none,trick_id=0,streams=1"
10656 timeout timeout
10658 Set timeout for socket I/O operations. Applicable only for HTTP
10659 output.
10660 index_correction index_correction
10662 Enable (1) or Disable (0) segment index correction logic.
10663 Applicable only when use_template is enabled and use_timeline is
10664 disabled.
10665 When enabled, the logic monitors the flow of segment indexes. If a
10667 streams's segment index value is not at the expected real time
10668 position, then the logic corrects that index value.
10669 Typically this logic is needed in live streaming use cases. The
10671 network bandwidth fluctuations are common during long run
10672 streaming. Each fluctuation can cause the segment indexes fall
10673 behind the expected real time position.
10674 format_options options_list
10676 Set container format (mp4/webm) options using a ":" separated list
10677 of key=value parameters. Values containing ":" special characters
10678 must be escaped.
10679 global_sidx global_sidx
10681 Write global SIDX atom. Applicable only for single file, mp4
10682 output, non-streaming mode.
10683 dash_segment_type dash_segment_type
10685 Possible values:
10686 auto
10688 If this flag is set, the dash segment files format will be
10689 selected based on the stream codec. This is the default mode.
10690 mp4 If this flag is set, the dash segment files will be in in
10692 ISOBMFF format.
10693 webm
10695 If this flag is set, the dash segment files will be in in WebM
10696 format.
10697 ignore_io_errors ignore_io_errors
10699 Ignore IO errors during open and write. Useful for long-duration
10700 runs with network output.
10701 lhls lhls
10703 Enable Low-latency HLS(LHLS). Adds #EXT-X-PREFETCH tag with current
10704 segment's URI. hls.js player folks are trying to standardize an
10705 open LHLS spec. The draft spec is available in
10706 https://github.com/video-dev/hlsjs-rfcs/blob/lhls-spec/proposals/0001-lhls.md
10707 This option tries to comply with the above open spec. It enables
10708 streaming and hls_playlist options automatically. This is an
10709 experimental feature.
10710 Note: This is not Apple's version LHLS. See
10712 <https://datatracker.ietf.org/doc/html/draft-pantos-hls-rfc8216bis>
10713 ldash ldash
10715 Enable Low-latency Dash by constraining the presence and values of
10716 some elements.
10717 master_m3u8_publish_rate master_m3u8_publish_rate
10719 Publish master playlist repeatedly every after specified number of
10720 segment intervals.
10721 write_prft write_prft
10723 Write Producer Reference Time elements on supported streams. This
10724 also enables writing prft boxes in the underlying muxer. Applicable
10725 only when the utc_url option is enabled. It's set to auto by
10726 default, in which case the muxer will attempt to enable it only in
10727 modes that require it.
10728 mpd_profile mpd_profile
10730 Set one or more manifest profiles.
10731 http_opts http_opts
10733 A :-separated list of key=value options to pass to the underlying
10734 HTTP protocol. Applicable only for HTTP output.
10735 target_latency target_latency
10737 Set an intended target latency in seconds (fractional value can be
10738 set) for serving. Applicable only when streaming and write_prft
10739 options are enabled. This is an informative fields clients can use
10740 to measure the latency of the service.
10741 min_playback_rate min_playback_rate
10743 Set the minimum playback rate indicated as appropriate for the
10744 purposes of automatically adjusting playback latency and buffer
10745 occupancy during normal playback by clients.
10746 max_playback_rate max_playback_rate
10748 Set the maximum playback rate indicated as appropriate for the
10749 purposes of automatically adjusting playback latency and buffer
10750 occupancy during normal playback by clients.
10751 update_period update_period
10753 Set the mpd update period ,for dynamic content.
10754 The unit is second.
10755 fifo
10757 The fifo pseudo-muxer allows the separation of encoding and muxing by
10758 using first-in-first-out queue and running the actual muxer in a
10759 separate thread. This is especially useful in combination with the tee
10760 muxer and can be used to send data to several destinations with
10761 different reliability/writing speed/latency.
10762 API users should be aware that callback functions (interrupt_callback,
10764 io_open and io_close) used within its AVFormatContext must be thread-
10765 safe.
10766 The behavior of the fifo muxer if the queue fills up or if the output
10768 fails is selectable,
10769 • output can be transparently restarted with configurable delay
10771 between retries based on real time or time of the processed stream.
10772 • encoding can be blocked during temporary failure, or continue
10774 transparently dropping packets in case fifo queue fills up.
10775 fifo_format
10777 Specify the format name. Useful if it cannot be guessed from the
10778 output name suffix.
10779 queue_size
10781 Specify size of the queue (number of packets). Default value is 60.
10782 format_opts
10784 Specify format options for the underlying muxer. Muxer options can
10785 be specified as a list of key=value pairs separated by ':'.
10786 drop_pkts_on_overflow bool
10788 If set to 1 (true), in case the fifo queue fills up, packets will
10789 be dropped rather than blocking the encoder. This makes it possible
10790 to continue streaming without delaying the input, at the cost of
10791 omitting part of the stream. By default this option is set to 0
10792 (false), so in such cases the encoder will be blocked until the
10793 muxer processes some of the packets and none of them is lost.
10794 attempt_recovery bool
10796 If failure occurs, attempt to recover the output. This is
10797 especially useful when used with network output, since it makes it
10798 possible to restart streaming transparently. By default this
10799 option is set to 0 (false).
10800 max_recovery_attempts
10802 Sets maximum number of successive unsuccessful recovery attempts
10803 after which the output fails permanently. By default this option is
10804 set to 0 (unlimited).
10805 recovery_wait_time duration
10807 Waiting time before the next recovery attempt after previous
10808 unsuccessful recovery attempt. Default value is 5 seconds.
10809 recovery_wait_streamtime bool
10811 If set to 0 (false), the real time is used when waiting for the
10812 recovery attempt (i.e. the recovery will be attempted after at
10813 least recovery_wait_time seconds). If set to 1 (true), the time of
10814 the processed stream is taken into account instead (i.e. the
10815 recovery will be attempted after at least recovery_wait_time
10816 seconds of the stream is omitted). By default, this option is set
10817 to 0 (false).
10818 recover_any_error bool
10820 If set to 1 (true), recovery will be attempted regardless of type
10821 of the error causing the failure. By default this option is set to
10822 0 (false) and in case of certain (usually permanent) errors the
10823 recovery is not attempted even when attempt_recovery is set to 1.
10824 restart_with_keyframe bool
10826 Specify whether to wait for the keyframe after recovering from
10827 queue overflow or failure. This option is set to 0 (false) by
10828 default.
10829 timeshift duration
10831 Buffer the specified amount of packets and delay writing the
10832 output. Note that queue_size must be big enough to store the
10833 packets for timeshift. At the end of the input the fifo buffer is
10834 flushed at realtime speed.
10835 Examples
10837 • Stream something to rtmp server, continue processing the stream at
10839 real-time rate even in case of temporary failure (network outage)
10840 and attempt to recover streaming every second indefinitely.
10841 ffmpeg -re -i ... -c:v libx264 -c:a aac -f fifo -fifo_format flv -map 0:v -map 0:a
10843 -drop_pkts_on_overflow 1 -attempt_recovery 1 -recovery_wait_time 1 rtmp://example.com/live/stream_name
10844 flv
10846 Adobe Flash Video Format muxer.
10847 This muxer accepts the following options:
10849 flvflags flags
10851 Possible values:
10852 aac_seq_header_detect
10854 Place AAC sequence header based on audio stream data.
10855 no_sequence_end
10857 Disable sequence end tag.
10858 no_metadata
10860 Disable metadata tag.
10861 no_duration_filesize
10863 Disable duration and filesize in metadata when they are equal
10864 to zero at the end of stream. (Be used to non-seekable living
10865 stream).
10866 add_keyframe_index
10868 Used to facilitate seeking; particularly for HTTP pseudo
10869 streaming.
10870 framecrc
10872 Per-packet CRC (Cyclic Redundancy Check) testing format.
10873 This muxer computes and prints the Adler-32 CRC for each audio and
10875 video packet. By default audio frames are converted to signed 16-bit
10876 raw audio and video frames to raw video before computing the CRC.
10877 The output of the muxer consists of a line for each audio and video
10879 packet of the form:
10880 <stream_index>, <packet_dts>, <packet_pts>, <packet_duration>, <packet_size>, 0x<CRC>
10882 CRC is a hexadecimal number 0-padded to 8 digits containing the CRC of
10884 the packet.
10885 Examples
10887 For example to compute the CRC of the audio and video frames in INPUT,
10889 converted to raw audio and video packets, and store it in the file
10890 out.crc:
10891 ffmpeg -i INPUT -f framecrc out.crc
10893 To print the information to stdout, use the command:
10895 ffmpeg -i INPUT -f framecrc -
10897 With ffmpeg, you can select the output format to which the audio and
10899 video frames are encoded before computing the CRC for each packet by
10900 specifying the audio and video codec. For example, to compute the CRC
10901 of each decoded input audio frame converted to PCM unsigned 8-bit and
10902 of each decoded input video frame converted to MPEG-2 video, use the
10903 command:
10904 ffmpeg -i INPUT -c:a pcm_u8 -c:v mpeg2video -f framecrc -
10906 See also the crc muxer.
10908 framehash
10910 Per-packet hash testing format.
10911 This muxer computes and prints a cryptographic hash for each audio and
10913 video packet. This can be used for packet-by-packet equality checks
10914 without having to individually do a binary comparison on each.
10915 By default audio frames are converted to signed 16-bit raw audio and
10917 video frames to raw video before computing the hash, but the output of
10918 explicit conversions to other codecs can also be used. It uses the
10919 SHA-256 cryptographic hash function by default, but supports several
10920 other algorithms.
10921 The output of the muxer consists of a line for each audio and video
10923 packet of the form:
10924 <stream_index>, <packet_dts>, <packet_pts>, <packet_duration>, <packet_size>, <hash>
10926 hash is a hexadecimal number representing the computed hash for the
10928 packet.
10929 hash algorithm
10931 Use the cryptographic hash function specified by the string
10932 algorithm. Supported values include "MD5", "murmur3", "RIPEMD128",
10933 "RIPEMD160", "RIPEMD256", "RIPEMD320", "SHA160", "SHA224", "SHA256"
10934 (default), "SHA512/224", "SHA512/256", "SHA384", "SHA512", "CRC32"
10935 and "adler32".
10936 Examples
10938 To compute the SHA-256 hash of the audio and video frames in INPUT,
10940 converted to raw audio and video packets, and store it in the file
10941 out.sha256:
10942 ffmpeg -i INPUT -f framehash out.sha256
10944 To print the information to stdout, using the MD5 hash function, use
10946 the command:
10947 ffmpeg -i INPUT -f framehash -hash md5 -
10949 See also the hash muxer.
10951 framemd5
10953 Per-packet MD5 testing format.
10954 This is a variant of the framehash muxer. Unlike that muxer, it
10956 defaults to using the MD5 hash function.
10957 Examples
10959 To compute the MD5 hash of the audio and video frames in INPUT,
10961 converted to raw audio and video packets, and store it in the file
10962 out.md5:
10963 ffmpeg -i INPUT -f framemd5 out.md5
10965 To print the information to stdout, use the command:
10967 ffmpeg -i INPUT -f framemd5 -
10969 See also the framehash and md5 muxers.
10971 gif
10973 Animated GIF muxer.
10974 It accepts the following options:
10976 loop
10978 Set the number of times to loop the output. Use "-1" for no loop, 0
10979 for looping indefinitely (default).
10980 final_delay
10982 Force the delay (expressed in centiseconds) after the last frame.
10983 Each frame ends with a delay until the next frame. The default is
10984 "-1", which is a special value to tell the muxer to re-use the
10985 previous delay. In case of a loop, you might want to customize this
10986 value to mark a pause for instance.
10987 For example, to encode a gif looping 10 times, with a 5 seconds delay
10989 between the loops:
10990 ffmpeg -i INPUT -loop 10 -final_delay 500 out.gif
10992 Note 1: if you wish to extract the frames into separate GIF files, you
10994 need to force the image2 muxer:
10995 ffmpeg -i INPUT -c:v gif -f image2 "out%d.gif"
10997 Note 2: the GIF format has a very large time base: the delay between
10999 two frames can therefore not be smaller than one centi second.
11000 hash
11002 Hash testing format.
11003 This muxer computes and prints a cryptographic hash of all the input
11005 audio and video frames. This can be used for equality checks without
11006 having to do a complete binary comparison.
11007 By default audio frames are converted to signed 16-bit raw audio and
11009 video frames to raw video before computing the hash, but the output of
11010 explicit conversions to other codecs can also be used. Timestamps are
11011 ignored. It uses the SHA-256 cryptographic hash function by default,
11012 but supports several other algorithms.
11013 The output of the muxer consists of a single line of the form:
11015 algo=hash, where algo is a short string representing the hash function
11016 used, and hash is a hexadecimal number representing the computed hash.
11017 hash algorithm
11019 Use the cryptographic hash function specified by the string
11020 algorithm. Supported values include "MD5", "murmur3", "RIPEMD128",
11021 "RIPEMD160", "RIPEMD256", "RIPEMD320", "SHA160", "SHA224", "SHA256"
11022 (default), "SHA512/224", "SHA512/256", "SHA384", "SHA512", "CRC32"
11023 and "adler32".
11024 Examples
11026 To compute the SHA-256 hash of the input converted to raw audio and
11028 video, and store it in the file out.sha256:
11029 ffmpeg -i INPUT -f hash out.sha256
11031 To print an MD5 hash to stdout use the command:
11033 ffmpeg -i INPUT -f hash -hash md5 -
11035 See also the framehash muxer.
11037 hls
11039 Apple HTTP Live Streaming muxer that segments MPEG-TS according to the
11040 HTTP Live Streaming (HLS) specification.
11041 It creates a playlist file, and one or more segment files. The output
11043 filename specifies the playlist filename.
11044 By default, the muxer creates a file for each segment produced. These
11046 files have the same name as the playlist, followed by a sequential
11047 number and a .ts extension.
11048 Make sure to require a closed GOP when encoding and to set the GOP size
11050 to fit your segment time constraint.
11051 For example, to convert an input file with ffmpeg:
11053 ffmpeg -i in.mkv -c:v h264 -flags +cgop -g 30 -hls_time 1 out.m3u8
11055 This example will produce the playlist, out.m3u8, and segment files:
11057 out0.ts, out1.ts, out2.ts, etc.
11058 See also the segment muxer, which provides a more generic and flexible
11060 implementation of a segmenter, and can be used to perform HLS
11061 segmentation.
11062 Options
11064 This muxer supports the following options:
11066 hls_init_time duration
11068 Set the initial target segment length. Default value is 0.
11069 duration must be a time duration specification, see the Time
11071 duration section in the ffmpeg-utils(1) manual.
11072 Segment will be cut on the next key frame after this time has
11074 passed on the first m3u8 list. After the initial playlist is
11075 filled ffmpeg will cut segments at duration equal to "hls_time"
11076 hls_time duration
11078 Set the target segment length. Default value is 2.
11079 duration must be a time duration specification, see the Time
11081 duration section in the ffmpeg-utils(1) manual. Segment will be
11082 cut on the next key frame after this time has passed.
11083 hls_list_size size
11085 Set the maximum number of playlist entries. If set to 0 the list
11086 file will contain all the segments. Default value is 5.
11087 hls_delete_threshold size
11089 Set the number of unreferenced segments to keep on disk before
11090 "hls_flags delete_segments" deletes them. Increase this to allow
11091 continue clients to download segments which were recently
11092 referenced in the playlist. Default value is 1, meaning segments
11093 older than "hls_list_size+1" will be deleted.
11094 hls_ts_options options_list
11096 Set output format options using a :-separated list of key=value
11097 parameters. Values containing ":" special characters must be
11098 escaped. "hls_ts_options" is deprecated, use hls_segment_options
11099 instead of it..
11100 hls_start_number_source
11102 Start the playlist sequence number ("#EXT-X-MEDIA-SEQUENCE")
11103 according to the specified source. Unless "hls_flags single_file"
11104 is set, it also specifies source of starting sequence numbers of
11105 segment and subtitle filenames. In any case, if "hls_flags
11106 append_list" is set and read playlist sequence number is greater
11107 than the specified start sequence number, then that value will be
11108 used as start value.
11109 It accepts the following values:
11111 generic (default)
11113 Set the starting sequence numbers according to start_number
11114 option value.
11115 epoch
11117 The start number will be the seconds since epoch (1970-01-01
11118 00:00:00)
11119 epoch_us
11121 The start number will be the microseconds since epoch
11122 (1970-01-01 00:00:00)
11123 datetime
11125 The start number will be based on the current date/time as
11126 YYYYmmddHHMMSS. e.g. 20161231235759.
11127 start_number number
11129 Start the playlist sequence number ("#EXT-X-MEDIA-SEQUENCE") from
11130 the specified number when hls_start_number_source value is generic.
11131 (This is the default case.) Unless "hls_flags single_file" is set,
11132 it also specifies starting sequence numbers of segment and subtitle
11133 filenames. Default value is 0.
11134 hls_allow_cache allowcache
11136 Explicitly set whether the client MAY (1) or MUST NOT (0) cache
11137 media segments.
11138 hls_base_url baseurl
11140 Append baseurl to every entry in the playlist. Useful to generate
11141 playlists with absolute paths.
11142 Note that the playlist sequence number must be unique for each
11144 segment and it is not to be confused with the segment filename
11145 sequence number which can be cyclic, for example if the wrap option
11146 is specified.
11147 hls_segment_filename filename
11149 Set the segment filename. Unless "hls_flags single_file" is set,
11150 filename is used as a string format with the segment number:
11151 ffmpeg -i in.nut -hls_segment_filename 'file%03d.ts' out.m3u8
11153 This example will produce the playlist, out.m3u8, and segment
11155 files: file000.ts, file001.ts, file002.ts, etc.
11156 filename may contain full path or relative path specification, but
11158 only the file name part without any path info will be contained in
11159 the m3u8 segment list. Should a relative path be specified, the
11160 path of the created segment files will be relative to the current
11161 working directory. When strftime_mkdir is set, the whole expanded
11162 value of filename will be written into the m3u8 segment list.
11163 When "var_stream_map" is set with two or more variant streams, the
11165 filename pattern must contain the string "%v", this string
11166 specifies the position of variant stream index in the generated
11167 segment file names.
11168 ffmpeg -i in.ts -b:v:0 1000k -b:v:1 256k -b:a:0 64k -b:a:1 32k \
11170 -map 0:v -map 0:a -map 0:v -map 0:a -f hls -var_stream_map "v:0,a:0 v:1,a:1" \
11171 -hls_segment_filename 'file_%v_%03d.ts' out_%v.m3u8
11172 This example will produce the playlists segment file sets:
11174 file_0_000.ts, file_0_001.ts, file_0_002.ts, etc. and
11175 file_1_000.ts, file_1_001.ts, file_1_002.ts, etc.
11176 The string "%v" may be present in the filename or in the last
11178 directory name containing the file, but only in one of them.
11179 (Additionally, %v may appear multiple times in the last sub-
11180 directory or filename.) If the string %v is present in the
11181 directory name, then sub-directories are created after expanding
11182 the directory name pattern. This enables creation of segments
11183 corresponding to different variant streams in subdirectories.
11184 ffmpeg -i in.ts -b:v:0 1000k -b:v:1 256k -b:a:0 64k -b:a:1 32k \
11186 -map 0:v -map 0:a -map 0:v -map 0:a -f hls -var_stream_map "v:0,a:0 v:1,a:1" \
11187 -hls_segment_filename 'vs%v/file_%03d.ts' vs%v/out.m3u8
11188 This example will produce the playlists segment file sets:
11190 vs0/file_000.ts, vs0/file_001.ts, vs0/file_002.ts, etc. and
11191 vs1/file_000.ts, vs1/file_001.ts, vs1/file_002.ts, etc.
11192 strftime
11194 Use strftime() on filename to expand the segment filename with
11195 localtime. The segment number is also available in this mode, but
11196 to use it, you need to specify second_level_segment_index hls_flag
11197 and %%d will be the specifier.
11198 ffmpeg -i in.nut -strftime 1 -hls_segment_filename 'file-%Y%m%d-%s.ts' out.m3u8
11200 This example will produce the playlist, out.m3u8, and segment
11202 files: file-20160215-1455569023.ts, file-20160215-1455569024.ts,
11203 etc. Note: On some systems/environments, the %s specifier is not
11204 available. See
11205 "strftime()" documentation.
11206 ffmpeg -i in.nut -strftime 1 -hls_flags second_level_segment_index -hls_segment_filename 'file-%Y%m%d-%%04d.ts' out.m3u8
11208 This example will produce the playlist, out.m3u8, and segment
11210 files: file-20160215-0001.ts, file-20160215-0002.ts, etc.
11211 strftime_mkdir
11213 Used together with -strftime_mkdir, it will create all
11214 subdirectories which is expanded in filename.
11215 ffmpeg -i in.nut -strftime 1 -strftime_mkdir 1 -hls_segment_filename '%Y%m%d/file-%Y%m%d-%s.ts' out.m3u8
11217 This example will create a directory 201560215 (if it does not
11219 exist), and then produce the playlist, out.m3u8, and segment files:
11220 20160215/file-20160215-1455569023.ts,
11221 20160215/file-20160215-1455569024.ts, etc.
11222 ffmpeg -i in.nut -strftime 1 -strftime_mkdir 1 -hls_segment_filename '%Y/%m/%d/file-%Y%m%d-%s.ts' out.m3u8
11224 This example will create a directory hierarchy 2016/02/15 (if any
11226 of them do not exist), and then produce the playlist, out.m3u8, and
11227 segment files: 2016/02/15/file-20160215-1455569023.ts,
11228 2016/02/15/file-20160215-1455569024.ts, etc.
11229 hls_segment_options options_list
11231 Set output format options using a :-separated list of key=value
11232 parameters. Values containing ":" special characters must be
11233 escaped.
11234 hls_key_info_file key_info_file
11236 Use the information in key_info_file for segment encryption. The
11237 first line of key_info_file specifies the key URI written to the
11238 playlist. The key URL is used to access the encryption key during
11239 playback. The second line specifies the path to the key file used
11240 to obtain the key during the encryption process. The key file is
11241 read as a single packed array of 16 octets in binary format. The
11242 optional third line specifies the initialization vector (IV) as a
11243 hexadecimal string to be used instead of the segment sequence
11244 number (default) for encryption. Changes to key_info_file will
11245 result in segment encryption with the new key/IV and an entry in
11246 the playlist for the new key URI/IV if "hls_flags periodic_rekey"
11247 is enabled.
11248 Key info file format:
11250 <key URI>
11252 <key file path>
11253 <IV> (optional)
11254 Example key URIs:
11256 http://server/file.key
11258 /path/to/file.key
11259 file.key
11260 Example key file paths:
11262 file.key
11264 /path/to/file.key
11265 Example IV:
11267 0123456789ABCDEF0123456789ABCDEF
11269 Key info file example:
11271 http://server/file.key
11273 /path/to/file.key
11274 0123456789ABCDEF0123456789ABCDEF
11275 Example shell script:
11277 #!/bin/sh
11279 BASE_URL=${1:-'.'}
11280 openssl rand 16 > file.key
11281 echo $BASE_URL/file.key > file.keyinfo
11282 echo file.key >> file.keyinfo
11283 echo $(openssl rand -hex 16) >> file.keyinfo
11284 ffmpeg -f lavfi -re -i testsrc -c:v h264 -hls_flags delete_segments \
11285 -hls_key_info_file file.keyinfo out.m3u8
11286 -hls_enc enc
11288 Enable (1) or disable (0) the AES128 encryption. When enabled
11289 every segment generated is encrypted and the encryption key is
11290 saved as playlist name.key.
11291 -hls_enc_key key
11293 16-octet key to encrypt the segments, by default it is randomly
11294 generated.
11295 -hls_enc_key_url keyurl
11297 If set, keyurl is prepended instead of baseurl to the key filename
11298 in the playlist.
11299 -hls_enc_iv iv
11301 16-octet initialization vector for every segment instead of the
11302 autogenerated ones.
11303 hls_segment_type flags
11305 Possible values:
11306 mpegts
11308 Output segment files in MPEG-2 Transport Stream format. This is
11309 compatible with all HLS versions.
11310 fmp4
11312 Output segment files in fragmented MP4 format, similar to MPEG-
11313 DASH. fmp4 files may be used in HLS version 7 and above.
11314 hls_fmp4_init_filename filename
11316 Set filename to the fragment files header file, default filename is
11317 init.mp4.
11318 Use "-strftime 1" on filename to expand the segment filename with
11320 localtime.
11321 ffmpeg -i in.nut -hls_segment_type fmp4 -strftime 1 -hls_fmp4_init_filename "%s_init.mp4" out.m3u8
11323 This will produce init like this 1602678741_init.mp4
11325 hls_fmp4_init_resend
11327 Resend init file after m3u8 file refresh every time, default is 0.
11328 When "var_stream_map" is set with two or more variant streams, the
11330 filename pattern must contain the string "%v", this string
11331 specifies the position of variant stream index in the generated
11332 init file names. The string "%v" may be present in the filename or
11333 in the last directory name containing the file. If the string is
11334 present in the directory name, then sub-directories are created
11335 after expanding the directory name pattern. This enables creation
11336 of init files corresponding to different variant streams in
11337 subdirectories.
11338 hls_flags flags
11340 Possible values:
11341 single_file
11343 If this flag is set, the muxer will store all segments in a
11344 single MPEG-TS file, and will use byte ranges in the playlist.
11345 HLS playlists generated with this way will have the version
11346 number 4. For example:
11347 ffmpeg -i in.nut -hls_flags single_file out.m3u8
11349 Will produce the playlist, out.m3u8, and a single segment file,
11351 out.ts.
11352 delete_segments
11354 Segment files removed from the playlist are deleted after a
11355 period of time equal to the duration of the segment plus the
11356 duration of the playlist.
11357 append_list
11359 Append new segments into the end of old segment list, and
11360 remove the "#EXT-X-ENDLIST" from the old segment list.
11361 round_durations
11363 Round the duration info in the playlist file segment info to
11364 integer values, instead of using floating point.
11365 discont_start
11367 Add the "#EXT-X-DISCONTINUITY" tag to the playlist, before the
11368 first segment's information.
11369 omit_endlist
11371 Do not append the "EXT-X-ENDLIST" tag at the end of the
11372 playlist.
11373 periodic_rekey
11375 The file specified by "hls_key_info_file" will be checked
11376 periodically and detect updates to the encryption info. Be sure
11377 to replace this file atomically, including the file containing
11378 the AES encryption key.
11379 independent_segments
11381 Add the "#EXT-X-INDEPENDENT-SEGMENTS" to playlists that has
11382 video segments and when all the segments of that playlist are
11383 guaranteed to start with a Key frame.
11384 iframes_only
11386 Add the "#EXT-X-I-FRAMES-ONLY" to playlists that has video
11387 segments and can play only I-frames in the "#EXT-X-BYTERANGE"
11388 mode.
11389 split_by_time
11391 Allow segments to start on frames other than keyframes. This
11392 improves behavior on some players when the time between
11393 keyframes is inconsistent, but may make things worse on others,
11394 and can cause some oddities during seeking. This flag should be
11395 used with the "hls_time" option.
11396 program_date_time
11398 Generate "EXT-X-PROGRAM-DATE-TIME" tags.
11399 second_level_segment_index
11401 Makes it possible to use segment indexes as %%d in
11402 hls_segment_filename expression besides date/time values when
11403 strftime is on. To get fixed width numbers with trailing
11404 zeroes, %%0xd format is available where x is the required
11405 width.
11406 second_level_segment_size
11408 Makes it possible to use segment sizes (counted in bytes) as
11409 %%s in hls_segment_filename expression besides date/time values
11410 when strftime is on. To get fixed width numbers with trailing
11411 zeroes, %%0xs format is available where x is the required
11412 width.
11413 second_level_segment_duration
11415 Makes it possible to use segment duration (calculated in
11416 microseconds) as %%t in hls_segment_filename expression besides
11417 date/time values when strftime is on. To get fixed width
11418 numbers with trailing zeroes, %%0xt format is available where x
11419 is the required width.
11420 ffmpeg -i sample.mpeg \
11422 -f hls -hls_time 3 -hls_list_size 5 \
11423 -hls_flags second_level_segment_index+second_level_segment_size+second_level_segment_duration \
11424 -strftime 1 -strftime_mkdir 1 -hls_segment_filename "segment_%Y%m%d%H%M%S_%%04d_%%08s_%%013t.ts" stream.m3u8
11425 This will produce segments like this:
11427 segment_20170102194334_0003_00122200_0000003000000.ts,
11428 segment_20170102194334_0004_00120072_0000003000000.ts etc.
11429 temp_file
11431 Write segment data to filename.tmp and rename to filename only
11432 once the segment is complete. A webserver serving up segments
11433 can be configured to reject requests to *.tmp to prevent access
11434 to in-progress segments before they have been added to the m3u8
11435 playlist. This flag also affects how m3u8 playlist files are
11436 created. If this flag is set, all playlist files will written
11437 into temporary file and renamed after they are complete,
11438 similarly as segments are handled. But playlists with "file"
11439 protocol and with type ("hls_playlist_type") other than "vod"
11440 are always written into temporary file regardless of this flag.
11441 Master playlist files ("master_pl_name"), if any, with "file"
11442 protocol, are always written into temporary file regardless of
11443 this flag if "master_pl_publish_rate" value is other than zero.
11444 hls_playlist_type event
11446 Emit "#EXT-X-PLAYLIST-TYPE:EVENT" in the m3u8 header. Forces
11447 hls_list_size to 0; the playlist can only be appended to.
11448 hls_playlist_type vod
11450 Emit "#EXT-X-PLAYLIST-TYPE:VOD" in the m3u8 header. Forces
11451 hls_list_size to 0; the playlist must not change.
11452 method
11454 Use the given HTTP method to create the hls files.
11455 ffmpeg -re -i in.ts -f hls -method PUT http://example.com/live/out.m3u8
11457 This example will upload all the mpegts segment files to the HTTP
11459 server using the HTTP PUT method, and update the m3u8 files every
11460 "refresh" times using the same method. Note that the HTTP server
11461 must support the given method for uploading files.
11462 http_user_agent
11464 Override User-Agent field in HTTP header. Applicable only for HTTP
11465 output.
11466 var_stream_map
11468 Map string which specifies how to group the audio, video and
11469 subtitle streams into different variant streams. The variant stream
11470 groups are separated by space. Expected string format is like this
11471 "a:0,v:0 a:1,v:1 ....". Here a:, v:, s: are the keys to specify
11472 audio, video and subtitle streams respectively. Allowed values are
11473 0 to 9 (limited just based on practical usage).
11474 When there are two or more variant streams, the output filename
11476 pattern must contain the string "%v", this string specifies the
11477 position of variant stream index in the output media playlist
11478 filenames. The string "%v" may be present in the filename or in the
11479 last directory name containing the file. If the string is present
11480 in the directory name, then sub-directories are created after
11481 expanding the directory name pattern. This enables creation of
11482 variant streams in subdirectories.
11483 ffmpeg -re -i in.ts -b:v:0 1000k -b:v:1 256k -b:a:0 64k -b:a:1 32k \
11485 -map 0:v -map 0:a -map 0:v -map 0:a -f hls -var_stream_map "v:0,a:0 v:1,a:1" \
11486 http://example.com/live/out_%v.m3u8
11487 This example creates two hls variant streams. The first variant
11489 stream will contain video stream of bitrate 1000k and audio stream
11490 of bitrate 64k and the second variant stream will contain video
11491 stream of bitrate 256k and audio stream of bitrate 32k. Here, two
11492 media playlist with file names out_0.m3u8 and out_1.m3u8 will be
11493 created. If you want something meaningful text instead of indexes
11494 in result names, you may specify names for each or some of the
11495 variants as in the following example.
11496 ffmpeg -re -i in.ts -b:v:0 1000k -b:v:1 256k -b:a:0 64k -b:a:1 32k \
11498 -map 0:v -map 0:a -map 0:v -map 0:a -f hls -var_stream_map "v:0,a:0,name:my_hd v:1,a:1,name:my_sd" \
11499 http://example.com/live/out_%v.m3u8
11500 This example creates two hls variant streams as in the previous
11502 one. But here, the two media playlist with file names
11503 out_my_hd.m3u8 and out_my_sd.m3u8 will be created.
11504 ffmpeg -re -i in.ts -b:v:0 1000k -b:v:1 256k -b:a:0 64k \
11506 -map 0:v -map 0:a -map 0:v -f hls -var_stream_map "v:0 a:0 v:1" \
11507 http://example.com/live/out_%v.m3u8
11508 This example creates three hls variant streams. The first variant
11510 stream will be a video only stream with video bitrate 1000k, the
11511 second variant stream will be an audio only stream with bitrate 64k
11512 and the third variant stream will be a video only stream with
11513 bitrate 256k. Here, three media playlist with file names
11514 out_0.m3u8, out_1.m3u8 and out_2.m3u8 will be created.
11515 ffmpeg -re -i in.ts -b:v:0 1000k -b:v:1 256k -b:a:0 64k -b:a:1 32k \
11517 -map 0:v -map 0:a -map 0:v -map 0:a -f hls -var_stream_map "v:0,a:0 v:1,a:1" \
11518 http://example.com/live/vs_%v/out.m3u8
11519 This example creates the variant streams in subdirectories. Here,
11521 the first media playlist is created at
11522 http://example.com/live/vs_0/out.m3u8 and the second one at
11523 http://example.com/live/vs_1/out.m3u8.
11524 ffmpeg -re -i in.ts -b:a:0 32k -b:a:1 64k -b:v:0 1000k -b:v:1 3000k \
11526 -map 0:a -map 0:a -map 0:v -map 0:v -f hls \
11527 -var_stream_map "a:0,agroup:aud_low a:1,agroup:aud_high v:0,agroup:aud_low v:1,agroup:aud_high" \
11528 -master_pl_name master.m3u8 \
11529 http://example.com/live/out_%v.m3u8
11530 This example creates two audio only and two video only variant
11532 streams. In addition to the #EXT-X-STREAM-INF tag for each variant
11533 stream in the master playlist, #EXT-X-MEDIA tag is also added for
11534 the two audio only variant streams and they are mapped to the two
11535 video only variant streams with audio group names 'aud_low' and
11536 'aud_high'.
11537 By default, a single hls variant containing all the encoded streams
11539 is created.
11540 ffmpeg -re -i in.ts -b:a:0 32k -b:a:1 64k -b:v:0 1000k \
11542 -map 0:a -map 0:a -map 0:v -f hls \
11543 -var_stream_map "a:0,agroup:aud_low,default:yes a:1,agroup:aud_low v:0,agroup:aud_low" \
11544 -master_pl_name master.m3u8 \
11545 http://example.com/live/out_%v.m3u8
11546 This example creates two audio only and one video only variant
11548 streams. In addition to the #EXT-X-STREAM-INF tag for each variant
11549 stream in the master playlist, #EXT-X-MEDIA tag is also added for
11550 the two audio only variant streams and they are mapped to the one
11551 video only variant streams with audio group name 'aud_low', and the
11552 audio group have default stat is NO or YES.
11553 By default, a single hls variant containing all the encoded streams
11555 is created.
11556 ffmpeg -re -i in.ts -b:a:0 32k -b:a:1 64k -b:v:0 1000k \
11558 -map 0:a -map 0:a -map 0:v -f hls \
11559 -var_stream_map "a:0,agroup:aud_low,default:yes,language:ENG a:1,agroup:aud_low,language:CHN v:0,agroup:aud_low" \
11560 -master_pl_name master.m3u8 \
11561 http://example.com/live/out_%v.m3u8
11562 This example creates two audio only and one video only variant
11564 streams. In addition to the #EXT-X-STREAM-INF tag for each variant
11565 stream in the master playlist, #EXT-X-MEDIA tag is also added for
11566 the two audio only variant streams and they are mapped to the one
11567 video only variant streams with audio group name 'aud_low', and the
11568 audio group have default stat is NO or YES, and one audio have and
11569 language is named ENG, the other audio language is named CHN.
11570 By default, a single hls variant containing all the encoded streams
11572 is created.
11573 ffmpeg -y -i input_with_subtitle.mkv \
11575 -b:v:0 5250k -c:v h264 -pix_fmt yuv420p -profile:v main -level 4.1 \
11576 -b:a:0 256k \
11577 -c:s webvtt -c:a mp2 -ar 48000 -ac 2 -map 0:v -map 0:a:0 -map 0:s:0 \
11578 -f hls -var_stream_map "v:0,a:0,s:0,sgroup:subtitle" \
11579 -master_pl_name master.m3u8 -t 300 -hls_time 10 -hls_init_time 4 -hls_list_size \
11580 10 -master_pl_publish_rate 10 -hls_flags \
11581 delete_segments+discont_start+split_by_time ./tmp/video.m3u8
11582 This example adds "#EXT-X-MEDIA" tag with "TYPE=SUBTITLES" in the
11584 master playlist with webvtt subtitle group name 'subtitle'. Please
11585 make sure the input file has one text subtitle stream at least.
11586 cc_stream_map
11588 Map string which specifies different closed captions groups and
11589 their attributes. The closed captions stream groups are separated
11590 by space. Expected string format is like this "ccgroup:<group
11591 name>,instreamid:<INSTREAM-ID>,language:<language code> ....".
11592 'ccgroup' and 'instreamid' are mandatory attributes. 'language' is
11593 an optional attribute. The closed captions groups configured using
11594 this option are mapped to different variant streams by providing
11595 the same 'ccgroup' name in the "var_stream_map" string. If
11596 "var_stream_map" is not set, then the first available ccgroup in
11597 "cc_stream_map" is mapped to the output variant stream. The
11598 examples for these two use cases are given below.
11599 ffmpeg -re -i in.ts -b:v 1000k -b:a 64k -a53cc 1 -f hls \
11601 -cc_stream_map "ccgroup:cc,instreamid:CC1,language:en" \
11602 -master_pl_name master.m3u8 \
11603 http://example.com/live/out.m3u8
11604 This example adds "#EXT-X-MEDIA" tag with "TYPE=CLOSED-CAPTIONS" in
11606 the master playlist with group name 'cc', language 'en' (english)
11607 and INSTREAM-ID 'CC1'. Also, it adds "CLOSED-CAPTIONS" attribute
11608 with group name 'cc' for the output variant stream.
11609 ffmpeg -re -i in.ts -b:v:0 1000k -b:v:1 256k -b:a:0 64k -b:a:1 32k \
11611 -a53cc:0 1 -a53cc:1 1\
11612 -map 0:v -map 0:a -map 0:v -map 0:a -f hls \
11613 -cc_stream_map "ccgroup:cc,instreamid:CC1,language:en ccgroup:cc,instreamid:CC2,language:sp" \
11614 -var_stream_map "v:0,a:0,ccgroup:cc v:1,a:1,ccgroup:cc" \
11615 -master_pl_name master.m3u8 \
11616 http://example.com/live/out_%v.m3u8
11617 This example adds two "#EXT-X-MEDIA" tags with
11619 "TYPE=CLOSED-CAPTIONS" in the master playlist for the INSTREAM-IDs
11620 'CC1' and 'CC2'. Also, it adds "CLOSED-CAPTIONS" attribute with
11621 group name 'cc' for the two output variant streams.
11622 master_pl_name
11624 Create HLS master playlist with the given name.
11625 ffmpeg -re -i in.ts -f hls -master_pl_name master.m3u8 http://example.com/live/out.m3u8
11627 This example creates HLS master playlist with name master.m3u8 and
11629 it is published at http://example.com/live/
11630 master_pl_publish_rate
11632 Publish master play list repeatedly every after specified number of
11633 segment intervals.
11634 ffmpeg -re -i in.ts -f hls -master_pl_name master.m3u8 \
11636 -hls_time 2 -master_pl_publish_rate 30 http://example.com/live/out.m3u8
11637 This example creates HLS master playlist with name master.m3u8 and
11639 keep publishing it repeatedly every after 30 segments i.e. every
11640 after 60s.
11641 http_persistent
11643 Use persistent HTTP connections. Applicable only for HTTP output.
11644 timeout
11646 Set timeout for socket I/O operations. Applicable only for HTTP
11647 output.
11648 -ignore_io_errors
11650 Ignore IO errors during open, write and delete. Useful for long-
11651 duration runs with network output.
11652 headers
11654 Set custom HTTP headers, can override built in default headers.
11655 Applicable only for HTTP output.
11656 ico
11658 ICO file muxer.
11659 Microsoft's icon file format (ICO) has some strict limitations that
11661 should be noted:
11662 • Size cannot exceed 256 pixels in any dimension
11664 • Only BMP and PNG images can be stored
11666 • If a BMP image is used, it must be one of the following pixel
11668 formats:
11669 BMP Bit Depth FFmpeg Pixel Format
11671 1bit pal8
11672 4bit pal8
11673 8bit pal8
11674 16bit rgb555le
11675 24bit bgr24
11676 32bit bgra
11677 • If a BMP image is used, it must use the BITMAPINFOHEADER DIB header
11679 • If a PNG image is used, it must use the rgba pixel format
11681 image2
11683 Image file muxer.
11684 The image file muxer writes video frames to image files.
11686 The output filenames are specified by a pattern, which can be used to
11688 produce sequentially numbered series of files. The pattern may contain
11689 the string "%d" or "%0Nd", this string specifies the position of the
11690 characters representing a numbering in the filenames. If the form
11691 "%0Nd" is used, the string representing the number in each filename is
11692 0-padded to N digits. The literal character '%' can be specified in the
11693 pattern with the string "%%".
11694 If the pattern contains "%d" or "%0Nd", the first filename of the file
11696 list specified will contain the number 1, all the following numbers
11697 will be sequential.
11698 The pattern may contain a suffix which is used to automatically
11700 determine the format of the image files to write.
11701 For example the pattern "img-%03d.bmp" will specify a sequence of
11703 filenames of the form img-001.bmp, img-002.bmp, ..., img-010.bmp, etc.
11704 The pattern "img%%-%d.jpg" will specify a sequence of filenames of the
11705 form img%-1.jpg, img%-2.jpg, ..., img%-10.jpg, etc.
11706 The image muxer supports the .Y.U.V image file format. This format is
11708 special in that that each image frame consists of three files, for each
11709 of the YUV420P components. To read or write this image file format,
11710 specify the name of the '.Y' file. The muxer will automatically open
11711 the '.U' and '.V' files as required.
11712 Options
11714 frame_pts
11716 If set to 1, expand the filename with pts from pkt->pts. Default
11717 value is 0.
11718 start_number
11720 Start the sequence from the specified number. Default value is 1.
11721 update
11723 If set to 1, the filename will always be interpreted as just a
11724 filename, not a pattern, and the corresponding file will be
11725 continuously overwritten with new images. Default value is 0.
11726 strftime
11728 If set to 1, expand the filename with date and time information
11729 from "strftime()". Default value is 0.
11730 atomic_writing
11732 Write output to a temporary file, which is renamed to target
11733 filename once writing is completed. Default is disabled.
11734 protocol_opts options_list
11736 Set protocol options as a :-separated list of key=value parameters.
11737 Values containing the ":" special character must be escaped.
11738 Examples
11740 The following example shows how to use ffmpeg for creating a sequence
11742 of files img-001.jpeg, img-002.jpeg, ..., taking one image every second
11743 from the input video:
11744 ffmpeg -i in.avi -vsync cfr -r 1 -f image2 'img-%03d.jpeg'
11746 Note that with ffmpeg, if the format is not specified with the "-f"
11748 option and the output filename specifies an image file format, the
11749 image2 muxer is automatically selected, so the previous command can be
11750 written as:
11751 ffmpeg -i in.avi -vsync cfr -r 1 'img-%03d.jpeg'
11753 Note also that the pattern must not necessarily contain "%d" or "%0Nd",
11755 for example to create a single image file img.jpeg from the start of
11756 the input video you can employ the command:
11757 ffmpeg -i in.avi -f image2 -frames:v 1 img.jpeg
11759 The strftime option allows you to expand the filename with date and
11761 time information. Check the documentation of the "strftime()" function
11762 for the syntax.
11763 For example to generate image files from the "strftime()"
11765 "%Y-%m-%d_%H-%M-%S" pattern, the following ffmpeg command can be used:
11766 ffmpeg -f v4l2 -r 1 -i /dev/video0 -f image2 -strftime 1 "%Y-%m-%d_%H-%M-%S.jpg"
11768 You can set the file name with current frame's PTS:
11770 ffmpeg -f v4l2 -r 1 -i /dev/video0 -copyts -f image2 -frame_pts true %d.jpg"
11772 A more complex example is to publish contents of your desktop directly
11774 to a WebDAV server every second:
11775 ffmpeg -f x11grab -framerate 1 -i :0.0 -q:v 6 -update 1 -protocol_opts method=PUT http://example.com/desktop.jpg
11777 matroska
11779 Matroska container muxer.
11780 This muxer implements the matroska and webm container specs.
11782 Metadata
11784 The recognized metadata settings in this muxer are:
11786 title
11788 Set title name provided to a single track. This gets mapped to the
11789 FileDescription element for a stream written as attachment.
11790 language
11792 Specify the language of the track in the Matroska languages form.
11793 The language can be either the 3 letters bibliographic ISO-639-2
11795 (ISO 639-2/B) form (like "fre" for French), or a language code
11796 mixed with a country code for specialities in languages (like "fre-
11797 ca" for Canadian French).
11798 stereo_mode
11800 Set stereo 3D video layout of two views in a single video track.
11801 The following values are recognized:
11803 mono
11805 video is not stereo
11806 left_right
11808 Both views are arranged side by side, Left-eye view is on the
11809 left
11810 bottom_top
11812 Both views are arranged in top-bottom orientation, Left-eye
11813 view is at bottom
11814 top_bottom
11816 Both views are arranged in top-bottom orientation, Left-eye
11817 view is on top
11818 checkerboard_rl
11820 Each view is arranged in a checkerboard interleaved pattern,
11821 Left-eye view being first
11822 checkerboard_lr
11824 Each view is arranged in a checkerboard interleaved pattern,
11825 Right-eye view being first
11826 row_interleaved_rl
11828 Each view is constituted by a row based interleaving, Right-eye
11829 view is first row
11830 row_interleaved_lr
11832 Each view is constituted by a row based interleaving, Left-eye
11833 view is first row
11834 col_interleaved_rl
11836 Both views are arranged in a column based interleaving manner,
11837 Right-eye view is first column
11838 col_interleaved_lr
11840 Both views are arranged in a column based interleaving manner,
11841 Left-eye view is first column
11842 anaglyph_cyan_red
11844 All frames are in anaglyph format viewable through red-cyan
11845 filters
11846 right_left
11848 Both views are arranged side by side, Right-eye view is on the
11849 left
11850 anaglyph_green_magenta
11852 All frames are in anaglyph format viewable through green-
11853 magenta filters
11854 block_lr
11856 Both eyes laced in one Block, Left-eye view is first
11857 block_rl
11859 Both eyes laced in one Block, Right-eye view is first
11860 For example a 3D WebM clip can be created using the following command
11862 line:
11863 ffmpeg -i sample_left_right_clip.mpg -an -c:v libvpx -metadata stereo_mode=left_right -y stereo_clip.webm
11865 Options
11867 This muxer supports the following options:
11869 reserve_index_space
11871 By default, this muxer writes the index for seeking (called cues in
11872 Matroska terms) at the end of the file, because it cannot know in
11873 advance how much space to leave for the index at the beginning of
11874 the file. However for some use cases -- e.g. streaming where
11875 seeking is possible but slow -- it is useful to put the index at
11876 the beginning of the file.
11877 If this option is set to a non-zero value, the muxer will reserve a
11879 given amount of space in the file header and then try to write the
11880 cues there when the muxing finishes. If the reserved space does not
11881 suffice, no Cues will be written, the file will be finalized and
11882 writing the trailer will return an error. A safe size for most use
11883 cases should be about 50kB per hour of video.
11884 Note that cues are only written if the output is seekable and this
11886 option will have no effect if it is not.
11887 default_mode
11889 This option controls how the FlagDefault of the output tracks will
11890 be set. It influences which tracks players should play by default.
11891 The default mode is passthrough.
11892 infer
11894 Every track with disposition default will have the FlagDefault
11895 set. Additionally, for each type of track (audio, video or
11896 subtitle), if no track with disposition default of this type
11897 exists, then the first track of this type will be marked as
11898 default (if existing). This ensures that the default flag is
11899 set in a sensible way even if the input originated from
11900 containers that lack the concept of default tracks.
11901 infer_no_subs
11903 This mode is the same as infer except that if no subtitle track
11904 with disposition default exists, no subtitle track will be
11905 marked as default.
11906 passthrough
11908 In this mode the FlagDefault is set if and only if the
11909 AV_DISPOSITION_DEFAULT flag is set in the disposition of the
11910 corresponding stream.
11911 flipped_raw_rgb
11913 If set to true, store positive height for raw RGB bitmaps, which
11914 indicates bitmap is stored bottom-up. Note that this option does
11915 not flip the bitmap which has to be done manually beforehand, e.g.
11916 by using the vflip filter. Default is false and indicates bitmap
11917 is stored top down.
11918 md5
11920 MD5 testing format.
11921 This is a variant of the hash muxer. Unlike that muxer, it defaults to
11923 using the MD5 hash function.
11924 Examples
11926 To compute the MD5 hash of the input converted to raw audio and video,
11928 and store it in the file out.md5:
11929 ffmpeg -i INPUT -f md5 out.md5
11931 You can print the MD5 to stdout with the command:
11933 ffmpeg -i INPUT -f md5 -
11935 See also the hash and framemd5 muxers.
11937 mov, mp4, ismv
11939 MOV/MP4/ISMV (Smooth Streaming) muxer.
11940 The mov/mp4/ismv muxer supports fragmentation. Normally, a MOV/MP4 file
11942 has all the metadata about all packets stored in one location (written
11943 at the end of the file, it can be moved to the start for better
11944 playback by adding faststart to the movflags, or using the qt-faststart
11945 tool). A fragmented file consists of a number of fragments, where
11946 packets and metadata about these packets are stored together. Writing a
11947 fragmented file has the advantage that the file is decodable even if
11948 the writing is interrupted (while a normal MOV/MP4 is undecodable if it
11949 is not properly finished), and it requires less memory when writing
11950 very long files (since writing normal MOV/MP4 files stores info about
11951 every single packet in memory until the file is closed). The downside
11952 is that it is less compatible with other applications.
11953 Options
11955 Fragmentation is enabled by setting one of the AVOptions that define
11957 how to cut the file into fragments:
11958 -moov_size bytes
11960 Reserves space for the moov atom at the beginning of the file
11961 instead of placing the moov atom at the end. If the space reserved
11962 is insufficient, muxing will fail.
11963 -movflags frag_keyframe
11965 Start a new fragment at each video keyframe.
11966 -frag_duration duration
11968 Create fragments that are duration microseconds long.
11969 -frag_size size
11971 Create fragments that contain up to size bytes of payload data.
11972 -movflags frag_custom
11974 Allow the caller to manually choose when to cut fragments, by
11975 calling "av_write_frame(ctx, NULL)" to write a fragment with the
11976 packets written so far. (This is only useful with other
11977 applications integrating libavformat, not from ffmpeg.)
11978 -min_frag_duration duration
11980 Don't create fragments that are shorter than duration microseconds
11981 long.
11982 If more than one condition is specified, fragments are cut when one of
11984 the specified conditions is fulfilled. The exception to this is
11985 "-min_frag_duration", which has to be fulfilled for any of the other
11986 conditions to apply.
11987 Additionally, the way the output file is written can be adjusted
11989 through a few other options:
11990 -movflags empty_moov
11992 Write an initial moov atom directly at the start of the file,
11993 without describing any samples in it. Generally, an mdat/moov pair
11994 is written at the start of the file, as a normal MOV/MP4 file,
11995 containing only a short portion of the file. With this option set,
11996 there is no initial mdat atom, and the moov atom only describes the
11997 tracks but has a zero duration.
11998 This option is implicitly set when writing ismv (Smooth Streaming)
12000 files.
12001 -movflags separate_moof
12003 Write a separate moof (movie fragment) atom for each track.
12004 Normally, packets for all tracks are written in a moof atom (which
12005 is slightly more efficient), but with this option set, the muxer
12006 writes one moof/mdat pair for each track, making it easier to
12007 separate tracks.
12008 This option is implicitly set when writing ismv (Smooth Streaming)
12010 files.
12011 -movflags skip_sidx
12013 Skip writing of sidx atom. When bitrate overhead due to sidx atom
12014 is high, this option could be used for cases where sidx atom is not
12015 mandatory. When global_sidx flag is enabled, this option will be
12016 ignored.
12017 -movflags faststart
12019 Run a second pass moving the index (moov atom) to the beginning of
12020 the file. This operation can take a while, and will not work in
12021 various situations such as fragmented output, thus it is not
12022 enabled by default.
12023 -movflags rtphint
12025 Add RTP hinting tracks to the output file.
12026 -movflags disable_chpl
12028 Disable Nero chapter markers (chpl atom). Normally, both Nero
12029 chapters and a QuickTime chapter track are written to the file.
12030 With this option set, only the QuickTime chapter track will be
12031 written. Nero chapters can cause failures when the file is
12032 reprocessed with certain tagging programs, like mp3Tag 2.61a and
12033 iTunes 11.3, most likely other versions are affected as well.
12034 -movflags omit_tfhd_offset
12036 Do not write any absolute base_data_offset in tfhd atoms. This
12037 avoids tying fragments to absolute byte positions in the
12038 file/streams.
12039 -movflags default_base_moof
12041 Similarly to the omit_tfhd_offset, this flag avoids writing the
12042 absolute base_data_offset field in tfhd atoms, but does so by using
12043 the new default-base-is-moof flag instead. This flag is new from
12044 14496-12:2012. This may make the fragments easier to parse in
12045 certain circumstances (avoiding basing track fragment location
12046 calculations on the implicit end of the previous track fragment).
12047 -write_tmcd
12049 Specify "on" to force writing a timecode track, "off" to disable it
12050 and "auto" to write a timecode track only for mov and mp4 output
12051 (default).
12052 -movflags negative_cts_offsets
12054 Enables utilization of version 1 of the CTTS box, in which the CTS
12055 offsets can be negative. This enables the initial sample to have
12056 DTS/CTS of zero, and reduces the need for edit lists for some cases
12057 such as video tracks with B-frames. Additionally, eases conformance
12058 with the DASH-IF interoperability guidelines.
12059 This option is implicitly set when writing ismv (Smooth Streaming)
12061 files.
12062 -write_prft
12064 Write producer time reference box (PRFT) with a specified time
12065 source for the NTP field in the PRFT box. Set value as wallclock to
12066 specify timesource as wallclock time and pts to specify timesource
12067 as input packets' PTS values.
12068 Setting value to pts is applicable only for a live encoding use
12070 case, where PTS values are set as as wallclock time at the source.
12071 For example, an encoding use case with decklink capture source
12072 where video_pts and audio_pts are set to abs_wallclock.
12073 -movie_timescale scale
12075 Set the timescale written in the movie header box ("mvhd"). Range
12076 is 1 to INT_MAX. Default is 1000.
12077 Example
12079 Smooth Streaming content can be pushed in real time to a publishing
12081 point on IIS with this muxer. Example:
12082 ffmpeg -re <<normal input/transcoding options>> -movflags isml+frag_keyframe -f ismv http://server/publishingpoint.isml/Streams(Encoder1)
12084 mp3
12086 The MP3 muxer writes a raw MP3 stream with the following optional
12087 features:
12088 • An ID3v2 metadata header at the beginning (enabled by default).
12090 Versions 2.3 and 2.4 are supported, the "id3v2_version" private
12091 option controls which one is used (3 or 4). Setting "id3v2_version"
12092 to 0 disables the ID3v2 header completely.
12093 The muxer supports writing attached pictures (APIC frames) to the
12095 ID3v2 header. The pictures are supplied to the muxer in form of a
12096 video stream with a single packet. There can be any number of those
12097 streams, each will correspond to a single APIC frame. The stream
12098 metadata tags title and comment map to APIC description and picture
12099 type respectively. See <http://id3.org/id3v2.4.0-frames> for
12100 allowed picture types.
12101 Note that the APIC frames must be written at the beginning, so the
12103 muxer will buffer the audio frames until it gets all the pictures.
12104 It is therefore advised to provide the pictures as soon as possible
12105 to avoid excessive buffering.
12106 • A Xing/LAME frame right after the ID3v2 header (if present). It is
12108 enabled by default, but will be written only if the output is
12109 seekable. The "write_xing" private option can be used to disable
12110 it. The frame contains various information that may be useful to
12111 the decoder, like the audio duration or encoder delay.
12112 • A legacy ID3v1 tag at the end of the file (disabled by default). It
12114 may be enabled with the "write_id3v1" private option, but as its
12115 capabilities are very limited, its usage is not recommended.
12116 Examples:
12118 Write an mp3 with an ID3v2.3 header and an ID3v1 footer:
12120 ffmpeg -i INPUT -id3v2_version 3 -write_id3v1 1 out.mp3
12122 To attach a picture to an mp3 file select both the audio and the
12124 picture stream with "map":
12125 ffmpeg -i input.mp3 -i cover.png -c copy -map 0 -map 1
12127 -metadata:s:v title="Album cover" -metadata:s:v comment="Cover (Front)" out.mp3
12128 Write a "clean" MP3 without any extra features:
12130 ffmpeg -i input.wav -write_xing 0 -id3v2_version 0 out.mp3
12132 mpegts
12134 MPEG transport stream muxer.
12135 This muxer implements ISO 13818-1 and part of ETSI EN 300 468.
12137 The recognized metadata settings in mpegts muxer are "service_provider"
12139 and "service_name". If they are not set the default for
12140 "service_provider" is FFmpeg and the default for "service_name" is
12141 Service01.
12142 Options
12144 The muxer options are:
12146 mpegts_transport_stream_id integer
12148 Set the transport_stream_id. This identifies a transponder in DVB.
12149 Default is 0x0001.
12150 mpegts_original_network_id integer
12152 Set the original_network_id. This is unique identifier of a network
12153 in DVB. Its main use is in the unique identification of a service
12154 through the path Original_Network_ID, Transport_Stream_ID. Default
12155 is 0x0001.
12156 mpegts_service_id integer
12158 Set the service_id, also known as program in DVB. Default is
12159 0x0001.
12160 mpegts_service_type integer
12162 Set the program service_type. Default is "digital_tv". Accepts the
12163 following options:
12164 hex_value
12166 Any hexadecimal value between 0x01 and 0xff as defined in ETSI
12167 300 468.
12168 digital_tv
12170 Digital TV service.
12171 digital_radio
12173 Digital Radio service.
12174 teletext
12176 Teletext service.
12177 advanced_codec_digital_radio
12179 Advanced Codec Digital Radio service.
12180 mpeg2_digital_hdtv
12182 MPEG2 Digital HDTV service.
12183 advanced_codec_digital_sdtv
12185 Advanced Codec Digital SDTV service.
12186 advanced_codec_digital_hdtv
12188 Advanced Codec Digital HDTV service.
12189 mpegts_pmt_start_pid integer
12191 Set the first PID for PMTs. Default is 0x1000, minimum is 0x0020,
12192 maximum is 0x1ffa. This option has no effect in m2ts mode where the
12193 PMT PID is fixed 0x0100.
12194 mpegts_start_pid integer
12196 Set the first PID for elementary streams. Default is 0x0100,
12197 minimum is 0x0020, maximum is 0x1ffa. This option has no effect in
12198 m2ts mode where the elementary stream PIDs are fixed.
12199 mpegts_m2ts_mode boolean
12201 Enable m2ts mode if set to 1. Default value is "-1" which disables
12202 m2ts mode.
12203 muxrate integer
12205 Set a constant muxrate. Default is VBR.
12206 pes_payload_size integer
12208 Set minimum PES packet payload in bytes. Default is 2930.
12209 mpegts_flags flags
12211 Set mpegts flags. Accepts the following options:
12212 resend_headers
12214 Reemit PAT/PMT before writing the next packet.
12215 latm
12217 Use LATM packetization for AAC.
12218 pat_pmt_at_frames
12220 Reemit PAT and PMT at each video frame.
12221 system_b
12223 Conform to System B (DVB) instead of System A (ATSC).
12224 initial_discontinuity
12226 Mark the initial packet of each stream as discontinuity.
12227 nit Emit NIT table.
12229 mpegts_copyts boolean
12231 Preserve original timestamps, if value is set to 1. Default value
12232 is "-1", which results in shifting timestamps so that they start
12233 from 0.
12234 omit_video_pes_length boolean
12236 Omit the PES packet length for video packets. Default is 1 (true).
12237 pcr_period integer
12239 Override the default PCR retransmission time in milliseconds.
12240 Default is "-1" which means that the PCR interval will be
12241 determined automatically: 20 ms is used for CBR streams, the
12242 highest multiple of the frame duration which is less than 100 ms is
12243 used for VBR streams.
12244 pat_period duration
12246 Maximum time in seconds between PAT/PMT tables. Default is 0.1.
12247 sdt_period duration
12249 Maximum time in seconds between SDT tables. Default is 0.5.
12250 nit_period duration
12252 Maximum time in seconds between NIT tables. Default is 0.5.
12253 tables_version integer
12255 Set PAT, PMT, SDT and NIT version (default 0, valid values are from
12256 0 to 31, inclusively). This option allows updating stream
12257 structure so that standard consumer may detect the change. To do
12258 so, reopen output "AVFormatContext" (in case of API usage) or
12259 restart ffmpeg instance, cyclically changing tables_version value:
12260 ffmpeg -i source1.ts -codec copy -f mpegts -tables_version 0 udp://1.1.1.1:1111
12262 ffmpeg -i source2.ts -codec copy -f mpegts -tables_version 1 udp://1.1.1.1:1111
12263 ...
12264 ffmpeg -i source3.ts -codec copy -f mpegts -tables_version 31 udp://1.1.1.1:1111
12265 ffmpeg -i source1.ts -codec copy -f mpegts -tables_version 0 udp://1.1.1.1:1111
12266 ffmpeg -i source2.ts -codec copy -f mpegts -tables_version 1 udp://1.1.1.1:1111
12267 ...
12268 Example
12270 ffmpeg -i file.mpg -c copy \
12272 -mpegts_original_network_id 0x1122 \
12273 -mpegts_transport_stream_id 0x3344 \
12274 -mpegts_service_id 0x5566 \
12275 -mpegts_pmt_start_pid 0x1500 \
12276 -mpegts_start_pid 0x150 \
12277 -metadata service_provider="Some provider" \
12278 -metadata service_name="Some Channel" \
12279 out.ts
12280 mxf, mxf_d10, mxf_opatom
12282 MXF muxer.
12283 Options
12285 The muxer options are:
12287 store_user_comments bool
12289 Set if user comments should be stored if available or never. IRT
12290 D-10 does not allow user comments. The default is thus to write
12291 them for mxf and mxf_opatom but not for mxf_d10
12292 null
12294 Null muxer.
12295 This muxer does not generate any output file, it is mainly useful for
12297 testing or benchmarking purposes.
12298 For example to benchmark decoding with ffmpeg you can use the command:
12300 ffmpeg -benchmark -i INPUT -f null out.null
12302 Note that the above command does not read or write the out.null file,
12304 but specifying the output file is required by the ffmpeg syntax.
12305 Alternatively you can write the command as:
12307 ffmpeg -benchmark -i INPUT -f null -
12309 nut
12311 -syncpoints flags
12312 Change the syncpoint usage in nut:
12313 default use the normal low-overhead seeking aids.
12315 none do not use the syncpoints at all, reducing the overhead but
12316 making the stream non-seekable;
12317 Use of this option is not recommended, as the resulting files are very damage
12318 sensitive and seeking is not possible. Also in general the overhead from
12319 syncpoints is negligible. Note, -C<write_index> 0 can be used to disable
12320 all growing data tables, allowing to mux endless streams with limited memory
12321 and without these disadvantages.
12322 timestamped extend the syncpoint with a wallclock field.
12324 The none and timestamped flags are experimental.
12326 -write_index bool
12328 Write index at the end, the default is to write an index.
12329 ffmpeg -i INPUT -f_strict experimental -syncpoints none - | processor
12331 ogg
12333 Ogg container muxer.
12334 -page_duration duration
12336 Preferred page duration, in microseconds. The muxer will attempt to
12337 create pages that are approximately duration microseconds long.
12338 This allows the user to compromise between seek granularity and
12339 container overhead. The default is 1 second. A value of 0 will fill
12340 all segments, making pages as large as possible. A value of 1 will
12341 effectively use 1 packet-per-page in most situations, giving a
12342 small seek granularity at the cost of additional container
12343 overhead.
12344 -serial_offset value
12346 Serial value from which to set the streams serial number. Setting
12347 it to different and sufficiently large values ensures that the
12348 produced ogg files can be safely chained.
12349 raw muxers
12351 Raw muxers accept a single stream matching the designated codec. They
12352 do not store timestamps or metadata. The recognized extension is the
12353 same as the muxer name unless indicated otherwise.
12354 ac3
12356 Dolby Digital, also known as AC-3, audio.
12358 adx
12360 CRI Middleware ADX audio.
12362 This muxer will write out the total sample count near the start of the
12364 first packet when the output is seekable and the count can be stored in
12365 32 bits.
12366 aptx
12368 aptX (Audio Processing Technology for Bluetooth) audio.
12370 aptx_hd
12372 aptX HD (Audio Processing Technology for Bluetooth) audio.
12374 Extensions: aptxhd
12376 avs2
12378 AVS2-P2/IEEE1857.4 video.
12380 Extensions: avs, avs2
12382 cavsvideo
12384 Chinese AVS (Audio Video Standard) video.
12386 Extensions: cavs
12388 codec2raw
12390 Codec 2 audio.
12392 No extension is registered so format name has to be supplied e.g. with
12394 the ffmpeg CLI tool "-f codec2raw".
12395 data
12397 Data muxer accepts a single stream with any codec of any type. The
12399 input stream has to be selected using the "-map" option with the ffmpeg
12400 CLI tool.
12401 No extension is registered so format name has to be supplied e.g. with
12403 the ffmpeg CLI tool "-f data".
12404 dirac
12406 BBC Dirac video. The Dirac Pro codec is a subset and is standardized as
12408 SMPTE VC-2.
12409 Extensions: drc, vc2
12411 dnxhd
12413 Avid DNxHD video. It is standardized as SMPTE VC-3. Accepts DNxHR
12415 streams.
12416 Extensions: dnxhd, dnxhr
12418 dts
12420 DTS Coherent Acoustics (DCA) audio.
12422 eac3
12424 Dolby Digital Plus, also known as Enhanced AC-3, audio.
12426 g722
12428 ITU-T G.722 audio.
12430 g723_1
12432 ITU-T G.723.1 audio.
12434 Extensions: tco, rco
12436 g726
12438 ITU-T G.726 big-endian ("left-justified") audio.
12440 No extension is registered so format name has to be supplied e.g. with
12442 the ffmpeg CLI tool "-f g726".
12443 g726le
12445 ITU-T G.726 little-endian ("right-justified") audio.
12447 No extension is registered so format name has to be supplied e.g. with
12449 the ffmpeg CLI tool "-f g726le".
12450 gsm
12452 Global System for Mobile Communications audio.
12454 h261
12456 ITU-T H.261 video.
12458 h263
12460 ITU-T H.263 / H.263-1996, H.263+ / H.263-1998 / H.263 version 2 video.
12462 h264
12464 ITU-T H.264 / MPEG-4 Part 10 AVC video. Bitstream shall be converted to
12466 Annex B syntax if it's in length-prefixed mode.
12467 Extensions: h264, 264
12469 hevc
12471 ITU-T H.265 / MPEG-H Part 2 HEVC video. Bitstream shall be converted to
12473 Annex B syntax if it's in length-prefixed mode.
12474 Extensions: hevc, h265, 265
12476 m4v
12478 MPEG-4 Part 2 video.
12480 mjpeg
12482 Motion JPEG video.
12484 Extensions: mjpg, mjpeg
12486 mlp
12488 Meridian Lossless Packing, also known as Packed PCM, audio.
12490 mp2
12492 MPEG-1 Audio Layer II audio.
12494 Extensions: mp2, m2a, mpa
12496 mpeg1video
12498 MPEG-1 Part 2 video.
12500 Extensions: mpg, mpeg, m1v
12502 mpeg2video
12504 ITU-T H.262 / MPEG-2 Part 2 video.
12506 Extensions: m2v
12508 obu
12510 AV1 low overhead Open Bitstream Units muxer. Temporal delimiter OBUs
12512 will be inserted in all temporal units of the stream.
12513 rawvideo
12515 Raw uncompressed video.
12517 Extensions: yuv, rgb
12519 sbc
12521 Bluetooth SIG low-complexity subband codec audio.
12523 Extensions: sbc, msbc
12525 truehd
12527 Dolby TrueHD audio.
12529 Extensions: thd
12531 vc1
12533 SMPTE 421M / VC-1 video.
12535 segment, stream_segment, ssegment
12537 Basic stream segmenter.
12538 This muxer outputs streams to a number of separate files of nearly
12540 fixed duration. Output filename pattern can be set in a fashion similar
12541 to image2, or by using a "strftime" template if the strftime option is
12542 enabled.
12543 "stream_segment" is a variant of the muxer used to write to streaming
12545 output formats, i.e. which do not require global headers, and is
12546 recommended for outputting e.g. to MPEG transport stream segments.
12547 "ssegment" is a shorter alias for "stream_segment".
12548 Every segment starts with a keyframe of the selected reference stream,
12550 which is set through the reference_stream option.
12551 Note that if you want accurate splitting for a video file, you need to
12553 make the input key frames correspond to the exact splitting times
12554 expected by the segmenter, or the segment muxer will start the new
12555 segment with the key frame found next after the specified start time.
12556 The segment muxer works best with a single constant frame rate video.
12558 Optionally it can generate a list of the created segments, by setting
12560 the option segment_list. The list type is specified by the
12561 segment_list_type option. The entry filenames in the segment list are
12562 set by default to the basename of the corresponding segment files.
12563 See also the hls muxer, which provides a more specific implementation
12565 for HLS segmentation.
12566 Options
12568 The segment muxer supports the following options:
12570 increment_tc 1|0
12572 if set to 1, increment timecode between each segment If this is
12573 selected, the input need to have a timecode in the first video
12574 stream. Default value is 0.
12575 reference_stream specifier
12577 Set the reference stream, as specified by the string specifier. If
12578 specifier is set to "auto", the reference is chosen automatically.
12579 Otherwise it must be a stream specifier (see the ``Stream
12580 specifiers'' chapter in the ffmpeg manual) which specifies the
12581 reference stream. The default value is "auto".
12582 segment_format format
12584 Override the inner container format, by default it is guessed by
12585 the filename extension.
12586 segment_format_options options_list
12588 Set output format options using a :-separated list of key=value
12589 parameters. Values containing the ":" special character must be
12590 escaped.
12591 segment_list name
12593 Generate also a listfile named name. If not specified no listfile
12594 is generated.
12595 segment_list_flags flags
12597 Set flags affecting the segment list generation.
12598 It currently supports the following flags:
12600 cache
12602 Allow caching (only affects M3U8 list files).
12603 live
12605 Allow live-friendly file generation.
12606 segment_list_size size
12608 Update the list file so that it contains at most size segments. If
12609 0 the list file will contain all the segments. Default value is 0.
12610 segment_list_entry_prefix prefix
12612 Prepend prefix to each entry. Useful to generate absolute paths.
12613 By default no prefix is applied.
12614 segment_list_type type
12616 Select the listing format.
12617 The following values are recognized:
12619 flat
12621 Generate a flat list for the created segments, one segment per
12622 line.
12623 csv, ext
12625 Generate a list for the created segments, one segment per line,
12626 each line matching the format (comma-separated values):
12627 <segment_filename>,<segment_start_time>,<segment_end_time>
12629 segment_filename is the name of the output file generated by
12631 the muxer according to the provided pattern. CSV escaping
12632 (according to RFC4180) is applied if required.
12633 segment_start_time and segment_end_time specify the segment
12635 start and end time expressed in seconds.
12636 A list file with the suffix ".csv" or ".ext" will auto-select
12638 this format.
12639 ext is deprecated in favor or csv.
12641 ffconcat
12643 Generate an ffconcat file for the created segments. The
12644 resulting file can be read using the FFmpeg concat demuxer.
12645 A list file with the suffix ".ffcat" or ".ffconcat" will auto-
12647 select this format.
12648 m3u8
12650 Generate an extended M3U8 file, version 3, compliant with
12651 <http://tools.ietf.org/id/draft-pantos-http-live-streaming>.
12652 A list file with the suffix ".m3u8" will auto-select this
12654 format.
12655 If not specified the type is guessed from the list file name
12657 suffix.
12658 segment_time time
12660 Set segment duration to time, the value must be a duration
12661 specification. Default value is "2". See also the segment_times
12662 option.
12663 Note that splitting may not be accurate, unless you force the
12665 reference stream key-frames at the given time. See the introductory
12666 notice and the examples below.
12667 segment_atclocktime 1|0
12669 If set to "1" split at regular clock time intervals starting from
12670 00:00 o'clock. The time value specified in segment_time is used for
12671 setting the length of the splitting interval.
12672 For example with segment_time set to "900" this makes it possible
12674 to create files at 12:00 o'clock, 12:15, 12:30, etc.
12675 Default value is "0".
12677 segment_clocktime_offset duration
12679 Delay the segment splitting times with the specified duration when
12680 using segment_atclocktime.
12681 For example with segment_time set to "900" and
12683 segment_clocktime_offset set to "300" this makes it possible to
12684 create files at 12:05, 12:20, 12:35, etc.
12685 Default value is "0".
12687 segment_clocktime_wrap_duration duration
12689 Force the segmenter to only start a new segment if a packet reaches
12690 the muxer within the specified duration after the segmenting clock
12691 time. This way you can make the segmenter more resilient to
12692 backward local time jumps, such as leap seconds or transition to
12693 standard time from daylight savings time.
12694 Default is the maximum possible duration which means starting a new
12696 segment regardless of the elapsed time since the last clock time.
12697 segment_time_delta delta
12699 Specify the accuracy time when selecting the start time for a
12700 segment, expressed as a duration specification. Default value is
12701 "0".
12702 When delta is specified a key-frame will start a new segment if its
12704 PTS satisfies the relation:
12705 PTS >= start_time - time_delta
12707 This option is useful when splitting video content, which is always
12709 split at GOP boundaries, in case a key frame is found just before
12710 the specified split time.
12711 In particular may be used in combination with the ffmpeg option
12713 force_key_frames. The key frame times specified by force_key_frames
12714 may not be set accurately because of rounding issues, with the
12715 consequence that a key frame time may result set just before the
12716 specified time. For constant frame rate videos a value of
12717 1/(2*frame_rate) should address the worst case mismatch between the
12718 specified time and the time set by force_key_frames.
12719 segment_times times
12721 Specify a list of split points. times contains a list of comma
12722 separated duration specifications, in increasing order. See also
12723 the segment_time option.
12724 segment_frames frames
12726 Specify a list of split video frame numbers. frames contains a list
12727 of comma separated integer numbers, in increasing order.
12728 This option specifies to start a new segment whenever a reference
12730 stream key frame is found and the sequential number (starting from
12731 0) of the frame is greater or equal to the next value in the list.
12732 segment_wrap limit
12734 Wrap around segment index once it reaches limit.
12735 segment_start_number number
12737 Set the sequence number of the first segment. Defaults to 0.
12738 strftime 1|0
12740 Use the "strftime" function to define the name of the new segments
12741 to write. If this is selected, the output segment name must contain
12742 a "strftime" function template. Default value is 0.
12743 break_non_keyframes 1|0
12745 If enabled, allow segments to start on frames other than keyframes.
12746 This improves behavior on some players when the time between
12747 keyframes is inconsistent, but may make things worse on others, and
12748 can cause some oddities during seeking. Defaults to 0.
12749 reset_timestamps 1|0
12751 Reset timestamps at the beginning of each segment, so that each
12752 segment will start with near-zero timestamps. It is meant to ease
12753 the playback of the generated segments. May not work with some
12754 combinations of muxers/codecs. It is set to 0 by default.
12755 initial_offset offset
12757 Specify timestamp offset to apply to the output packet timestamps.
12758 The argument must be a time duration specification, and defaults to
12759 0.
12760 write_empty_segments 1|0
12762 If enabled, write an empty segment if there are no packets during
12763 the period a segment would usually span. Otherwise, the segment
12764 will be filled with the next packet written. Defaults to 0.
12765 Make sure to require a closed GOP when encoding and to set the GOP size
12767 to fit your segment time constraint.
12768 Examples
12770 • Remux the content of file in.mkv to a list of segments out-000.nut,
12772 out-001.nut, etc., and write the list of generated segments to
12773 out.list:
12774 ffmpeg -i in.mkv -codec hevc -flags +cgop -g 60 -map 0 -f segment -segment_list out.list out%03d.nut
12776 • Segment input and set output format options for the output
12778 segments:
12779 ffmpeg -i in.mkv -f segment -segment_time 10 -segment_format_options movflags=+faststart out%03d.mp4
12781 • Segment the input file according to the split points specified by
12783 the segment_times option:
12784 ffmpeg -i in.mkv -codec copy -map 0 -f segment -segment_list out.csv -segment_times 1,2,3,5,8,13,21 out%03d.nut
12786 • Use the ffmpeg force_key_frames option to force key frames in the
12788 input at the specified location, together with the segment option
12789 segment_time_delta to account for possible roundings operated when
12790 setting key frame times.
12791 ffmpeg -i in.mkv -force_key_frames 1,2,3,5,8,13,21 -codec:v mpeg4 -codec:a pcm_s16le -map 0 \
12793 -f segment -segment_list out.csv -segment_times 1,2,3,5,8,13,21 -segment_time_delta 0.05 out%03d.nut
12794 In order to force key frames on the input file, transcoding is
12796 required.
12797 • Segment the input file by splitting the input file according to the
12799 frame numbers sequence specified with the segment_frames option:
12800 ffmpeg -i in.mkv -codec copy -map 0 -f segment -segment_list out.csv -segment_frames 100,200,300,500,800 out%03d.nut
12802 • Convert the in.mkv to TS segments using the "libx264" and "aac"
12804 encoders:
12805 ffmpeg -i in.mkv -map 0 -codec:v libx264 -codec:a aac -f ssegment -segment_list out.list out%03d.ts
12807 • Segment the input file, and create an M3U8 live playlist (can be
12809 used as live HLS source):
12810 ffmpeg -re -i in.mkv -codec copy -map 0 -f segment -segment_list playlist.m3u8 \
12812 -segment_list_flags +live -segment_time 10 out%03d.mkv
12813 smoothstreaming
12815 Smooth Streaming muxer generates a set of files (Manifest, chunks)
12816 suitable for serving with conventional web server.
12817 window_size
12819 Specify the number of fragments kept in the manifest. Default 0
12820 (keep all).
12821 extra_window_size
12823 Specify the number of fragments kept outside of the manifest before
12824 removing from disk. Default 5.
12825 lookahead_count
12827 Specify the number of lookahead fragments. Default 2.
12828 min_frag_duration
12830 Specify the minimum fragment duration (in microseconds). Default
12831 5000000.
12832 remove_at_exit
12834 Specify whether to remove all fragments when finished. Default 0
12835 (do not remove).
12836 streamhash
12838 Per stream hash testing format.
12839 This muxer computes and prints a cryptographic hash of all the input
12841 frames, on a per-stream basis. This can be used for equality checks
12842 without having to do a complete binary comparison.
12843 By default audio frames are converted to signed 16-bit raw audio and
12845 video frames to raw video before computing the hash, but the output of
12846 explicit conversions to other codecs can also be used. Timestamps are
12847 ignored. It uses the SHA-256 cryptographic hash function by default,
12848 but supports several other algorithms.
12849 The output of the muxer consists of one line per stream of the form:
12851 streamindex,streamtype,algo=hash, where streamindex is the index of the
12852 mapped stream, streamtype is a single character indicating the type of
12853 stream, algo is a short string representing the hash function used, and
12854 hash is a hexadecimal number representing the computed hash.
12855 hash algorithm
12857 Use the cryptographic hash function specified by the string
12858 algorithm. Supported values include "MD5", "murmur3", "RIPEMD128",
12859 "RIPEMD160", "RIPEMD256", "RIPEMD320", "SHA160", "SHA224", "SHA256"
12860 (default), "SHA512/224", "SHA512/256", "SHA384", "SHA512", "CRC32"
12861 and "adler32".
12862 Examples
12864 To compute the SHA-256 hash of the input converted to raw audio and
12866 video, and store it in the file out.sha256:
12867 ffmpeg -i INPUT -f streamhash out.sha256
12869 To print an MD5 hash to stdout use the command:
12871 ffmpeg -i INPUT -f streamhash -hash md5 -
12873 See also the hash and framehash muxers.
12875 tee
12877 The tee muxer can be used to write the same data to several outputs,
12878 such as files or streams. It can be used, for example, to stream a
12879 video over a network and save it to disk at the same time.
12880 It is different from specifying several outputs to the ffmpeg command-
12882 line tool. With the tee muxer, the audio and video data will be encoded
12883 only once. With conventional multiple outputs, multiple encoding
12884 operations in parallel are initiated, which can be a very expensive
12885 process. The tee muxer is not useful when using the libavformat API
12886 directly because it is then possible to feed the same packets to
12887 several muxers directly.
12888 Since the tee muxer does not represent any particular output format,
12890 ffmpeg cannot auto-select output streams. So all streams intended for
12891 output must be specified using "-map". See the examples below.
12892 Some encoders may need different options depending on the output
12894 format; the auto-detection of this can not work with the tee muxer, so
12895 they need to be explicitly specified. The main example is the
12896 global_header flag.
12897 The slave outputs are specified in the file name given to the muxer,
12899 separated by '|'. If any of the slave name contains the '|' separator,
12900 leading or trailing spaces or any special character, those must be
12901 escaped (see the "Quoting and escaping" section in the ffmpeg-utils(1)
12902 manual).
12903 Options
12905 use_fifo bool
12907 If set to 1, slave outputs will be processed in separate threads
12908 using the fifo muxer. This allows to compensate for different
12909 speed/latency/reliability of outputs and setup transparent
12910 recovery. By default this feature is turned off.
12911 fifo_options
12913 Options to pass to fifo pseudo-muxer instances. See fifo.
12914 Muxer options can be specified for each slave by prepending them as a
12916 list of key=value pairs separated by ':', between square brackets. If
12917 the options values contain a special character or the ':' separator,
12918 they must be escaped; note that this is a second level escaping.
12919 The following special options are also recognized:
12921 f Specify the format name. Required if it cannot be guessed from the
12923 output URL.
12924 bsfs[/spec]
12926 Specify a list of bitstream filters to apply to the specified
12927 output.
12928 It is possible to specify to which streams a given bitstream filter
12930 applies, by appending a stream specifier to the option separated by
12931 "/". spec must be a stream specifier (see Format stream
12932 specifiers).
12933 If the stream specifier is not specified, the bitstream filters
12935 will be applied to all streams in the output. This will cause that
12936 output operation to fail if the output contains streams to which
12937 the bitstream filter cannot be applied e.g. "h264_mp4toannexb"
12938 being applied to an output containing an audio stream.
12939 Options for a bitstream filter must be specified in the form of
12941 "opt=value".
12942 Several bitstream filters can be specified, separated by ",".
12944 use_fifo bool
12946 This allows to override tee muxer use_fifo option for individual
12947 slave muxer.
12948 fifo_options
12950 This allows to override tee muxer fifo_options for individual slave
12951 muxer. See fifo.
12952 select
12954 Select the streams that should be mapped to the slave output,
12955 specified by a stream specifier. If not specified, this defaults to
12956 all the mapped streams. This will cause that output operation to
12957 fail if the output format does not accept all mapped streams.
12958 You may use multiple stream specifiers separated by commas (",")
12960 e.g.: "a:0,v"
12961 onfail
12963 Specify behaviour on output failure. This can be set to either
12964 "abort" (which is default) or "ignore". "abort" will cause whole
12965 process to fail in case of failure on this slave output. "ignore"
12966 will ignore failure on this output, so other outputs will continue
12967 without being affected.
12968 Examples
12970 • Encode something and both archive it in a WebM file and stream it
12972 as MPEG-TS over UDP:
12973 ffmpeg -i ... -c:v libx264 -c:a mp2 -f tee -map 0:v -map 0:a
12975 "archive-20121107.mkv|[f=mpegts]udp://10.0.1.255:1234/"
12976 • As above, but continue streaming even if output to local file fails
12978 (for example local drive fills up):
12979 ffmpeg -i ... -c:v libx264 -c:a mp2 -f tee -map 0:v -map 0:a
12981 "[onfail=ignore]archive-20121107.mkv|[f=mpegts]udp://10.0.1.255:1234/"
12982 • Use ffmpeg to encode the input, and send the output to three
12984 different destinations. The "dump_extra" bitstream filter is used
12985 to add extradata information to all the output video keyframes
12986 packets, as requested by the MPEG-TS format. The select option is
12987 applied to out.aac in order to make it contain only audio packets.
12988 ffmpeg -i ... -map 0 -flags +global_header -c:v libx264 -c:a aac
12990 -f tee "[bsfs/v=dump_extra=freq=keyframe]out.ts|[movflags=+faststart]out.mp4|[select=a]out.aac"
12991 • As above, but select only stream "a:1" for the audio output. Note
12993 that a second level escaping must be performed, as ":" is a special
12994 character used to separate options.
12995 ffmpeg -i ... -map 0 -flags +global_header -c:v libx264 -c:a aac
12997 -f tee "[bsfs/v=dump_extra=freq=keyframe]out.ts|[movflags=+faststart]out.mp4|[select=\'a:1\']out.aac"
12998 webm_chunk
13000 WebM Live Chunk Muxer.
13001 This muxer writes out WebM headers and chunks as separate files which
13003 can be consumed by clients that support WebM Live streams via DASH.
13004 Options
13006 This muxer supports the following options:
13008 chunk_start_index
13010 Index of the first chunk (defaults to 0).
13011 header
13013 Filename of the header where the initialization data will be
13014 written.
13015 audio_chunk_duration
13017 Duration of each audio chunk in milliseconds (defaults to 5000).
13018 Example
13020 ffmpeg -f v4l2 -i /dev/video0 \
13022 -f alsa -i hw:0 \
13023 -map 0:0 \
13024 -c:v libvpx-vp9 \
13025 -s 640x360 -keyint_min 30 -g 30 \
13026 -f webm_chunk \
13027 -header webm_live_video_360.hdr \
13028 -chunk_start_index 1 \
13029 webm_live_video_360_%d.chk \
13030 -map 1:0 \
13031 -c:a libvorbis \
13032 -b:a 128k \
13033 -f webm_chunk \
13034 -header webm_live_audio_128.hdr \
13035 -chunk_start_index 1 \
13036 -audio_chunk_duration 1000 \
13037 webm_live_audio_128_%d.chk
13038 webm_dash_manifest
13040 WebM DASH Manifest muxer.
13041 This muxer implements the WebM DASH Manifest specification to generate
13043 the DASH manifest XML. It also supports manifest generation for DASH
13044 live streams.
13045 For more information see:
13047 • WebM DASH Specification:
13049 <https://sites.google.com/a/webmproject.org/wiki/adaptive-streaming/webm-dash-specification>
13050 • ISO DASH Specification:
13052 <http://standards.iso.org/ittf/PubliclyAvailableStandards/c065274_ISO_IEC_23009-1_2014.zip>
13053 Options
13055 This muxer supports the following options:
13057 adaptation_sets
13059 This option has the following syntax: "id=x,streams=a,b,c
13060 id=y,streams=d,e" where x and y are the unique identifiers of the
13061 adaptation sets and a,b,c,d and e are the indices of the
13062 corresponding audio and video streams. Any number of adaptation
13063 sets can be added using this option.
13064 live
13066 Set this to 1 to create a live stream DASH Manifest. Default: 0.
13067 chunk_start_index
13069 Start index of the first chunk. This will go in the startNumber
13070 attribute of the SegmentTemplate element in the manifest. Default:
13071 0.
13072 chunk_duration_ms
13074 Duration of each chunk in milliseconds. This will go in the
13075 duration attribute of the SegmentTemplate element in the manifest.
13076 Default: 1000.
13077 utc_timing_url
13079 URL of the page that will return the UTC timestamp in ISO format.
13080 This will go in the value attribute of the UTCTiming element in the
13081 manifest. Default: None.
13082 time_shift_buffer_depth
13084 Smallest time (in seconds) shifting buffer for which any
13085 Representation is guaranteed to be available. This will go in the
13086 timeShiftBufferDepth attribute of the MPD element. Default: 60.
13087 minimum_update_period
13089 Minimum update period (in seconds) of the manifest. This will go in
13090 the minimumUpdatePeriod attribute of the MPD element. Default: 0.
13091 Example
13093 ffmpeg -f webm_dash_manifest -i video1.webm \
13095 -f webm_dash_manifest -i video2.webm \
13096 -f webm_dash_manifest -i audio1.webm \
13097 -f webm_dash_manifest -i audio2.webm \
13098 -map 0 -map 1 -map 2 -map 3 \
13099 -c copy \
13100 -f webm_dash_manifest \
13101 -adaptation_sets "id=0,streams=0,1 id=1,streams=2,3" \
13102 manifest.xml
13103
13105 FFmpeg is able to dump metadata from media files into a simple
13106 UTF-8-encoded INI-like text file and then load it back using the
13107 metadata muxer/demuxer.
13108 The file format is as follows:
13110 1. A file consists of a header and a number of metadata tags divided
13112 into sections, each on its own line.
13113 2. The header is a ;FFMETADATA string, followed by a version number
13115 (now 1).
13116 3. Metadata tags are of the form key=value
13118 4. Immediately after header follows global metadata
13120 5. After global metadata there may be sections with
13122 per-stream/per-chapter metadata.
13123 6. A section starts with the section name in uppercase (i.e. STREAM or
13125 CHAPTER) in brackets ([, ]) and ends with next section or end of
13126 file.
13127 7. At the beginning of a chapter section there may be an optional
13129 timebase to be used for start/end values. It must be in form
13130 TIMEBASE=num/den, where num and den are integers. If the timebase
13131 is missing then start/end times are assumed to be in nanoseconds.
13132 Next a chapter section must contain chapter start and end times in
13134 form START=num, END=num, where num is a positive integer.
13135 8. Empty lines and lines starting with ; or # are ignored.
13137 9. Metadata keys or values containing special characters (=, ;, #, \
13139 and a newline) must be escaped with a backslash \.
13140 10. Note that whitespace in metadata (e.g. foo = bar) is considered to
13142 be a part of the tag (in the example above key is foo , value is
13143 bar).
13144 A ffmetadata file might look like this:
13146 ;FFMETADATA1
13148 title=bike\\shed
13149 ;this is a comment
13150 artist=FFmpeg troll team
13151 [CHAPTER]
13153 TIMEBASE=1/1000
13154 START=0
13155 #chapter ends at 0:01:00
13156 END=60000
13157 title=chapter \#1
13158 [STREAM]
13159 title=multi\
13160 line
13161 By using the ffmetadata muxer and demuxer it is possible to extract
13163 metadata from an input file to an ffmetadata file, and then transcode
13164 the file into an output file with the edited ffmetadata file.
13165 Extracting an ffmetadata file with ffmpeg goes as follows:
13167 ffmpeg -i INPUT -f ffmetadata FFMETADATAFILE
13169 Reinserting edited metadata information from the FFMETADATAFILE file
13171 can be done as:
13172 ffmpeg -i INPUT -i FFMETADATAFILE -map_metadata 1 -codec copy OUTPUT
13174
13176 The libavformat library provides some generic global options, which can
13177 be set on all the protocols. In addition each protocol may support so-
13178 called private options, which are specific for that component.
13179 Options may be set by specifying -option value in the FFmpeg tools, or
13181 by setting the value explicitly in the "AVFormatContext" options or
13182 using the libavutil/opt.h API for programmatic use.
13183 The list of supported options follows:
13185 protocol_whitelist list (input)
13187 Set a ","-separated list of allowed protocols. "ALL" matches all
13188 protocols. Protocols prefixed by "-" are disabled. All protocols
13189 are allowed by default but protocols used by an another protocol
13190 (nested protocols) are restricted to a per protocol subset.
13191
13193 Protocols are configured elements in FFmpeg that enable access to
13194 resources that require specific protocols.
13195 When you configure your FFmpeg build, all the supported protocols are
13197 enabled by default. You can list all available ones using the configure
13198 option "--list-protocols".
13199 You can disable all the protocols using the configure option
13201 "--disable-protocols", and selectively enable a protocol using the
13202 option "--enable-protocol=PROTOCOL", or you can disable a particular
13203 protocol using the option "--disable-protocol=PROTOCOL".
13204 The option "-protocols" of the ff* tools will display the list of
13206 supported protocols.
13207 All protocols accept the following options:
13209 rw_timeout
13211 Maximum time to wait for (network) read/write operations to
13212 complete, in microseconds.
13213 A description of the currently available protocols follows.
13215 amqp
13217 Advanced Message Queueing Protocol (AMQP) version 0-9-1 is a broker
13218 based publish-subscribe communication protocol.
13219 FFmpeg must be compiled with --enable-librabbitmq to support AMQP. A
13221 separate AMQP broker must also be run. An example open-source AMQP
13222 broker is RabbitMQ.
13223 After starting the broker, an FFmpeg client may stream data to the
13225 broker using the command:
13226 ffmpeg -re -i input -f mpegts amqp://[[user]:[password]@]hostname[:port][/vhost]
13228 Where hostname and port (default is 5672) is the address of the broker.
13230 The client may also set a user/password for authentication. The default
13231 for both fields is "guest". Name of virtual host on broker can be set
13232 with vhost. The default value is "/".
13233 Muliple subscribers may stream from the broker using the command:
13235 ffplay amqp://[[user]:[password]@]hostname[:port][/vhost]
13237 In RabbitMQ all data published to the broker flows through a specific
13239 exchange, and each subscribing client has an assigned queue/buffer.
13240 When a packet arrives at an exchange, it may be copied to a client's
13241 queue depending on the exchange and routing_key fields.
13242 The following options are supported:
13244 exchange
13246 Sets the exchange to use on the broker. RabbitMQ has several
13247 predefined exchanges: "amq.direct" is the default exchange, where
13248 the publisher and subscriber must have a matching routing_key;
13249 "amq.fanout" is the same as a broadcast operation (i.e. the data is
13250 forwarded to all queues on the fanout exchange independent of the
13251 routing_key); and "amq.topic" is similar to "amq.direct", but
13252 allows for more complex pattern matching (refer to the RabbitMQ
13253 documentation).
13254 routing_key
13256 Sets the routing key. The default value is "amqp". The routing key
13257 is used on the "amq.direct" and "amq.topic" exchanges to decide
13258 whether packets are written to the queue of a subscriber.
13259 pkt_size
13261 Maximum size of each packet sent/received to the broker. Default is
13262 131072. Minimum is 4096 and max is any large value (representable
13263 by an int). When receiving packets, this sets an internal buffer
13264 size in FFmpeg. It should be equal to or greater than the size of
13265 the published packets to the broker. Otherwise the received message
13266 may be truncated causing decoding errors.
13267 connection_timeout
13269 The timeout in seconds during the initial connection to the broker.
13270 The default value is rw_timeout, or 5 seconds if rw_timeout is not
13271 set.
13272 delivery_mode mode
13274 Sets the delivery mode of each message sent to broker. The
13275 following values are accepted:
13276 persistent
13278 Delivery mode set to "persistent" (2). This is the default
13279 value. Messages may be written to the broker's disk depending
13280 on its setup.
13281 non-persistent
13283 Delivery mode set to "non-persistent" (1). Messages will stay
13284 in broker's memory unless the broker is under memory pressure.
13285 async
13287 Asynchronous data filling wrapper for input stream.
13288 Fill data in a background thread, to decouple I/O operation from demux
13290 thread.
13291 async:<URL>
13293 async:http://host/resource
13294 async:cache:http://host/resource
13295 bluray
13297 Read BluRay playlist.
13298 The accepted options are:
13300 angle
13302 BluRay angle
13303 chapter
13305 Start chapter (1...N)
13306 playlist
13308 Playlist to read (BDMV/PLAYLIST/?????.mpls)
13309 Examples:
13311 Read longest playlist from BluRay mounted to /mnt/bluray:
13313 bluray:/mnt/bluray
13315 Read angle 2 of playlist 4 from BluRay mounted to /mnt/bluray, start
13317 from chapter 2:
13318 -playlist 4 -angle 2 -chapter 2 bluray:/mnt/bluray
13320 cache
13322 Caching wrapper for input stream.
13323 Cache the input stream to temporary file. It brings seeking capability
13325 to live streams.
13326 The accepted options are:
13328 read_ahead_limit
13330 Amount in bytes that may be read ahead when seeking isn't
13331 supported. Range is -1 to INT_MAX. -1 for unlimited. Default is
13332 65536.
13333 URL Syntax is
13335 cache:<URL>
13337 concat
13339 Physical concatenation protocol.
13340 Read and seek from many resources in sequence as if they were a unique
13342 resource.
13343 A URL accepted by this protocol has the syntax:
13345 concat:<URL1>|<URL2>|...|<URLN>
13347 where URL1, URL2, ..., URLN are the urls of the resource to be
13349 concatenated, each one possibly specifying a distinct protocol.
13350 For example to read a sequence of files split1.mpeg, split2.mpeg,
13352 split3.mpeg with ffplay use the command:
13353 ffplay concat:split1.mpeg\|split2.mpeg\|split3.mpeg
13355 Note that you may need to escape the character "|" which is special for
13357 many shells.
13358 concatf
13360 Physical concatenation protocol using a line break delimited list of
13361 resources.
13362 Read and seek from many resources in sequence as if they were a unique
13364 resource.
13365 A URL accepted by this protocol has the syntax:
13367 concatf:<URL>
13369 where URL is the url containing a line break delimited list of
13371 resources to be concatenated, each one possibly specifying a distinct
13372 protocol. Special characters must be escaped with backslash or single
13373 quotes. See the "Quoting and escaping" section in the ffmpeg-utils(1)
13374 manual.
13375 For example to read a sequence of files split1.mpeg, split2.mpeg,
13377 split3.mpeg listed in separate lines within a file split.txt with
13378 ffplay use the command:
13379 ffplay concatf:split.txt
13381 Where split.txt contains the lines:
13383 split1.mpeg
13385 split2.mpeg
13386 split3.mpeg
13387 crypto
13389 AES-encrypted stream reading protocol.
13390 The accepted options are:
13392 key Set the AES decryption key binary block from given hexadecimal
13394 representation.
13395 iv Set the AES decryption initialization vector binary block from
13397 given hexadecimal representation.
13398 Accepted URL formats:
13400 crypto:<URL>
13402 crypto+<URL>
13403 data
13405 Data in-line in the URI. See
13406 <http://en.wikipedia.org/wiki/Data_URI_scheme>.
13407 For example, to convert a GIF file given inline with ffmpeg:
13409 ffmpeg -i "data:image/gif;base64,R0lGODdhCAAIAMIEAAAAAAAA//8AAP//AP///////////////ywAAAAACAAIAAADF0gEDLojDgdGiJdJqUX02iB4E8Q9jUMkADs=" smiley.png
13411 file
13413 File access protocol.
13414 Read from or write to a file.
13416 A file URL can have the form:
13418 file:<filename>
13420 where filename is the path of the file to read.
13422 An URL that does not have a protocol prefix will be assumed to be a
13424 file URL. Depending on the build, an URL that looks like a Windows path
13425 with the drive letter at the beginning will also be assumed to be a
13426 file URL (usually not the case in builds for unix-like systems).
13427 For example to read from a file input.mpeg with ffmpeg use the command:
13429 ffmpeg -i file:input.mpeg output.mpeg
13431 This protocol accepts the following options:
13433 truncate
13435 Truncate existing files on write, if set to 1. A value of 0
13436 prevents truncating. Default value is 1.
13437 blocksize
13439 Set I/O operation maximum block size, in bytes. Default value is
13440 "INT_MAX", which results in not limiting the requested block size.
13441 Setting this value reasonably low improves user termination request
13442 reaction time, which is valuable for files on slow medium.
13443 follow
13445 If set to 1, the protocol will retry reading at the end of the
13446 file, allowing reading files that still are being written. In order
13447 for this to terminate, you either need to use the rw_timeout
13448 option, or use the interrupt callback (for API users).
13449 seekable
13451 Controls if seekability is advertised on the file. 0 means non-
13452 seekable, -1 means auto (seekable for normal files, non-seekable
13453 for named pipes).
13454 Many demuxers handle seekable and non-seekable resources
13456 differently, overriding this might speed up opening certain files
13457 at the cost of losing some features (e.g. accurate seeking).
13458 ftp
13460 FTP (File Transfer Protocol).
13461 Read from or write to remote resources using FTP protocol.
13463 Following syntax is required.
13465 ftp://[user[:password]@]server[:port]/path/to/remote/resource.mpeg
13467 This protocol accepts the following options.
13469 timeout
13471 Set timeout in microseconds of socket I/O operations used by the
13472 underlying low level operation. By default it is set to -1, which
13473 means that the timeout is not specified.
13474 ftp-user
13476 Set a user to be used for authenticating to the FTP server. This is
13477 overridden by the user in the FTP URL.
13478 ftp-password
13480 Set a password to be used for authenticating to the FTP server.
13481 This is overridden by the password in the FTP URL, or by ftp-
13482 anonymous-password if no user is set.
13483 ftp-anonymous-password
13485 Password used when login as anonymous user. Typically an e-mail
13486 address should be used.
13487 ftp-write-seekable
13489 Control seekability of connection during encoding. If set to 1 the
13490 resource is supposed to be seekable, if set to 0 it is assumed not
13491 to be seekable. Default value is 0.
13492 NOTE: Protocol can be used as output, but it is recommended to not do
13494 it, unless special care is taken (tests, customized server
13495 configuration etc.). Different FTP servers behave in different way
13496 during seek operation. ff* tools may produce incomplete content due to
13497 server limitations.
13498 gopher
13500 Gopher protocol.
13501 gophers
13503 Gophers protocol.
13504 The Gopher protocol with TLS encapsulation.
13506 hls
13508 Read Apple HTTP Live Streaming compliant segmented stream as a uniform
13509 one. The M3U8 playlists describing the segments can be remote HTTP
13510 resources or local files, accessed using the standard file protocol.
13511 The nested protocol is declared by specifying "+proto" after the hls
13512 URI scheme name, where proto is either "file" or "http".
13513 hls+http://host/path/to/remote/resource.m3u8
13515 hls+file://path/to/local/resource.m3u8
13516 Using this protocol is discouraged - the hls demuxer should work just
13518 as well (if not, please report the issues) and is more complete. To
13519 use the hls demuxer instead, simply use the direct URLs to the m3u8
13520 files.
13521 http
13523 HTTP (Hyper Text Transfer Protocol).
13524 This protocol accepts the following options:
13526 seekable
13528 Control seekability of connection. If set to 1 the resource is
13529 supposed to be seekable, if set to 0 it is assumed not to be
13530 seekable, if set to -1 it will try to autodetect if it is seekable.
13531 Default value is -1.
13532 chunked_post
13534 If set to 1 use chunked Transfer-Encoding for posts, default is 1.
13535 content_type
13537 Set a specific content type for the POST messages or for listen
13538 mode.
13539 http_proxy
13541 set HTTP proxy to tunnel through e.g. http://example.com:1234
13542 headers
13544 Set custom HTTP headers, can override built in default headers. The
13545 value must be a string encoding the headers.
13546 multiple_requests
13548 Use persistent connections if set to 1, default is 0.
13549 post_data
13551 Set custom HTTP post data.
13552 referer
13554 Set the Referer header. Include 'Referer: URL' header in HTTP
13555 request.
13556 user_agent
13558 Override the User-Agent header. If not specified the protocol will
13559 use a string describing the libavformat build. ("Lavf/<version>")
13560 reconnect_at_eof
13562 If set then eof is treated like an error and causes reconnection,
13563 this is useful for live / endless streams.
13564 reconnect_streamed
13566 If set then even streamed/non seekable streams will be reconnected
13567 on errors.
13568 reconnect_on_network_error
13570 Reconnect automatically in case of TCP/TLS errors during connect.
13571 reconnect_on_http_error
13573 A comma separated list of HTTP status codes to reconnect on. The
13574 list can include specific status codes (e.g. '503') or the strings
13575 '4xx' / '5xx'.
13576 reconnect_delay_max
13578 Sets the maximum delay in seconds after which to give up
13579 reconnecting
13580 mime_type
13582 Export the MIME type.
13583 http_version
13585 Exports the HTTP response version number. Usually "1.0" or "1.1".
13586 icy If set to 1 request ICY (SHOUTcast) metadata from the server. If
13588 the server supports this, the metadata has to be retrieved by the
13589 application by reading the icy_metadata_headers and
13590 icy_metadata_packet options. The default is 1.
13591 icy_metadata_headers
13593 If the server supports ICY metadata, this contains the ICY-specific
13594 HTTP reply headers, separated by newline characters.
13595 icy_metadata_packet
13597 If the server supports ICY metadata, and icy was set to 1, this
13598 contains the last non-empty metadata packet sent by the server. It
13599 should be polled in regular intervals by applications interested in
13600 mid-stream metadata updates.
13601 cookies
13603 Set the cookies to be sent in future requests. The format of each
13604 cookie is the same as the value of a Set-Cookie HTTP response
13605 field. Multiple cookies can be delimited by a newline character.
13606 offset
13608 Set initial byte offset.
13609 end_offset
13611 Try to limit the request to bytes preceding this offset.
13612 method
13614 When used as a client option it sets the HTTP method for the
13615 request.
13616 When used as a server option it sets the HTTP method that is going
13618 to be expected from the client(s). If the expected and the
13619 received HTTP method do not match the client will be given a Bad
13620 Request response. When unset the HTTP method is not checked for
13621 now. This will be replaced by autodetection in the future.
13622 listen
13624 If set to 1 enables experimental HTTP server. This can be used to
13625 send data when used as an output option, or read data from a client
13626 with HTTP POST when used as an input option. If set to 2 enables
13627 experimental multi-client HTTP server. This is not yet implemented
13628 in ffmpeg.c and thus must not be used as a command line option.
13629 # Server side (sending):
13631 ffmpeg -i somefile.ogg -c copy -listen 1 -f ogg http://<server>:<port>
13632 # Client side (receiving):
13634 ffmpeg -i http://<server>:<port> -c copy somefile.ogg
13635 # Client can also be done with wget:
13637 wget http://<server>:<port> -O somefile.ogg
13638 # Server side (receiving):
13640 ffmpeg -listen 1 -i http://<server>:<port> -c copy somefile.ogg
13641 # Client side (sending):
13643 ffmpeg -i somefile.ogg -chunked_post 0 -c copy -f ogg http://<server>:<port>
13644 # Client can also be done with wget:
13646 wget --post-file=somefile.ogg http://<server>:<port>
13647 send_expect_100
13649 Send an Expect: 100-continue header for POST. If set to 1 it will
13650 send, if set to 0 it won't, if set to -1 it will try to send if it
13651 is applicable. Default value is -1.
13652 auth_type
13654 Set HTTP authentication type. No option for Digest, since this
13655 method requires getting nonce parameters from the server first and
13656 can't be used straight away like Basic.
13657 none
13659 Choose the HTTP authentication type automatically. This is the
13660 default.
13661 basic
13663 Choose the HTTP basic authentication.
13664 Basic authentication sends a Base64-encoded string that
13666 contains a user name and password for the client. Base64 is not
13667 a form of encryption and should be considered the same as
13668 sending the user name and password in clear text (Base64 is a
13669 reversible encoding). If a resource needs to be protected,
13670 strongly consider using an authentication scheme other than
13671 basic authentication. HTTPS/TLS should be used with basic
13672 authentication. Without these additional security
13673 enhancements, basic authentication should not be used to
13674 protect sensitive or valuable information.
13675 HTTP Cookies
13677 Some HTTP requests will be denied unless cookie values are passed in
13679 with the request. The cookies option allows these cookies to be
13680 specified. At the very least, each cookie must specify a value along
13681 with a path and domain. HTTP requests that match both the domain and
13682 path will automatically include the cookie value in the HTTP Cookie
13683 header field. Multiple cookies can be delimited by a newline.
13684 The required syntax to play a stream specifying a cookie is:
13686 ffplay -cookies "nlqptid=nltid=tsn; path=/; domain=somedomain.com;" http://somedomain.com/somestream.m3u8
13688 Icecast
13690 Icecast protocol (stream to Icecast servers)
13691 This protocol accepts the following options:
13693 ice_genre
13695 Set the stream genre.
13696 ice_name
13698 Set the stream name.
13699 ice_description
13701 Set the stream description.
13702 ice_url
13704 Set the stream website URL.
13705 ice_public
13707 Set if the stream should be public. The default is 0 (not public).
13708 user_agent
13710 Override the User-Agent header. If not specified a string of the
13711 form "Lavf/<version>" will be used.
13712 password
13714 Set the Icecast mountpoint password.
13715 content_type
13717 Set the stream content type. This must be set if it is different
13718 from audio/mpeg.
13719 legacy_icecast
13721 This enables support for Icecast versions < 2.4.0, that do not
13722 support the HTTP PUT method but the SOURCE method.
13723 tls Establish a TLS (HTTPS) connection to Icecast.
13725 icecast://[<username>[:<password>]@]<server>:<port>/<mountpoint>
13727 mmst
13729 MMS (Microsoft Media Server) protocol over TCP.
13730 mmsh
13732 MMS (Microsoft Media Server) protocol over HTTP.
13733 The required syntax is:
13735 mmsh://<server>[:<port>][/<app>][/<playpath>]
13737 md5
13739 MD5 output protocol.
13740 Computes the MD5 hash of the data to be written, and on close writes
13742 this to the designated output or stdout if none is specified. It can be
13743 used to test muxers without writing an actual file.
13744 Some examples follow.
13746 # Write the MD5 hash of the encoded AVI file to the file output.avi.md5.
13748 ffmpeg -i input.flv -f avi -y md5:output.avi.md5
13749 # Write the MD5 hash of the encoded AVI file to stdout.
13751 ffmpeg -i input.flv -f avi -y md5:
13752 Note that some formats (typically MOV) require the output protocol to
13754 be seekable, so they will fail with the MD5 output protocol.
13755 pipe
13757 UNIX pipe access protocol.
13758 Read and write from UNIX pipes.
13760 The accepted syntax is:
13762 pipe:[<number>]
13764 number is the number corresponding to the file descriptor of the pipe
13766 (e.g. 0 for stdin, 1 for stdout, 2 for stderr). If number is not
13767 specified, by default the stdout file descriptor will be used for
13768 writing, stdin for reading.
13769 For example to read from stdin with ffmpeg:
13771 cat test.wav | ffmpeg -i pipe:0
13773 # ...this is the same as...
13774 cat test.wav | ffmpeg -i pipe:
13775 For writing to stdout with ffmpeg:
13777 ffmpeg -i test.wav -f avi pipe:1 | cat > test.avi
13779 # ...this is the same as...
13780 ffmpeg -i test.wav -f avi pipe: | cat > test.avi
13781 This protocol accepts the following options:
13783 blocksize
13785 Set I/O operation maximum block size, in bytes. Default value is
13786 "INT_MAX", which results in not limiting the requested block size.
13787 Setting this value reasonably low improves user termination request
13788 reaction time, which is valuable if data transmission is slow.
13789 Note that some formats (typically MOV), require the output protocol to
13791 be seekable, so they will fail with the pipe output protocol.
13792 prompeg
13794 Pro-MPEG Code of Practice #3 Release 2 FEC protocol.
13795 The Pro-MPEG CoP#3 FEC is a 2D parity-check forward error correction
13797 mechanism for MPEG-2 Transport Streams sent over RTP.
13798 This protocol must be used in conjunction with the "rtp_mpegts" muxer
13800 and the "rtp" protocol.
13801 The required syntax is:
13803 -f rtp_mpegts -fec prompeg=<option>=<val>... rtp://<hostname>:<port>
13805 The destination UDP ports are "port + 2" for the column FEC stream and
13807 "port + 4" for the row FEC stream.
13808 This protocol accepts the following options:
13810 l=n The number of columns (4-20, LxD <= 100)
13812 d=n The number of rows (4-20, LxD <= 100)
13814 Example usage:
13816 -f rtp_mpegts -fec prompeg=l=8:d=4 rtp://<hostname>:<port>
13818 rist
13820 Reliable Internet Streaming Transport protocol
13821 The accepted options are:
13823 rist_profile
13825 Supported values:
13826 simple
13828 main
13829 This one is default.
13830 advanced
13832 buffer_size
13833 Set internal RIST buffer size in milliseconds for retransmission of
13834 data. Default value is 0 which means the librist default (1 sec).
13835 Maximum value is 30 seconds.
13836 pkt_size
13838 Set maximum packet size for sending data. 1316 by default.
13839 log_level
13841 Set loglevel for RIST logging messages. You only need to set this
13842 if you explicitly want to enable debug level messages or packet
13843 loss simulation, otherwise the regular loglevel is respected.
13844 secret
13846 Set override of encryption secret, by default is unset.
13847 encryption
13849 Set encryption type, by default is disabled. Acceptable values are
13850 128 and 256.
13851 rtmp
13853 Real-Time Messaging Protocol.
13854 The Real-Time Messaging Protocol (RTMP) is used for streaming
13856 multimedia content across a TCP/IP network.
13857 The required syntax is:
13859 rtmp://[<username>:<password>@]<server>[:<port>][/<app>][/<instance>][/<playpath>]
13861 The accepted parameters are:
13863 username
13865 An optional username (mostly for publishing).
13866 password
13868 An optional password (mostly for publishing).
13869 server
13871 The address of the RTMP server.
13872 port
13874 The number of the TCP port to use (by default is 1935).
13875 app It is the name of the application to access. It usually corresponds
13877 to the path where the application is installed on the RTMP server
13878 (e.g. /ondemand/, /flash/live/, etc.). You can override the value
13879 parsed from the URI through the "rtmp_app" option, too.
13880 playpath
13882 It is the path or name of the resource to play with reference to
13883 the application specified in app, may be prefixed by "mp4:". You
13884 can override the value parsed from the URI through the
13885 "rtmp_playpath" option, too.
13886 listen
13888 Act as a server, listening for an incoming connection.
13889 timeout
13891 Maximum time to wait for the incoming connection. Implies listen.
13892 Additionally, the following parameters can be set via command line
13894 options (or in code via "AVOption"s):
13895 rtmp_app
13897 Name of application to connect on the RTMP server. This option
13898 overrides the parameter specified in the URI.
13899 rtmp_buffer
13901 Set the client buffer time in milliseconds. The default is 3000.
13902 rtmp_conn
13904 Extra arbitrary AMF connection parameters, parsed from a string,
13905 e.g. like "B:1 S:authMe O:1 NN:code:1.23 NS:flag:ok O:0". Each
13906 value is prefixed by a single character denoting the type, B for
13907 Boolean, N for number, S for string, O for object, or Z for null,
13908 followed by a colon. For Booleans the data must be either 0 or 1
13909 for FALSE or TRUE, respectively. Likewise for Objects the data
13910 must be 0 or 1 to end or begin an object, respectively. Data items
13911 in subobjects may be named, by prefixing the type with 'N' and
13912 specifying the name before the value (i.e. "NB:myFlag:1"). This
13913 option may be used multiple times to construct arbitrary AMF
13914 sequences.
13915 rtmp_flashver
13917 Version of the Flash plugin used to run the SWF player. The default
13918 is LNX 9,0,124,2. (When publishing, the default is FMLE/3.0
13919 (compatible; <libavformat version>).)
13920 rtmp_flush_interval
13922 Number of packets flushed in the same request (RTMPT only). The
13923 default is 10.
13924 rtmp_live
13926 Specify that the media is a live stream. No resuming or seeking in
13927 live streams is possible. The default value is "any", which means
13928 the subscriber first tries to play the live stream specified in the
13929 playpath. If a live stream of that name is not found, it plays the
13930 recorded stream. The other possible values are "live" and
13931 "recorded".
13932 rtmp_pageurl
13934 URL of the web page in which the media was embedded. By default no
13935 value will be sent.
13936 rtmp_playpath
13938 Stream identifier to play or to publish. This option overrides the
13939 parameter specified in the URI.
13940 rtmp_subscribe
13942 Name of live stream to subscribe to. By default no value will be
13943 sent. It is only sent if the option is specified or if rtmp_live
13944 is set to live.
13945 rtmp_swfhash
13947 SHA256 hash of the decompressed SWF file (32 bytes).
13948 rtmp_swfsize
13950 Size of the decompressed SWF file, required for SWFVerification.
13951 rtmp_swfurl
13953 URL of the SWF player for the media. By default no value will be
13954 sent.
13955 rtmp_swfverify
13957 URL to player swf file, compute hash/size automatically.
13958 rtmp_tcurl
13960 URL of the target stream. Defaults to proto://host[:port]/app.
13961 tcp_nodelay=1|0
13963 Set TCP_NODELAY to disable Nagle's algorithm. Default value is 0.
13964 Remark: Writing to the socket is currently not optimized to
13966 minimize system calls and reduces the efficiency / effect of
13967 TCP_NODELAY.
13968 For example to read with ffplay a multimedia resource named "sample"
13970 from the application "vod" from an RTMP server "myserver":
13971 ffplay rtmp://myserver/vod/sample
13973 To publish to a password protected server, passing the playpath and app
13975 names separately:
13976 ffmpeg -re -i <input> -f flv -rtmp_playpath some/long/path -rtmp_app long/app/name rtmp://username:password@myserver/
13978 rtmpe
13980 Encrypted Real-Time Messaging Protocol.
13981 The Encrypted Real-Time Messaging Protocol (RTMPE) is used for
13983 streaming multimedia content within standard cryptographic primitives,
13984 consisting of Diffie-Hellman key exchange and HMACSHA256, generating a
13985 pair of RC4 keys.
13986 rtmps
13988 Real-Time Messaging Protocol over a secure SSL connection.
13989 The Real-Time Messaging Protocol (RTMPS) is used for streaming
13991 multimedia content across an encrypted connection.
13992 rtmpt
13994 Real-Time Messaging Protocol tunneled through HTTP.
13995 The Real-Time Messaging Protocol tunneled through HTTP (RTMPT) is used
13997 for streaming multimedia content within HTTP requests to traverse
13998 firewalls.
13999 rtmpte
14001 Encrypted Real-Time Messaging Protocol tunneled through HTTP.
14002 The Encrypted Real-Time Messaging Protocol tunneled through HTTP
14004 (RTMPTE) is used for streaming multimedia content within HTTP requests
14005 to traverse firewalls.
14006 rtmpts
14008 Real-Time Messaging Protocol tunneled through HTTPS.
14009 The Real-Time Messaging Protocol tunneled through HTTPS (RTMPTS) is
14011 used for streaming multimedia content within HTTPS requests to traverse
14012 firewalls.
14013 libsmbclient
14015 libsmbclient permits one to manipulate CIFS/SMB network resources.
14016 Following syntax is required.
14018 smb://[[domain:]user[:password@]]server[/share[/path[/file]]]
14020 This protocol accepts the following options.
14022 timeout
14024 Set timeout in milliseconds of socket I/O operations used by the
14025 underlying low level operation. By default it is set to -1, which
14026 means that the timeout is not specified.
14027 truncate
14029 Truncate existing files on write, if set to 1. A value of 0
14030 prevents truncating. Default value is 1.
14031 workgroup
14033 Set the workgroup used for making connections. By default workgroup
14034 is not specified.
14035 For more information see: <http://www.samba.org/>.
14037 libssh
14039 Secure File Transfer Protocol via libssh
14040 Read from or write to remote resources using SFTP protocol.
14042 Following syntax is required.
14044 sftp://[user[:password]@]server[:port]/path/to/remote/resource.mpeg
14046 This protocol accepts the following options.
14048 timeout
14050 Set timeout of socket I/O operations used by the underlying low
14051 level operation. By default it is set to -1, which means that the
14052 timeout is not specified.
14053 truncate
14055 Truncate existing files on write, if set to 1. A value of 0
14056 prevents truncating. Default value is 1.
14057 private_key
14059 Specify the path of the file containing private key to use during
14060 authorization. By default libssh searches for keys in the ~/.ssh/
14061 directory.
14062 Example: Play a file stored on remote server.
14064 ffplay sftp://user:password@server_address:22/home/user/resource.mpeg
14066 librtmp rtmp, rtmpe, rtmps, rtmpt, rtmpte
14068 Real-Time Messaging Protocol and its variants supported through
14069 librtmp.
14070 Requires the presence of the librtmp headers and library during
14072 configuration. You need to explicitly configure the build with
14073 "--enable-librtmp". If enabled this will replace the native RTMP
14074 protocol.
14075 This protocol provides most client functions and a few server functions
14077 needed to support RTMP, RTMP tunneled in HTTP (RTMPT), encrypted RTMP
14078 (RTMPE), RTMP over SSL/TLS (RTMPS) and tunneled variants of these
14079 encrypted types (RTMPTE, RTMPTS).
14080 The required syntax is:
14082 <rtmp_proto>://<server>[:<port>][/<app>][/<playpath>] <options>
14084 where rtmp_proto is one of the strings "rtmp", "rtmpt", "rtmpe",
14086 "rtmps", "rtmpte", "rtmpts" corresponding to each RTMP variant, and
14087 server, port, app and playpath have the same meaning as specified for
14088 the RTMP native protocol. options contains a list of space-separated
14089 options of the form key=val.
14090 See the librtmp manual page (man 3 librtmp) for more information.
14092 For example, to stream a file in real-time to an RTMP server using
14094 ffmpeg:
14095 ffmpeg -re -i myfile -f flv rtmp://myserver/live/mystream
14097 To play the same stream using ffplay:
14099 ffplay "rtmp://myserver/live/mystream live=1"
14101 rtp
14103 Real-time Transport Protocol.
14104 The required syntax for an RTP URL is:
14106 rtp://hostname[:port][?option=val...]
14107 port specifies the RTP port to use.
14109 The following URL options are supported:
14111 ttl=n
14113 Set the TTL (Time-To-Live) value (for multicast only).
14114 rtcpport=n
14116 Set the remote RTCP port to n.
14117 localrtpport=n
14119 Set the local RTP port to n.
14120 localrtcpport=n'
14122 Set the local RTCP port to n.
14123 pkt_size=n
14125 Set max packet size (in bytes) to n.
14126 buffer_size=size
14128 Set the maximum UDP socket buffer size in bytes.
14129 connect=0|1
14131 Do a "connect()" on the UDP socket (if set to 1) or not (if set to
14132 0).
14133 sources=ip[,ip]
14135 List allowed source IP addresses.
14136 block=ip[,ip]
14138 List disallowed (blocked) source IP addresses.
14139 write_to_source=0|1
14141 Send packets to the source address of the latest received packet
14142 (if set to 1) or to a default remote address (if set to 0).
14143 localport=n
14145 Set the local RTP port to n.
14146 localaddr=addr
14148 Local IP address of a network interface used for sending packets or
14149 joining multicast groups.
14150 timeout=n
14152 Set timeout (in microseconds) of socket I/O operations to n.
14153 This is a deprecated option. Instead, localrtpport should be used.
14155 Important notes:
14157 1. If rtcpport is not set the RTCP port will be set to the RTP port
14159 value plus 1.
14160 2. If localrtpport (the local RTP port) is not set any available port
14162 will be used for the local RTP and RTCP ports.
14163 3. If localrtcpport (the local RTCP port) is not set it will be set to
14165 the local RTP port value plus 1.
14166 rtsp
14168 Real-Time Streaming Protocol.
14169 RTSP is not technically a protocol handler in libavformat, it is a
14171 demuxer and muxer. The demuxer supports both normal RTSP (with data
14172 transferred over RTP; this is used by e.g. Apple and Microsoft) and
14173 Real-RTSP (with data transferred over RDT).
14174 The muxer can be used to send a stream using RTSP ANNOUNCE to a server
14176 supporting it (currently Darwin Streaming Server and Mischa
14177 Spiegelmock's <https://github.com/revmischa/rtsp-server>).
14178 The required syntax for a RTSP url is:
14180 rtsp://<hostname>[:<port>]/<path>
14182 Options can be set on the ffmpeg/ffplay command line, or set in code
14184 via "AVOption"s or in "avformat_open_input".
14185 The following options are supported.
14187 initial_pause
14189 Do not start playing the stream immediately if set to 1. Default
14190 value is 0.
14191 rtsp_transport
14193 Set RTSP transport protocols.
14194 It accepts the following values:
14196 udp Use UDP as lower transport protocol.
14198 tcp Use TCP (interleaving within the RTSP control channel) as lower
14200 transport protocol.
14201 udp_multicast
14203 Use UDP multicast as lower transport protocol.
14204 http
14206 Use HTTP tunneling as lower transport protocol, which is useful
14207 for passing proxies.
14208 Multiple lower transport protocols may be specified, in that case
14210 they are tried one at a time (if the setup of one fails, the next
14211 one is tried). For the muxer, only the tcp and udp options are
14212 supported.
14213 rtsp_flags
14215 Set RTSP flags.
14216 The following values are accepted:
14218 filter_src
14220 Accept packets only from negotiated peer address and port.
14221 listen
14223 Act as a server, listening for an incoming connection.
14224 prefer_tcp
14226 Try TCP for RTP transport first, if TCP is available as RTSP
14227 RTP transport.
14228 Default value is none.
14230 allowed_media_types
14232 Set media types to accept from the server.
14233 The following flags are accepted:
14235 video
14237 audio
14238 data
14239 By default it accepts all media types.
14241 min_port
14243 Set minimum local UDP port. Default value is 5000.
14244 max_port
14246 Set maximum local UDP port. Default value is 65000.
14247 listen_timeout
14249 Set maximum timeout (in seconds) to establish an initial
14250 connection. Setting listen_timeout > 0 sets rtsp_flags to listen.
14251 Default is -1 which means an infinite timeout when listen mode is
14252 set.
14253 reorder_queue_size
14255 Set number of packets to buffer for handling of reordered packets.
14256 timeout
14258 Set socket TCP I/O timeout in microseconds.
14259 user_agent
14261 Override User-Agent header. If not specified, it defaults to the
14262 libavformat identifier string.
14263 When receiving data over UDP, the demuxer tries to reorder received
14265 packets (since they may arrive out of order, or packets may get lost
14266 totally). This can be disabled by setting the maximum demuxing delay to
14267 zero (via the "max_delay" field of AVFormatContext).
14268 When watching multi-bitrate Real-RTSP streams with ffplay, the streams
14270 to display can be chosen with "-vst" n and "-ast" n for video and audio
14271 respectively, and can be switched on the fly by pressing "v" and "a".
14272 Examples
14274 The following examples all make use of the ffplay and ffmpeg tools.
14276 • Watch a stream over UDP, with a max reordering delay of 0.5
14278 seconds:
14279 ffplay -max_delay 500000 -rtsp_transport udp rtsp://server/video.mp4
14281 • Watch a stream tunneled over HTTP:
14283 ffplay -rtsp_transport http rtsp://server/video.mp4
14285 • Send a stream in realtime to a RTSP server, for others to watch:
14287 ffmpeg -re -i <input> -f rtsp -muxdelay 0.1 rtsp://server/live.sdp
14289 • Receive a stream in realtime:
14291 ffmpeg -rtsp_flags listen -i rtsp://ownaddress/live.sdp <output>
14293 sap
14295 Session Announcement Protocol (RFC 2974). This is not technically a
14296 protocol handler in libavformat, it is a muxer and demuxer. It is used
14297 for signalling of RTP streams, by announcing the SDP for the streams
14298 regularly on a separate port.
14299 Muxer
14301 The syntax for a SAP url given to the muxer is:
14303 sap://<destination>[:<port>][?<options>]
14305 The RTP packets are sent to destination on port port, or to port 5004
14307 if no port is specified. options is a "&"-separated list. The
14308 following options are supported:
14309 announce_addr=address
14311 Specify the destination IP address for sending the announcements
14312 to. If omitted, the announcements are sent to the commonly used
14313 SAP announcement multicast address 224.2.127.254 (sap.mcast.net),
14314 or ff0e::2:7ffe if destination is an IPv6 address.
14315 announce_port=port
14317 Specify the port to send the announcements on, defaults to 9875 if
14318 not specified.
14319 ttl=ttl
14321 Specify the time to live value for the announcements and RTP
14322 packets, defaults to 255.
14323 same_port=0|1
14325 If set to 1, send all RTP streams on the same port pair. If zero
14326 (the default), all streams are sent on unique ports, with each
14327 stream on a port 2 numbers higher than the previous. VLC/Live555
14328 requires this to be set to 1, to be able to receive the stream.
14329 The RTP stack in libavformat for receiving requires all streams to
14330 be sent on unique ports.
14331 Example command lines follow.
14333 To broadcast a stream on the local subnet, for watching in VLC:
14335 ffmpeg -re -i <input> -f sap sap://224.0.0.255?same_port=1
14337 Similarly, for watching in ffplay:
14339 ffmpeg -re -i <input> -f sap sap://224.0.0.255
14341 And for watching in ffplay, over IPv6:
14343 ffmpeg -re -i <input> -f sap sap://[ff0e::1:2:3:4]
14345 Demuxer
14347 The syntax for a SAP url given to the demuxer is:
14349 sap://[<address>][:<port>]
14351 address is the multicast address to listen for announcements on, if
14353 omitted, the default 224.2.127.254 (sap.mcast.net) is used. port is the
14354 port that is listened on, 9875 if omitted.
14355 The demuxers listens for announcements on the given address and port.
14357 Once an announcement is received, it tries to receive that particular
14358 stream.
14359 Example command lines follow.
14361 To play back the first stream announced on the normal SAP multicast
14363 address:
14364 ffplay sap://
14366 To play back the first stream announced on one the default IPv6 SAP
14368 multicast address:
14369 ffplay sap://[ff0e::2:7ffe]
14371 sctp
14373 Stream Control Transmission Protocol.
14374 The accepted URL syntax is:
14376 sctp://<host>:<port>[?<options>]
14378 The protocol accepts the following options:
14380 listen
14382 If set to any value, listen for an incoming connection. Outgoing
14383 connection is done by default.
14384 max_streams
14386 Set the maximum number of streams. By default no limit is set.
14387 srt
14389 Haivision Secure Reliable Transport Protocol via libsrt.
14390 The supported syntax for a SRT URL is:
14392 srt://<hostname>:<port>[?<options>]
14394 options contains a list of &-separated options of the form key=val.
14396 or
14398 <options> srt://<hostname>:<port>
14400 options contains a list of '-key val' options.
14402 This protocol accepts the following options.
14404 connect_timeout=milliseconds
14406 Connection timeout; SRT cannot connect for RTT > 1500 msec (2
14407 handshake exchanges) with the default connect timeout of 3 seconds.
14408 This option applies to the caller and rendezvous connection modes.
14409 The connect timeout is 10 times the value set for the rendezvous
14410 mode (which can be used as a workaround for this connection problem
14411 with earlier versions).
14412 ffs=bytes
14414 Flight Flag Size (Window Size), in bytes. FFS is actually an
14415 internal parameter and you should set it to not less than
14416 recv_buffer_size and mss. The default value is relatively large,
14417 therefore unless you set a very large receiver buffer, you do not
14418 need to change this option. Default value is 25600.
14419 inputbw=bytes/seconds
14421 Sender nominal input rate, in bytes per seconds. Used along with
14422 oheadbw, when maxbw is set to relative (0), to calculate maximum
14423 sending rate when recovery packets are sent along with the main
14424 media stream: inputbw * (100 + oheadbw) / 100 if inputbw is not set
14425 while maxbw is set to relative (0), the actual input rate is
14426 evaluated inside the library. Default value is 0.
14427 iptos=tos
14429 IP Type of Service. Applies to sender only. Default value is 0xB8.
14430 ipttl=ttl
14432 IP Time To Live. Applies to sender only. Default value is 64.
14433 latency=microseconds
14435 Timestamp-based Packet Delivery Delay. Used to absorb bursts of
14436 missed packet retransmissions. This flag sets both rcvlatency and
14437 peerlatency to the same value. Note that prior to version 1.3.0
14438 this is the only flag to set the latency, however this is
14439 effectively equivalent to setting peerlatency, when side is sender
14440 and rcvlatency when side is receiver, and the bidirectional stream
14441 sending is not supported.
14442 listen_timeout=microseconds
14444 Set socket listen timeout.
14445 maxbw=bytes/seconds
14447 Maximum sending bandwidth, in bytes per seconds. -1 infinite
14448 (CSRTCC limit is 30mbps) 0 relative to input rate (see inputbw) >0
14449 absolute limit value Default value is 0 (relative)
14450 mode=caller|listener|rendezvous
14452 Connection mode. caller opens client connection. listener starts
14453 server to listen for incoming connections. rendezvous use Rendez-
14454 Vous connection mode. Default value is caller.
14455 mss=bytes
14457 Maximum Segment Size, in bytes. Used for buffer allocation and rate
14458 calculation using a packet counter assuming fully filled packets.
14459 The smallest MSS between the peers is used. This is 1500 by default
14460 in the overall internet. This is the maximum size of the UDP
14461 packet and can be only decreased, unless you have some unusual
14462 dedicated network settings. Default value is 1500.
14463 nakreport=1|0
14465 If set to 1, Receiver will send `UMSG_LOSSREPORT` messages
14466 periodically until a lost packet is retransmitted or intentionally
14467 dropped. Default value is 1.
14468 oheadbw=percents
14470 Recovery bandwidth overhead above input rate, in percents. See
14471 inputbw. Default value is 25%.
14472 passphrase=string
14474 HaiCrypt Encryption/Decryption Passphrase string, length from 10 to
14475 79 characters. The passphrase is the shared secret between the
14476 sender and the receiver. It is used to generate the Key Encrypting
14477 Key using PBKDF2 (Password-Based Key Derivation Function). It is
14478 used only if pbkeylen is non-zero. It is used on the receiver only
14479 if the received data is encrypted. The configured passphrase
14480 cannot be recovered (write-only).
14481 enforced_encryption=1|0
14483 If true, both connection parties must have the same password set
14484 (including empty, that is, with no encryption). If the password
14485 doesn't match or only one side is unencrypted, the connection is
14486 rejected. Default is true.
14487 kmrefreshrate=packets
14489 The number of packets to be transmitted after which the encryption
14490 key is switched to a new key. Default is -1. -1 means auto
14491 (0x1000000 in srt library). The range for this option is integers
14492 in the 0 - "INT_MAX".
14493 kmpreannounce=packets
14495 The interval between when a new encryption key is sent and when
14496 switchover occurs. This value also applies to the subsequent
14497 interval between when switchover occurs and when the old encryption
14498 key is decommissioned. Default is -1. -1 means auto (0x1000 in srt
14499 library). The range for this option is integers in the 0 -
14500 "INT_MAX".
14501 snddropdelay=microseconds
14503 The sender's extra delay before dropping packets. This delay is
14504 added to the default drop delay time interval value.
14505 Special value -1: Do not drop packets on the sender at all.
14507 payload_size=bytes
14509 Sets the maximum declared size of a packet transferred during the
14510 single call to the sending function in Live mode. Use 0 if this
14511 value isn't used (which is default in file mode). Default is -1
14512 (automatic), which typically means MPEG-TS; if you are going to use
14513 SRT to send any different kind of payload, such as, for example,
14514 wrapping a live stream in very small frames, then you can use a
14515 bigger maximum frame size, though not greater than 1456 bytes.
14516 pkt_size=bytes
14518 Alias for payload_size.
14519 peerlatency=microseconds
14521 The latency value (as described in rcvlatency) that is set by the
14522 sender side as a minimum value for the receiver.
14523 pbkeylen=bytes
14525 Sender encryption key length, in bytes. Only can be set to 0, 16,
14526 24 and 32. Enable sender encryption if not 0. Not required on
14527 receiver (set to 0), key size obtained from sender in HaiCrypt
14528 handshake. Default value is 0.
14529 rcvlatency=microseconds
14531 The time that should elapse since the moment when the packet was
14532 sent and the moment when it's delivered to the receiver application
14533 in the receiving function. This time should be a buffer time large
14534 enough to cover the time spent for sending, unexpectedly extended
14535 RTT time, and the time needed to retransmit the lost UDP packet.
14536 The effective latency value will be the maximum of this options'
14537 value and the value of peerlatency set by the peer side. Before
14538 version 1.3.0 this option is only available as latency.
14539 recv_buffer_size=bytes
14541 Set UDP receive buffer size, expressed in bytes.
14542 send_buffer_size=bytes
14544 Set UDP send buffer size, expressed in bytes.
14545 timeout=microseconds
14547 Set raise error timeouts for read, write and connect operations.
14548 Note that the SRT library has internal timeouts which can be
14549 controlled separately, the value set here is only a cap on those.
14550 tlpktdrop=1|0
14552 Too-late Packet Drop. When enabled on receiver, it skips missing
14553 packets that have not been delivered in time and delivers the
14554 following packets to the application when their time-to-play has
14555 come. It also sends a fake ACK to the sender. When enabled on
14556 sender and enabled on the receiving peer, the sender drops the
14557 older packets that have no chance of being delivered in time. It
14558 was automatically enabled in the sender if the receiver supports
14559 it.
14560 sndbuf=bytes
14562 Set send buffer size, expressed in bytes.
14563 rcvbuf=bytes
14565 Set receive buffer size, expressed in bytes.
14566 Receive buffer must not be greater than ffs.
14568 lossmaxttl=packets
14570 The value up to which the Reorder Tolerance may grow. When Reorder
14571 Tolerance is > 0, then packet loss report is delayed until that
14572 number of packets come in. Reorder Tolerance increases every time a
14573 "belated" packet has come, but it wasn't due to retransmission
14574 (that is, when UDP packets tend to come out of order), with the
14575 difference between the latest sequence and this packet's sequence,
14576 and not more than the value of this option. By default it's 0,
14577 which means that this mechanism is turned off, and the loss report
14578 is always sent immediately upon experiencing a "gap" in sequences.
14579 minversion
14581 The minimum SRT version that is required from the peer. A
14582 connection to a peer that does not satisfy the minimum version
14583 requirement will be rejected.
14584 The version format in hex is 0xXXYYZZ for x.y.z in human readable
14586 form.
14587 streamid=string
14589 A string limited to 512 characters that can be set on the socket
14590 prior to connecting. This stream ID will be able to be retrieved by
14591 the listener side from the socket that is returned from srt_accept
14592 and was connected by a socket with that set stream ID. SRT does not
14593 enforce any special interpretation of the contents of this string.
14594 This option doesnXt make sense in Rendezvous connection; the result
14595 might be that simply one side will override the value from the
14596 other side and itXs the matter of luck which one would win
14597 srt_streamid=string
14599 Alias for streamid to avoid conflict with ffmpeg command line
14600 option.
14601 smoother=live|file
14603 The type of Smoother used for the transmission for that socket,
14604 which is responsible for the transmission and congestion control.
14605 The Smoother type must be exactly the same on both connecting
14606 parties, otherwise the connection is rejected.
14607 messageapi=1|0
14609 When set, this socket uses the Message API, otherwise it uses
14610 Buffer API. Note that in live mode (see transtype) thereXs only
14611 message API available. In File mode you can chose to use one of two
14612 modes:
14613 Stream API (default, when this option is false). In this mode you
14615 may send as many data as you wish with one sending instruction, or
14616 even use dedicated functions that read directly from a file. The
14617 internal facility will take care of any speed and congestion
14618 control. When receiving, you can also receive as many data as
14619 desired, the data not extracted will be waiting for the next call.
14620 There is no boundary between data portions in the Stream mode.
14621 Message API. In this mode your single sending instruction passes
14623 exactly one piece of data that has boundaries (a message). Contrary
14624 to Live mode, this message may span across multiple UDP packets and
14625 the only size limitation is that it shall fit as a whole in the
14626 sending buffer. The receiver shall use as large buffer as necessary
14627 to receive the message, otherwise the message will not be given up.
14628 When the message is not complete (not all packets received or there
14629 was a packet loss) it will not be given up.
14630 transtype=live|file
14632 Sets the transmission type for the socket, in particular, setting
14633 this option sets multiple other parameters to their default values
14634 as required for a particular transmission type.
14635 live: Set options as for live transmission. In this mode, you
14637 should send by one sending instruction only so many data that fit
14638 in one UDP packet, and limited to the value defined first in
14639 payload_size (1316 is default in this mode). There is no speed
14640 control in this mode, only the bandwidth control, if configured, in
14641 order to not exceed the bandwidth with the overhead transmission
14642 (retransmitted and control packets).
14643 file: Set options as for non-live transmission. See messageapi for
14645 further explanations
14646 linger=seconds
14648 The number of seconds that the socket waits for unsent data when
14649 closing. Default is -1. -1 means auto (off with 0 seconds in live
14650 mode, on with 180 seconds in file mode). The range for this option
14651 is integers in the 0 - "INT_MAX".
14652 tsbpd=1|0
14654 When true, use Timestamp-based Packet Delivery mode. The default
14655 behavior depends on the transmission type: enabled in live mode,
14656 disabled in file mode.
14657 For more information see: <https://github.com/Haivision/srt>.
14659 srtp
14661 Secure Real-time Transport Protocol.
14662 The accepted options are:
14664 srtp_in_suite
14666 srtp_out_suite
14667 Select input and output encoding suites.
14668 Supported values:
14670 AES_CM_128_HMAC_SHA1_80
14672 SRTP_AES128_CM_HMAC_SHA1_80
14673 AES_CM_128_HMAC_SHA1_32
14674 SRTP_AES128_CM_HMAC_SHA1_32
14675 srtp_in_params
14676 srtp_out_params
14677 Set input and output encoding parameters, which are expressed by a
14678 base64-encoded representation of a binary block. The first 16 bytes
14679 of this binary block are used as master key, the following 14 bytes
14680 are used as master salt.
14681 subfile
14683 Virtually extract a segment of a file or another stream. The
14684 underlying stream must be seekable.
14685 Accepted options:
14687 start
14689 Start offset of the extracted segment, in bytes.
14690 end End offset of the extracted segment, in bytes. If set to 0,
14692 extract till end of file.
14693 Examples:
14695 Extract a chapter from a DVD VOB file (start and end sectors obtained
14697 externally and multiplied by 2048):
14698 subfile,,start,153391104,end,268142592,,:/media/dvd/VIDEO_TS/VTS_08_1.VOB
14700 Play an AVI file directly from a TAR archive:
14702 subfile,,start,183241728,end,366490624,,:archive.tar
14704 Play a MPEG-TS file from start offset till end:
14706 subfile,,start,32815239,end,0,,:video.ts
14708 tee
14710 Writes the output to multiple protocols. The individual outputs are
14711 separated by |
14712 tee:file://path/to/local/this.avi|file://path/to/local/that.avi
14714 tcp
14716 Transmission Control Protocol.
14717 The required syntax for a TCP url is:
14719 tcp://<hostname>:<port>[?<options>]
14721 options contains a list of &-separated options of the form key=val.
14723 The list of supported options follows.
14725 listen=2|1|0
14727 Listen for an incoming connection. 0 disables listen, 1 enables
14728 listen in single client mode, 2 enables listen in multi-client
14729 mode. Default value is 0.
14730 timeout=microseconds
14732 Set raise error timeout, expressed in microseconds.
14733 This option is only relevant in read mode: if no data arrived in
14735 more than this time interval, raise error.
14736 listen_timeout=milliseconds
14738 Set listen timeout, expressed in milliseconds.
14739 recv_buffer_size=bytes
14741 Set receive buffer size, expressed bytes.
14742 send_buffer_size=bytes
14744 Set send buffer size, expressed bytes.
14745 tcp_nodelay=1|0
14747 Set TCP_NODELAY to disable Nagle's algorithm. Default value is 0.
14748 Remark: Writing to the socket is currently not optimized to
14750 minimize system calls and reduces the efficiency / effect of
14751 TCP_NODELAY.
14752 tcp_mss=bytes
14754 Set maximum segment size for outgoing TCP packets, expressed in
14755 bytes.
14756 The following example shows how to setup a listening TCP connection
14758 with ffmpeg, which is then accessed with ffplay:
14759 ffmpeg -i <input> -f <format> tcp://<hostname>:<port>?listen
14761 ffplay tcp://<hostname>:<port>
14762 tls
14764 Transport Layer Security (TLS) / Secure Sockets Layer (SSL)
14765 The required syntax for a TLS/SSL url is:
14767 tls://<hostname>:<port>[?<options>]
14769 The following parameters can be set via command line options (or in
14771 code via "AVOption"s):
14772 ca_file, cafile=filename
14774 A file containing certificate authority (CA) root certificates to
14775 treat as trusted. If the linked TLS library contains a default this
14776 might not need to be specified for verification to work, but not
14777 all libraries and setups have defaults built in. The file must be
14778 in OpenSSL PEM format.
14779 tls_verify=1|0
14781 If enabled, try to verify the peer that we are communicating with.
14782 Note, if using OpenSSL, this currently only makes sure that the
14783 peer certificate is signed by one of the root certificates in the
14784 CA database, but it does not validate that the certificate actually
14785 matches the host name we are trying to connect to. (With other
14786 backends, the host name is validated as well.)
14787 This is disabled by default since it requires a CA database to be
14789 provided by the caller in many cases.
14790 cert_file, cert=filename
14792 A file containing a certificate to use in the handshake with the
14793 peer. (When operating as server, in listen mode, this is more
14794 often required by the peer, while client certificates only are
14795 mandated in certain setups.)
14796 key_file, key=filename
14798 A file containing the private key for the certificate.
14799 listen=1|0
14801 If enabled, listen for connections on the provided port, and assume
14802 the server role in the handshake instead of the client role.
14803 http_proxy
14805 The HTTP proxy to tunnel through, e.g. "http://example.com:1234".
14806 The proxy must support the CONNECT method.
14807 Example command lines:
14809 To create a TLS/SSL server that serves an input stream.
14811 ffmpeg -i <input> -f <format> tls://<hostname>:<port>?listen&cert=<server.crt>&key=<server.key>
14813 To play back a stream from the TLS/SSL server using ffplay:
14815 ffplay tls://<hostname>:<port>
14817 udp
14819 User Datagram Protocol.
14820 The required syntax for an UDP URL is:
14822 udp://<hostname>:<port>[?<options>]
14824 options contains a list of &-separated options of the form key=val.
14826 In case threading is enabled on the system, a circular buffer is used
14828 to store the incoming data, which allows one to reduce loss of data due
14829 to UDP socket buffer overruns. The fifo_size and overrun_nonfatal
14830 options are related to this buffer.
14831 The list of supported options follows.
14833 buffer_size=size
14835 Set the UDP maximum socket buffer size in bytes. This is used to
14836 set either the receive or send buffer size, depending on what the
14837 socket is used for. Default is 32 KB for output, 384 KB for input.
14838 See also fifo_size.
14839 bitrate=bitrate
14841 If set to nonzero, the output will have the specified constant
14842 bitrate if the input has enough packets to sustain it.
14843 burst_bits=bits
14845 When using bitrate this specifies the maximum number of bits in
14846 packet bursts.
14847 localport=port
14849 Override the local UDP port to bind with.
14850 localaddr=addr
14852 Local IP address of a network interface used for sending packets or
14853 joining multicast groups.
14854 pkt_size=size
14856 Set the size in bytes of UDP packets.
14857 reuse=1|0
14859 Explicitly allow or disallow reusing UDP sockets.
14860 ttl=ttl
14862 Set the time to live value (for multicast only).
14863 connect=1|0
14865 Initialize the UDP socket with "connect()". In this case, the
14866 destination address can't be changed with ff_udp_set_remote_url
14867 later. If the destination address isn't known at the start, this
14868 option can be specified in ff_udp_set_remote_url, too. This allows
14869 finding out the source address for the packets with getsockname,
14870 and makes writes return with AVERROR(ECONNREFUSED) if "destination
14871 unreachable" is received. For receiving, this gives the benefit of
14872 only receiving packets from the specified peer address/port.
14873 sources=address[,address]
14875 Only receive packets sent from the specified addresses. In case of
14876 multicast, also subscribe to multicast traffic coming from these
14877 addresses only.
14878 block=address[,address]
14880 Ignore packets sent from the specified addresses. In case of
14881 multicast, also exclude the source addresses in the multicast
14882 subscription.
14883 fifo_size=units
14885 Set the UDP receiving circular buffer size, expressed as a number
14886 of packets with size of 188 bytes. If not specified defaults to
14887 7*4096.
14888 overrun_nonfatal=1|0
14890 Survive in case of UDP receiving circular buffer overrun. Default
14891 value is 0.
14892 timeout=microseconds
14894 Set raise error timeout, expressed in microseconds.
14895 This option is only relevant in read mode: if no data arrived in
14897 more than this time interval, raise error.
14898 broadcast=1|0
14900 Explicitly allow or disallow UDP broadcasting.
14901 Note that broadcasting may not work properly on networks having a
14903 broadcast storm protection.
14904 Examples
14906 • Use ffmpeg to stream over UDP to a remote endpoint:
14908 ffmpeg -i <input> -f <format> udp://<hostname>:<port>
14910 • Use ffmpeg to stream in mpegts format over UDP using 188 sized UDP
14912 packets, using a large input buffer:
14913 ffmpeg -i <input> -f mpegts udp://<hostname>:<port>?pkt_size=188&buffer_size=65535
14915 • Use ffmpeg to receive over UDP from a remote endpoint:
14917 ffmpeg -i udp://[<multicast-address>]:<port> ...
14919 unix
14921 Unix local socket
14922 The required syntax for a Unix socket URL is:
14924 unix://<filepath>
14926 The following parameters can be set via command line options (or in
14928 code via "AVOption"s):
14929 timeout
14931 Timeout in ms.
14932 listen
14934 Create the Unix socket in listening mode.
14935 zmq
14937 ZeroMQ asynchronous messaging using the libzmq library.
14938 This library supports unicast streaming to multiple clients without
14940 relying on an external server.
14941 The required syntax for streaming or connecting to a stream is:
14943 zmq:tcp://ip-address:port
14945 Example: Create a localhost stream on port 5555:
14947 ffmpeg -re -i input -f mpegts zmq:tcp://127.0.0.1:5555
14949 Multiple clients may connect to the stream using:
14951 ffplay zmq:tcp://127.0.0.1:5555
14953 Streaming to multiple clients is implemented using a ZeroMQ Pub-Sub
14955 pattern. The server side binds to a port and publishes data. Clients
14956 connect to the server (via IP address/port) and subscribe to the
14957 stream. The order in which the server and client start generally does
14958 not matter.
14959 ffmpeg must be compiled with the --enable-libzmq option to support this
14961 protocol.
14962 Options can be set on the ffmpeg/ffplay command line. The following
14964 options are supported:
14965 pkt_size
14967 Forces the maximum packet size for sending/receiving data. The
14968 default value is 131,072 bytes. On the server side, this sets the
14969 maximum size of sent packets via ZeroMQ. On the clients, it sets an
14970 internal buffer size for receiving packets. Note that pkt_size on
14971 the clients should be equal to or greater than pkt_size on the
14972 server. Otherwise the received message may be truncated causing
14973 decoding errors.
14974
14976 The libavdevice library provides the same interface as libavformat.
14977 Namely, an input device is considered like a demuxer, and an output
14978 device like a muxer, and the interface and generic device options are
14979 the same provided by libavformat (see the ffmpeg-formats manual).
14980 In addition each input or output device may support so-called private
14982 options, which are specific for that component.
14983 Options may be set by specifying -option value in the FFmpeg tools, or
14985 by setting the value explicitly in the device "AVFormatContext" options
14986 or using the libavutil/opt.h API for programmatic use.
14987
14989 Input devices are configured elements in FFmpeg which enable accessing
14990 the data coming from a multimedia device attached to your system.
14991 When you configure your FFmpeg build, all the supported input devices
14993 are enabled by default. You can list all available ones using the
14994 configure option "--list-indevs".
14995 You can disable all the input devices using the configure option
14997 "--disable-indevs", and selectively enable an input device using the
14998 option "--enable-indev=INDEV", or you can disable a particular input
14999 device using the option "--disable-indev=INDEV".
15000 The option "-devices" of the ff* tools will display the list of
15002 supported input devices.
15003 A description of the currently available input devices follows.
15005 alsa
15007 ALSA (Advanced Linux Sound Architecture) input device.
15008 To enable this input device during configuration you need libasound
15010 installed on your system.
15011 This device allows capturing from an ALSA device. The name of the
15013 device to capture has to be an ALSA card identifier.
15014 An ALSA identifier has the syntax:
15016 hw:<CARD>[,<DEV>[,<SUBDEV>]]
15018 where the DEV and SUBDEV components are optional.
15020 The three arguments (in order: CARD,DEV,SUBDEV) specify card number or
15022 identifier, device number and subdevice number (-1 means any).
15023 To see the list of cards currently recognized by your system check the
15025 files /proc/asound/cards and /proc/asound/devices.
15026 For example to capture with ffmpeg from an ALSA device with card id 0,
15028 you may run the command:
15029 ffmpeg -f alsa -i hw:0 alsaout.wav
15031 For more information see:
15033 <http://www.alsa-project.org/alsa-doc/alsa-lib/pcm.html>
15034 Options
15036 sample_rate
15038 Set the sample rate in Hz. Default is 48000.
15039 channels
15041 Set the number of channels. Default is 2.
15042 android_camera
15044 Android camera input device.
15045 This input devices uses the Android Camera2 NDK API which is available
15047 on devices with API level 24+. The availability of android_camera is
15048 autodetected during configuration.
15049 This device allows capturing from all cameras on an Android device,
15051 which are integrated into the Camera2 NDK API.
15052 The available cameras are enumerated internally and can be selected
15054 with the camera_index parameter. The input file string is discarded.
15055 Generally the back facing camera has index 0 while the front facing
15057 camera has index 1.
15058 Options
15060 video_size
15062 Set the video size given as a string such as 640x480 or hd720.
15063 Falls back to the first available configuration reported by Android
15064 if requested video size is not available or by default.
15065 framerate
15067 Set the video framerate. Falls back to the first available
15068 configuration reported by Android if requested framerate is not
15069 available or by default (-1).
15070 camera_index
15072 Set the index of the camera to use. Default is 0.
15073 input_queue_size
15075 Set the maximum number of frames to buffer. Default is 5.
15076 avfoundation
15078 AVFoundation input device.
15079 AVFoundation is the currently recommended framework by Apple for
15081 streamgrabbing on OSX >= 10.7 as well as on iOS.
15082 The input filename has to be given in the following syntax:
15084 -i "[[VIDEO]:[AUDIO]]"
15086 The first entry selects the video input while the latter selects the
15088 audio input. The stream has to be specified by the device name or the
15089 device index as shown by the device list. Alternatively, the video
15090 and/or audio input device can be chosen by index using the
15091 B<-video_device_index E<lt>INDEXE<gt>>
15093 and/or
15095 B<-audio_device_index E<lt>INDEXE<gt>>
15097 , overriding any device name or index given in the input filename.
15099 All available devices can be enumerated by using -list_devices true,
15101 listing all device names and corresponding indices.
15102 There are two device name aliases:
15104 "default"
15106 Select the AVFoundation default device of the corresponding type.
15107 "none"
15109 Do not record the corresponding media type. This is equivalent to
15110 specifying an empty device name or index.
15111 Options
15113 AVFoundation supports the following options:
15115 -list_devices <TRUE|FALSE>
15117 If set to true, a list of all available input devices is given
15118 showing all device names and indices.
15119 -video_device_index <INDEX>
15121 Specify the video device by its index. Overrides anything given in
15122 the input filename.
15123 -audio_device_index <INDEX>
15125 Specify the audio device by its index. Overrides anything given in
15126 the input filename.
15127 -pixel_format <FORMAT>
15129 Request the video device to use a specific pixel format. If the
15130 specified format is not supported, a list of available formats is
15131 given and the first one in this list is used instead. Available
15132 pixel formats are: "monob, rgb555be, rgb555le, rgb565be, rgb565le,
15133 rgb24, bgr24, 0rgb, bgr0, 0bgr, rgb0,
15134 bgr48be, uyvy422, yuva444p, yuva444p16le, yuv444p, yuv422p16,
15135 yuv422p10, yuv444p10,
15136 yuv420p, nv12, yuyv422, gray"
15137 -framerate
15139 Set the grabbing frame rate. Default is "ntsc", corresponding to a
15140 frame rate of "30000/1001".
15141 -video_size
15143 Set the video frame size.
15144 -capture_cursor
15146 Capture the mouse pointer. Default is 0.
15147 -capture_mouse_clicks
15149 Capture the screen mouse clicks. Default is 0.
15150 -capture_raw_data
15152 Capture the raw device data. Default is 0. Using this option may
15153 result in receiving the underlying data delivered to the
15154 AVFoundation framework. E.g. for muxed devices that sends raw DV
15155 data to the framework (like tape-based camcorders), setting this
15156 option to false results in extracted video frames captured in the
15157 designated pixel format only. Setting this option to true results
15158 in receiving the raw DV stream untouched.
15159 Examples
15161 • Print the list of AVFoundation supported devices and exit:
15163 $ ffmpeg -f avfoundation -list_devices true -i ""
15165 • Record video from video device 0 and audio from audio device 0 into
15167 out.avi:
15168 $ ffmpeg -f avfoundation -i "0:0" out.avi
15170 • Record video from video device 2 and audio from audio device 1 into
15172 out.avi:
15173 $ ffmpeg -f avfoundation -video_device_index 2 -i ":1" out.avi
15175 • Record video from the system default video device using the pixel
15177 format bgr0 and do not record any audio into out.avi:
15178 $ ffmpeg -f avfoundation -pixel_format bgr0 -i "default:none" out.avi
15180 • Record raw DV data from a suitable input device and write the
15182 output into out.dv:
15183 $ ffmpeg -f avfoundation -capture_raw_data true -i "zr100:none" out.dv
15185 bktr
15187 BSD video input device.
15188 Options
15190 framerate
15192 Set the frame rate.
15193 video_size
15195 Set the video frame size. Default is "vga".
15196 standard
15198 Available values are:
15199 pal
15201 ntsc
15202 secam
15203 paln
15204 palm
15205 ntscj
15206 decklink
15208 The decklink input device provides capture capabilities for Blackmagic
15209 DeckLink devices.
15210 To enable this input device, you need the Blackmagic DeckLink SDK and
15212 you need to configure with the appropriate "--extra-cflags" and
15213 "--extra-ldflags". On Windows, you need to run the IDL files through
15214 widl.
15215 DeckLink is very picky about the formats it supports. Pixel format of
15217 the input can be set with raw_format. Framerate and video size must be
15218 determined for your device with -list_formats 1. Audio sample rate is
15219 always 48 kHz and the number of channels can be 2, 8 or 16. Note that
15220 all audio channels are bundled in one single audio track.
15221 Options
15223 list_devices
15225 If set to true, print a list of devices and exit. Defaults to
15226 false. This option is deprecated, please use the "-sources" option
15227 of ffmpeg to list the available input devices.
15228 list_formats
15230 If set to true, print a list of supported formats and exit.
15231 Defaults to false.
15232 format_code <FourCC>
15234 This sets the input video format to the format given by the FourCC.
15235 To see the supported values of your device(s) use list_formats.
15236 Note that there is a FourCC 'pal ' that can also be used as pal (3
15237 letters). Default behavior is autodetection of the input video
15238 format, if the hardware supports it.
15239 raw_format
15241 Set the pixel format of the captured video. Available values are:
15242 auto
15244 This is the default which means 8-bit YUV 422 or 8-bit ARGB if
15245 format autodetection is used, 8-bit YUV 422 otherwise.
15246 uyvy422
15248 8-bit YUV 422.
15249 yuv422p10
15251 10-bit YUV 422.
15252 argb
15254 8-bit RGB.
15255 bgra
15257 8-bit RGB.
15258 rgb10
15260 10-bit RGB.
15261 teletext_lines
15263 If set to nonzero, an additional teletext stream will be captured
15264 from the vertical ancillary data. Both SD PAL (576i) and HD (1080i
15265 or 1080p) sources are supported. In case of HD sources, OP47
15266 packets are decoded.
15267 This option is a bitmask of the SD PAL VBI lines captured,
15269 specifically lines 6 to 22, and lines 318 to 335. Line 6 is the LSB
15270 in the mask. Selected lines which do not contain teletext
15271 information will be ignored. You can use the special all constant
15272 to select all possible lines, or standard to skip lines 6, 318 and
15273 319, which are not compatible with all receivers.
15274 For SD sources, ffmpeg needs to be compiled with
15276 "--enable-libzvbi". For HD sources, on older (pre-4K) DeckLink card
15277 models you have to capture in 10 bit mode.
15278 channels
15280 Defines number of audio channels to capture. Must be 2, 8 or 16.
15281 Defaults to 2.
15282 duplex_mode
15284 Sets the decklink device duplex/profile mode. Must be unset, half,
15285 full, one_sub_device_full, one_sub_device_half,
15286 two_sub_device_full, four_sub_device_half Defaults to unset.
15287 Note: DeckLink SDK 11.0 have replaced the duplex property by a
15289 profile property. For the DeckLink Duo 2 and DeckLink Quad 2, a
15290 profile is shared between any 2 sub-devices that utilize the same
15291 connectors. For the DeckLink 8K Pro, a profile is shared between
15292 all 4 sub-devices. So DeckLink 8K Pro support four profiles.
15293 Valid profile modes for DeckLink 8K Pro(with DeckLink SDK >= 11.0):
15295 one_sub_device_full, one_sub_device_half, two_sub_device_full,
15296 four_sub_device_half
15297 Valid profile modes for DeckLink Quad 2 and DeckLink Duo 2: half,
15299 full
15300 timecode_format
15302 Timecode type to include in the frame and video stream metadata.
15303 Must be none, rp188vitc, rp188vitc2, rp188ltc, rp188hfr, rp188any,
15304 vitc, vitc2, or serial. Defaults to none (not included).
15305 In order to properly support 50/60 fps timecodes, the ordering of
15307 the queried timecode types for rp188any is HFR, VITC1, VITC2 and
15308 LTC for >30 fps content. Note that this is slightly different to
15309 the ordering used by the DeckLink API, which is HFR, VITC1, LTC,
15310 VITC2.
15311 video_input
15313 Sets the video input source. Must be unset, sdi, hdmi, optical_sdi,
15314 component, composite or s_video. Defaults to unset.
15315 audio_input
15317 Sets the audio input source. Must be unset, embedded, aes_ebu,
15318 analog, analog_xlr, analog_rca or microphone. Defaults to unset.
15319 video_pts
15321 Sets the video packet timestamp source. Must be video, audio,
15322 reference, wallclock or abs_wallclock. Defaults to video.
15323 audio_pts
15325 Sets the audio packet timestamp source. Must be video, audio,
15326 reference, wallclock or abs_wallclock. Defaults to audio.
15327 draw_bars
15329 If set to true, color bars are drawn in the event of a signal loss.
15330 Defaults to true.
15331 queue_size
15333 Sets maximum input buffer size in bytes. If the buffering reaches
15334 this value, incoming frames will be dropped. Defaults to
15335 1073741824.
15336 audio_depth
15338 Sets the audio sample bit depth. Must be 16 or 32. Defaults to 16.
15339 decklink_copyts
15341 If set to true, timestamps are forwarded as they are without
15342 removing the initial offset. Defaults to false.
15343 timestamp_align
15345 Capture start time alignment in seconds. If set to nonzero, input
15346 frames are dropped till the system timestamp aligns with configured
15347 value. Alignment difference of up to one frame duration is
15348 tolerated. This is useful for maintaining input synchronization
15349 across N different hardware devices deployed for 'N-way'
15350 redundancy. The system time of different hardware devices should be
15351 synchronized with protocols such as NTP or PTP, before using this
15352 option. Note that this method is not foolproof. In some border
15353 cases input synchronization may not happen due to thread scheduling
15354 jitters in the OS. Either sync could go wrong by 1 frame or in a
15355 rarer case timestamp_align seconds. Defaults to 0.
15356 wait_for_tc (bool)
15358 Drop frames till a frame with timecode is received. Sometimes
15359 serial timecode isn't received with the first input frame. If that
15360 happens, the stored stream timecode will be inaccurate. If this
15361 option is set to true, input frames are dropped till a frame with
15362 timecode is received. Option timecode_format must be specified.
15363 Defaults to false.
15364 enable_klv(bool)
15366 If set to true, extracts KLV data from VANC and outputs KLV
15367 packets. KLV VANC packets are joined based on MID and PSC fields
15368 and aggregated into one KLV packet. Defaults to false.
15369 Examples
15371 • List input devices:
15373 ffmpeg -sources decklink
15375 • List supported formats:
15377 ffmpeg -f decklink -list_formats 1 -i 'Intensity Pro'
15379 • Capture video clip at 1080i50:
15381 ffmpeg -format_code Hi50 -f decklink -i 'Intensity Pro' -c:a copy -c:v copy output.avi
15383 • Capture video clip at 1080i50 10 bit:
15385 ffmpeg -raw_format yuv422p10 -format_code Hi50 -f decklink -i 'UltraStudio Mini Recorder' -c:a copy -c:v copy output.avi
15387 • Capture video clip at 1080i50 with 16 audio channels:
15389 ffmpeg -channels 16 -format_code Hi50 -f decklink -i 'UltraStudio Mini Recorder' -c:a copy -c:v copy output.avi
15391 dshow
15393 Windows DirectShow input device.
15394 DirectShow support is enabled when FFmpeg is built with the mingw-w64
15396 project. Currently only audio and video devices are supported.
15397 Multiple devices may be opened as separate inputs, but they may also be
15399 opened on the same input, which should improve synchronism between
15400 them.
15401 The input name should be in the format:
15403 <TYPE>=<NAME>[:<TYPE>=<NAME>]
15405 where TYPE can be either audio or video, and NAME is the device's name
15407 or alternative name..
15408 Options
15410 If no options are specified, the device's defaults are used. If the
15412 device does not support the requested options, it will fail to open.
15413 video_size
15415 Set the video size in the captured video.
15416 framerate
15418 Set the frame rate in the captured video.
15419 sample_rate
15421 Set the sample rate (in Hz) of the captured audio.
15422 sample_size
15424 Set the sample size (in bits) of the captured audio.
15425 channels
15427 Set the number of channels in the captured audio.
15428 list_devices
15430 If set to true, print a list of devices and exit.
15431 list_options
15433 If set to true, print a list of selected device's options and exit.
15434 video_device_number
15436 Set video device number for devices with the same name (starts at
15437 0, defaults to 0).
15438 audio_device_number
15440 Set audio device number for devices with the same name (starts at
15441 0, defaults to 0).
15442 pixel_format
15444 Select pixel format to be used by DirectShow. This may only be set
15445 when the video codec is not set or set to rawvideo.
15446 audio_buffer_size
15448 Set audio device buffer size in milliseconds (which can directly
15449 impact latency, depending on the device). Defaults to using the
15450 audio device's default buffer size (typically some multiple of
15451 500ms). Setting this value too low can degrade performance. See
15452 also
15453 <http://msdn.microsoft.com/en-us/library/windows/desktop/dd377582(v=vs.85).aspx>
15454 video_pin_name
15456 Select video capture pin to use by name or alternative name.
15457 audio_pin_name
15459 Select audio capture pin to use by name or alternative name.
15460 crossbar_video_input_pin_number
15462 Select video input pin number for crossbar device. This will be
15463 routed to the crossbar device's Video Decoder output pin. Note
15464 that changing this value can affect future invocations (sets a new
15465 default) until system reboot occurs.
15466 crossbar_audio_input_pin_number
15468 Select audio input pin number for crossbar device. This will be
15469 routed to the crossbar device's Audio Decoder output pin. Note
15470 that changing this value can affect future invocations (sets a new
15471 default) until system reboot occurs.
15472 show_video_device_dialog
15474 If set to true, before capture starts, popup a display dialog to
15475 the end user, allowing them to change video filter properties and
15476 configurations manually. Note that for crossbar devices, adjusting
15477 values in this dialog may be needed at times to toggle between PAL
15478 (25 fps) and NTSC (29.97) input frame rates, sizes, interlacing,
15479 etc. Changing these values can enable different scan rates/frame
15480 rates and avoiding green bars at the bottom, flickering scan lines,
15481 etc. Note that with some devices, changing these properties can
15482 also affect future invocations (sets new defaults) until system
15483 reboot occurs.
15484 show_audio_device_dialog
15486 If set to true, before capture starts, popup a display dialog to
15487 the end user, allowing them to change audio filter properties and
15488 configurations manually.
15489 show_video_crossbar_connection_dialog
15491 If set to true, before capture starts, popup a display dialog to
15492 the end user, allowing them to manually modify crossbar pin
15493 routings, when it opens a video device.
15494 show_audio_crossbar_connection_dialog
15496 If set to true, before capture starts, popup a display dialog to
15497 the end user, allowing them to manually modify crossbar pin
15498 routings, when it opens an audio device.
15499 show_analog_tv_tuner_dialog
15501 If set to true, before capture starts, popup a display dialog to
15502 the end user, allowing them to manually modify TV channels and
15503 frequencies.
15504 show_analog_tv_tuner_audio_dialog
15506 If set to true, before capture starts, popup a display dialog to
15507 the end user, allowing them to manually modify TV audio (like mono
15508 vs. stereo, Language A,B or C).
15509 audio_device_load
15511 Load an audio capture filter device from file instead of searching
15512 it by name. It may load additional parameters too, if the filter
15513 supports the serialization of its properties to. To use this an
15514 audio capture source has to be specified, but it can be anything
15515 even fake one.
15516 audio_device_save
15518 Save the currently used audio capture filter device and its
15519 parameters (if the filter supports it) to a file. If a file with
15520 the same name exists it will be overwritten.
15521 video_device_load
15523 Load a video capture filter device from file instead of searching
15524 it by name. It may load additional parameters too, if the filter
15525 supports the serialization of its properties to. To use this a
15526 video capture source has to be specified, but it can be anything
15527 even fake one.
15528 video_device_save
15530 Save the currently used video capture filter device and its
15531 parameters (if the filter supports it) to a file. If a file with
15532 the same name exists it will be overwritten.
15533 use_video_device_timestamps
15535 If set to false, the timestamp for video frames will be derived
15536 from the wallclock instead of the timestamp provided by the capture
15537 device. This allows working around devices that provide unreliable
15538 timestamps.
15539 Examples
15541 • Print the list of DirectShow supported devices and exit:
15543 $ ffmpeg -list_devices true -f dshow -i dummy
15545 • Open video device Camera:
15547 $ ffmpeg -f dshow -i video="Camera"
15549 • Open second video device with name Camera:
15551 $ ffmpeg -f dshow -video_device_number 1 -i video="Camera"
15553 • Open video device Camera and audio device Microphone:
15555 $ ffmpeg -f dshow -i video="Camera":audio="Microphone"
15557 • Print the list of supported options in selected device and exit:
15559 $ ffmpeg -list_options true -f dshow -i video="Camera"
15561 • Specify pin names to capture by name or alternative name, specify
15563 alternative device name:
15564 $ ffmpeg -f dshow -audio_pin_name "Audio Out" -video_pin_name 2 -i video=video="@device_pnp_\\?\pci#ven_1a0a&dev_6200&subsys_62021461&rev_01#4&e2c7dd6&0&00e1#{65e8773d-8f56-11d0-a3b9-00a0c9223196}\{ca465100-deb0-4d59-818f-8c477184adf6}":audio="Microphone"
15566 • Configure a crossbar device, specifying crossbar pins, allow user
15568 to adjust video capture properties at startup:
15569 $ ffmpeg -f dshow -show_video_device_dialog true -crossbar_video_input_pin_number 0
15571 -crossbar_audio_input_pin_number 3 -i video="AVerMedia BDA Analog Capture":audio="AVerMedia BDA Analog Capture"
15572 fbdev
15574 Linux framebuffer input device.
15575 The Linux framebuffer is a graphic hardware-independent abstraction
15577 layer to show graphics on a computer monitor, typically on the console.
15578 It is accessed through a file device node, usually /dev/fb0.
15579 For more detailed information read the file
15581 Documentation/fb/framebuffer.txt included in the Linux source tree.
15582 See also <http://linux-fbdev.sourceforge.net/>, and fbset(1).
15584 To record from the framebuffer device /dev/fb0 with ffmpeg:
15586 ffmpeg -f fbdev -framerate 10 -i /dev/fb0 out.avi
15588 You can take a single screenshot image with the command:
15590 ffmpeg -f fbdev -framerate 1 -i /dev/fb0 -frames:v 1 screenshot.jpeg
15592 Options
15594 framerate
15596 Set the frame rate. Default is 25.
15597 gdigrab
15599 Win32 GDI-based screen capture device.
15600 This device allows you to capture a region of the display on Windows.
15602 There are two options for the input filename:
15604 desktop
15606 or
15608 title=<window_title>
15610 The first option will capture the entire desktop, or a fixed region of
15612 the desktop. The second option will instead capture the contents of a
15613 single window, regardless of its position on the screen.
15614 For example, to grab the entire desktop using ffmpeg:
15616 ffmpeg -f gdigrab -framerate 6 -i desktop out.mpg
15618 Grab a 640x480 region at position "10,20":
15620 ffmpeg -f gdigrab -framerate 6 -offset_x 10 -offset_y 20 -video_size vga -i desktop out.mpg
15622 Grab the contents of the window named "Calculator"
15624 ffmpeg -f gdigrab -framerate 6 -i title=Calculator out.mpg
15626 Options
15628 draw_mouse
15630 Specify whether to draw the mouse pointer. Use the value 0 to not
15631 draw the pointer. Default value is 1.
15632 framerate
15634 Set the grabbing frame rate. Default value is "ntsc", corresponding
15635 to a frame rate of "30000/1001".
15636 show_region
15638 Show grabbed region on screen.
15639 If show_region is specified with 1, then the grabbing region will
15641 be indicated on screen. With this option, it is easy to know what
15642 is being grabbed if only a portion of the screen is grabbed.
15643 Note that show_region is incompatible with grabbing the contents of
15645 a single window.
15646 For example:
15648 ffmpeg -f gdigrab -show_region 1 -framerate 6 -video_size cif -offset_x 10 -offset_y 20 -i desktop out.mpg
15650 video_size
15652 Set the video frame size. The default is to capture the full screen
15653 if desktop is selected, or the full window size if
15654 title=window_title is selected.
15655 offset_x
15657 When capturing a region with video_size, set the distance from the
15658 left edge of the screen or desktop.
15659 Note that the offset calculation is from the top left corner of the
15661 primary monitor on Windows. If you have a monitor positioned to the
15662 left of your primary monitor, you will need to use a negative
15663 offset_x value to move the region to that monitor.
15664 offset_y
15666 When capturing a region with video_size, set the distance from the
15667 top edge of the screen or desktop.
15668 Note that the offset calculation is from the top left corner of the
15670 primary monitor on Windows. If you have a monitor positioned above
15671 your primary monitor, you will need to use a negative offset_y
15672 value to move the region to that monitor.
15673 iec61883
15675 FireWire DV/HDV input device using libiec61883.
15676 To enable this input device, you need libiec61883, libraw1394 and
15678 libavc1394 installed on your system. Use the configure option
15679 "--enable-libiec61883" to compile with the device enabled.
15680 The iec61883 capture device supports capturing from a video device
15682 connected via IEEE1394 (FireWire), using libiec61883 and the new Linux
15683 FireWire stack (juju). This is the default DV/HDV input method in Linux
15684 Kernel 2.6.37 and later, since the old FireWire stack was removed.
15685 Specify the FireWire port to be used as input file, or "auto" to choose
15687 the first port connected.
15688 Options
15690 dvtype
15692 Override autodetection of DV/HDV. This should only be used if auto
15693 detection does not work, or if usage of a different device type
15694 should be prohibited. Treating a DV device as HDV (or vice versa)
15695 will not work and result in undefined behavior. The values auto,
15696 dv and hdv are supported.
15697 dvbuffer
15699 Set maximum size of buffer for incoming data, in frames. For DV,
15700 this is an exact value. For HDV, it is not frame exact, since HDV
15701 does not have a fixed frame size.
15702 dvguid
15704 Select the capture device by specifying its GUID. Capturing will
15705 only be performed from the specified device and fails if no device
15706 with the given GUID is found. This is useful to select the input if
15707 multiple devices are connected at the same time. Look at
15708 /sys/bus/firewire/devices to find out the GUIDs.
15709 Examples
15711 • Grab and show the input of a FireWire DV/HDV device.
15713 ffplay -f iec61883 -i auto
15715 • Grab and record the input of a FireWire DV/HDV device, using a
15717 packet buffer of 100000 packets if the source is HDV.
15718 ffmpeg -f iec61883 -i auto -dvbuffer 100000 out.mpg
15720 jack
15722 JACK input device.
15723 To enable this input device during configuration you need libjack
15725 installed on your system.
15726 A JACK input device creates one or more JACK writable clients, one for
15728 each audio channel, with name client_name:input_N, where client_name is
15729 the name provided by the application, and N is a number which
15730 identifies the channel. Each writable client will send the acquired
15731 data to the FFmpeg input device.
15732 Once you have created one or more JACK readable clients, you need to
15734 connect them to one or more JACK writable clients.
15735 To connect or disconnect JACK clients you can use the jack_connect and
15737 jack_disconnect programs, or do it through a graphical interface, for
15738 example with qjackctl.
15739 To list the JACK clients and their properties you can invoke the
15741 command jack_lsp.
15742 Follows an example which shows how to capture a JACK readable client
15744 with ffmpeg.
15745 # Create a JACK writable client with name "ffmpeg".
15747 $ ffmpeg -f jack -i ffmpeg -y out.wav
15748 # Start the sample jack_metro readable client.
15750 $ jack_metro -b 120 -d 0.2 -f 4000
15751 # List the current JACK clients.
15753 $ jack_lsp -c
15754 system:capture_1
15755 system:capture_2
15756 system:playback_1
15757 system:playback_2
15758 ffmpeg:input_1
15759 metro:120_bpm
15760 # Connect metro to the ffmpeg writable client.
15762 $ jack_connect metro:120_bpm ffmpeg:input_1
15763 For more information read: <http://jackaudio.org/>
15765 Options
15767 channels
15769 Set the number of channels. Default is 2.
15770 kmsgrab
15772 KMS video input device.
15773 Captures the KMS scanout framebuffer associated with a specified CRTC
15775 or plane as a DRM object that can be passed to other hardware
15776 functions.
15777 Requires either DRM master or CAP_SYS_ADMIN to run.
15779 If you don't understand what all of that means, you probably don't want
15781 this. Look at x11grab instead.
15782 Options
15784 device
15786 DRM device to capture on. Defaults to /dev/dri/card0.
15787 format
15789 Pixel format of the framebuffer. This can be autodetected if you
15790 are running Linux 5.7 or later, but needs to be provided for
15791 earlier versions. Defaults to bgr0, which is the most common
15792 format used by the Linux console and Xorg X server.
15793 format_modifier
15795 Format modifier to signal on output frames. This is necessary to
15796 import correctly into some APIs. It can be autodetected if you are
15797 running Linux 5.7 or later, but will need to be provided explicitly
15798 when needed in earlier versions. See the libdrm documentation for
15799 possible values.
15800 crtc_id
15802 KMS CRTC ID to define the capture source. The first active plane
15803 on the given CRTC will be used.
15804 plane_id
15806 KMS plane ID to define the capture source. Defaults to the first
15807 active plane found if neither crtc_id nor plane_id are specified.
15808 framerate
15810 Framerate to capture at. This is not synchronised to any page
15811 flipping or framebuffer changes - it just defines the interval at
15812 which the framebuffer is sampled. Sampling faster than the
15813 framebuffer update rate will generate independent frames with the
15814 same content. Defaults to 30.
15815 Examples
15817 • Capture from the first active plane, download the result to normal
15819 frames and encode. This will only work if the framebuffer is both
15820 linear and mappable - if not, the result may be scrambled or fail
15821 to download.
15822 ffmpeg -f kmsgrab -i - -vf 'hwdownload,format=bgr0' output.mp4
15824 • Capture from CRTC ID 42 at 60fps, map the result to VAAPI, convert
15826 to NV12 and encode as H.264.
15827 ffmpeg -crtc_id 42 -framerate 60 -f kmsgrab -i - -vf 'hwmap=derive_device=vaapi,scale_vaapi=w=1920:h=1080:format=nv12' -c:v h264_vaapi output.mp4
15829 • To capture only part of a plane the output can be cropped - this
15831 can be used to capture a single window, as long as it has a known
15832 absolute position and size. For example, to capture and encode the
15833 middle quarter of a 1920x1080 plane:
15834 ffmpeg -f kmsgrab -i - -vf 'hwmap=derive_device=vaapi,crop=960:540:480:270,scale_vaapi=960:540:nv12' -c:v h264_vaapi output.mp4
15836 lavfi
15838 Libavfilter input virtual device.
15839 This input device reads data from the open output pads of a libavfilter
15841 filtergraph.
15842 For each filtergraph open output, the input device will create a
15844 corresponding stream which is mapped to the generated output. Currently
15845 only video data is supported. The filtergraph is specified through the
15846 option graph.
15847 Options
15849 graph
15851 Specify the filtergraph to use as input. Each video open output
15852 must be labelled by a unique string of the form "outN", where N is
15853 a number starting from 0 corresponding to the mapped input stream
15854 generated by the device. The first unlabelled output is
15855 automatically assigned to the "out0" label, but all the others need
15856 to be specified explicitly.
15857 The suffix "+subcc" can be appended to the output label to create
15859 an extra stream with the closed captions packets attached to that
15860 output (experimental; only for EIA-608 / CEA-708 for now). The
15861 subcc streams are created after all the normal streams, in the
15862 order of the corresponding stream. For example, if there is
15863 "out19+subcc", "out7+subcc" and up to "out42", the stream #43 is
15864 subcc for stream #7 and stream #44 is subcc for stream #19.
15865 If not specified defaults to the filename specified for the input
15867 device.
15868 graph_file
15870 Set the filename of the filtergraph to be read and sent to the
15871 other filters. Syntax of the filtergraph is the same as the one
15872 specified by the option graph.
15873 dumpgraph
15875 Dump graph to stderr.
15876 Examples
15878 • Create a color video stream and play it back with ffplay:
15880 ffplay -f lavfi -graph "color=c=pink [out0]" dummy
15882 • As the previous example, but use filename for specifying the graph
15884 description, and omit the "out0" label:
15885 ffplay -f lavfi color=c=pink
15887 • Create three different video test filtered sources and play them:
15889 ffplay -f lavfi -graph "testsrc [out0]; testsrc,hflip [out1]; testsrc,negate [out2]" test3
15891 • Read an audio stream from a file using the amovie source and play
15893 it back with ffplay:
15894 ffplay -f lavfi "amovie=test.wav"
15896 • Read an audio stream and a video stream and play it back with
15898 ffplay:
15899 ffplay -f lavfi "movie=test.avi[out0];amovie=test.wav[out1]"
15901 • Dump decoded frames to images and closed captions to a file
15903 (experimental):
15904 ffmpeg -f lavfi -i "movie=test.ts[out0+subcc]" -map v frame%08d.png -map s -c copy -f rawvideo subcc.bin
15906 libcdio
15908 Audio-CD input device based on libcdio.
15909 To enable this input device during configuration you need libcdio
15911 installed on your system. It requires the configure option
15912 "--enable-libcdio".
15913 This device allows playing and grabbing from an Audio-CD.
15915 For example to copy with ffmpeg the entire Audio-CD in /dev/sr0, you
15917 may run the command:
15918 ffmpeg -f libcdio -i /dev/sr0 cd.wav
15920 Options
15922 speed
15924 Set drive reading speed. Default value is 0.
15925 The speed is specified CD-ROM speed units. The speed is set through
15927 the libcdio "cdio_cddap_speed_set" function. On many CD-ROM drives,
15928 specifying a value too large will result in using the fastest
15929 speed.
15930 paranoia_mode
15932 Set paranoia recovery mode flags. It accepts one of the following
15933 values:
15934 disable
15936 verify
15937 overlap
15938 neverskip
15939 full
15940 Default value is disable.
15942 For more information about the available recovery modes, consult
15944 the paranoia project documentation.
15945 libdc1394
15947 IIDC1394 input device, based on libdc1394 and libraw1394.
15948 Requires the configure option "--enable-libdc1394".
15950 Options
15952 framerate
15954 Set the frame rate. Default is "ntsc", corresponding to a frame
15955 rate of "30000/1001".
15956 pixel_format
15958 Select the pixel format. Default is "uyvy422".
15959 video_size
15961 Set the video size given as a string such as "640x480" or "hd720".
15962 Default is "qvga".
15963 openal
15965 The OpenAL input device provides audio capture on all systems with a
15966 working OpenAL 1.1 implementation.
15967 To enable this input device during configuration, you need OpenAL
15969 headers and libraries installed on your system, and need to configure
15970 FFmpeg with "--enable-openal".
15971 OpenAL headers and libraries should be provided as part of your OpenAL
15973 implementation, or as an additional download (an SDK). Depending on
15974 your installation you may need to specify additional flags via the
15975 "--extra-cflags" and "--extra-ldflags" for allowing the build system to
15976 locate the OpenAL headers and libraries.
15977 An incomplete list of OpenAL implementations follows:
15979 Creative
15981 The official Windows implementation, providing hardware
15982 acceleration with supported devices and software fallback. See
15983 <http://openal.org/>.
15984 OpenAL Soft
15986 Portable, open source (LGPL) software implementation. Includes
15987 backends for the most common sound APIs on the Windows, Linux,
15988 Solaris, and BSD operating systems. See
15989 <http://kcat.strangesoft.net/openal.html>.
15990 Apple
15992 OpenAL is part of Core Audio, the official Mac OS X Audio
15993 interface. See
15994 <http://developer.apple.com/technologies/mac/audio-and-video.html>
15995 This device allows one to capture from an audio input device handled
15997 through OpenAL.
15998 You need to specify the name of the device to capture in the provided
16000 filename. If the empty string is provided, the device will
16001 automatically select the default device. You can get the list of the
16002 supported devices by using the option list_devices.
16003 Options
16005 channels
16007 Set the number of channels in the captured audio. Only the values 1
16008 (monaural) and 2 (stereo) are currently supported. Defaults to 2.
16009 sample_size
16011 Set the sample size (in bits) of the captured audio. Only the
16012 values 8 and 16 are currently supported. Defaults to 16.
16013 sample_rate
16015 Set the sample rate (in Hz) of the captured audio. Defaults to
16016 44.1k.
16017 list_devices
16019 If set to true, print a list of devices and exit. Defaults to
16020 false.
16021 Examples
16023 Print the list of OpenAL supported devices and exit:
16025 $ ffmpeg -list_devices true -f openal -i dummy out.ogg
16027 Capture from the OpenAL device DR-BT101 via PulseAudio:
16029 $ ffmpeg -f openal -i 'DR-BT101 via PulseAudio' out.ogg
16031 Capture from the default device (note the empty string '' as filename):
16033 $ ffmpeg -f openal -i '' out.ogg
16035 Capture from two devices simultaneously, writing to two different
16037 files, within the same ffmpeg command:
16038 $ ffmpeg -f openal -i 'DR-BT101 via PulseAudio' out1.ogg -f openal -i 'ALSA Default' out2.ogg
16040 Note: not all OpenAL implementations support multiple simultaneous
16042 capture - try the latest OpenAL Soft if the above does not work.
16043 oss
16045 Open Sound System input device.
16046 The filename to provide to the input device is the device node
16048 representing the OSS input device, and is usually set to /dev/dsp.
16049 For example to grab from /dev/dsp using ffmpeg use the command:
16051 ffmpeg -f oss -i /dev/dsp /tmp/oss.wav
16053 For more information about OSS see:
16055 <http://manuals.opensound.com/usersguide/dsp.html>
16056 Options
16058 sample_rate
16060 Set the sample rate in Hz. Default is 48000.
16061 channels
16063 Set the number of channels. Default is 2.
16064 pulse
16066 PulseAudio input device.
16067 To enable this output device you need to configure FFmpeg with
16069 "--enable-libpulse".
16070 The filename to provide to the input device is a source device or the
16072 string "default"
16073 To list the PulseAudio source devices and their properties you can
16075 invoke the command pactl list sources.
16076 More information about PulseAudio can be found on
16078 <http://www.pulseaudio.org>.
16079 Options
16081 server
16083 Connect to a specific PulseAudio server, specified by an IP
16084 address. Default server is used when not provided.
16085 name
16087 Specify the application name PulseAudio will use when showing
16088 active clients, by default it is the "LIBAVFORMAT_IDENT" string.
16089 stream_name
16091 Specify the stream name PulseAudio will use when showing active
16092 streams, by default it is "record".
16093 sample_rate
16095 Specify the samplerate in Hz, by default 48kHz is used.
16096 channels
16098 Specify the channels in use, by default 2 (stereo) is set.
16099 frame_size
16101 Specify the number of bytes per frame, by default it is set to
16102 1024.
16103 fragment_size
16105 Specify the minimal buffering fragment in PulseAudio, it will
16106 affect the audio latency. By default it is unset.
16107 wallclock
16109 Set the initial PTS using the current time. Default is 1.
16110 Examples
16112 Record a stream from default device:
16114 ffmpeg -f pulse -i default /tmp/pulse.wav
16116 sndio
16118 sndio input device.
16119 To enable this input device during configuration you need libsndio
16121 installed on your system.
16122 The filename to provide to the input device is the device node
16124 representing the sndio input device, and is usually set to /dev/audio0.
16125 For example to grab from /dev/audio0 using ffmpeg use the command:
16127 ffmpeg -f sndio -i /dev/audio0 /tmp/oss.wav
16129 Options
16131 sample_rate
16133 Set the sample rate in Hz. Default is 48000.
16134 channels
16136 Set the number of channels. Default is 2.
16137 video4linux2, v4l2
16139 Video4Linux2 input video device.
16140 "v4l2" can be used as alias for "video4linux2".
16142 If FFmpeg is built with v4l-utils support (by using the
16144 "--enable-libv4l2" configure option), it is possible to use it with the
16145 "-use_libv4l2" input device option.
16146 The name of the device to grab is a file device node, usually Linux
16148 systems tend to automatically create such nodes when the device (e.g.
16149 an USB webcam) is plugged into the system, and has a name of the kind
16150 /dev/videoN, where N is a number associated to the device.
16151 Video4Linux2 devices usually support a limited set of widthxheight
16153 sizes and frame rates. You can check which are supported using
16154 -list_formats all for Video4Linux2 devices. Some devices, like TV
16155 cards, support one or more standards. It is possible to list all the
16156 supported standards using -list_standards all.
16157 The time base for the timestamps is 1 microsecond. Depending on the
16159 kernel version and configuration, the timestamps may be derived from
16160 the real time clock (origin at the Unix Epoch) or the monotonic clock
16161 (origin usually at boot time, unaffected by NTP or manual changes to
16162 the clock). The -timestamps abs or -ts abs option can be used to force
16163 conversion into the real time clock.
16164 Some usage examples of the video4linux2 device with ffmpeg and ffplay:
16166 • List supported formats for a video4linux2 device:
16168 ffplay -f video4linux2 -list_formats all /dev/video0
16170 • Grab and show the input of a video4linux2 device:
16172 ffplay -f video4linux2 -framerate 30 -video_size hd720 /dev/video0
16174 • Grab and record the input of a video4linux2 device, leave the frame
16176 rate and size as previously set:
16177 ffmpeg -f video4linux2 -input_format mjpeg -i /dev/video0 out.mpeg
16179 For more information about Video4Linux, check <http://linuxtv.org/>.
16181 Options
16183 standard
16185 Set the standard. Must be the name of a supported standard. To get
16186 a list of the supported standards, use the list_standards option.
16187 channel
16189 Set the input channel number. Default to -1, which means using the
16190 previously selected channel.
16191 video_size
16193 Set the video frame size. The argument must be a string in the form
16194 WIDTHxHEIGHT or a valid size abbreviation.
16195 pixel_format
16197 Select the pixel format (only valid for raw video input).
16198 input_format
16200 Set the preferred pixel format (for raw video) or a codec name.
16201 This option allows one to select the input format, when several are
16202 available.
16203 framerate
16205 Set the preferred video frame rate.
16206 list_formats
16208 List available formats (supported pixel formats, codecs, and frame
16209 sizes) and exit.
16210 Available values are:
16212 all Show all available (compressed and non-compressed) formats.
16214 raw Show only raw video (non-compressed) formats.
16216 compressed
16218 Show only compressed formats.
16219 list_standards
16221 List supported standards and exit.
16222 Available values are:
16224 all Show all supported standards.
16226 timestamps, ts
16228 Set type of timestamps for grabbed frames.
16229 Available values are:
16231 default
16233 Use timestamps from the kernel.
16234 abs Use absolute timestamps (wall clock).
16236 mono2abs
16238 Force conversion from monotonic to absolute timestamps.
16239 Default value is "default".
16241 use_libv4l2
16243 Use libv4l2 (v4l-utils) conversion functions. Default is 0.
16244 vfwcap
16246 VfW (Video for Windows) capture input device.
16247 The filename passed as input is the capture driver number, ranging from
16249 0 to 9. You may use "list" as filename to print a list of drivers. Any
16250 other filename will be interpreted as device number 0.
16251 Options
16253 video_size
16255 Set the video frame size.
16256 framerate
16258 Set the grabbing frame rate. Default value is "ntsc", corresponding
16259 to a frame rate of "30000/1001".
16260 x11grab
16262 X11 video input device.
16263 To enable this input device during configuration you need libxcb
16265 installed on your system. It will be automatically detected during
16266 configuration.
16267 This device allows one to capture a region of an X11 display.
16269 The filename passed as input has the syntax:
16271 [<hostname>]:<display_number>.<screen_number>[+<x_offset>,<y_offset>]
16273 hostname:display_number.screen_number specifies the X11 display name of
16275 the screen to grab from. hostname can be omitted, and defaults to
16276 "localhost". The environment variable DISPLAY contains the default
16277 display name.
16278 x_offset and y_offset specify the offsets of the grabbed area with
16280 respect to the top-left border of the X11 screen. They default to 0.
16281 Check the X11 documentation (e.g. man X) for more detailed information.
16283 Use the xdpyinfo program for getting basic information about the
16285 properties of your X11 display (e.g. grep for "name" or "dimensions").
16286 For example to grab from :0.0 using ffmpeg:
16288 ffmpeg -f x11grab -framerate 25 -video_size cif -i :0.0 out.mpg
16290 Grab at position "10,20":
16292 ffmpeg -f x11grab -framerate 25 -video_size cif -i :0.0+10,20 out.mpg
16294 Options
16296 select_region
16298 Specify whether to select the grabbing area graphically using the
16299 pointer. A value of 1 prompts the user to select the grabbing area
16300 graphically by clicking and dragging. A single click with no
16301 dragging will select the whole screen. A region with zero width or
16302 height will also select the whole screen. This option overwrites
16303 the video_size, grab_x, and grab_y options. Default value is 0.
16304 draw_mouse
16306 Specify whether to draw the mouse pointer. A value of 0 specifies
16307 not to draw the pointer. Default value is 1.
16308 follow_mouse
16310 Make the grabbed area follow the mouse. The argument can be
16311 "centered" or a number of pixels PIXELS.
16312 When it is specified with "centered", the grabbing region follows
16314 the mouse pointer and keeps the pointer at the center of region;
16315 otherwise, the region follows only when the mouse pointer reaches
16316 within PIXELS (greater than zero) to the edge of region.
16317 For example:
16319 ffmpeg -f x11grab -follow_mouse centered -framerate 25 -video_size cif -i :0.0 out.mpg
16321 To follow only when the mouse pointer reaches within 100 pixels to
16323 edge:
16324 ffmpeg -f x11grab -follow_mouse 100 -framerate 25 -video_size cif -i :0.0 out.mpg
16326 framerate
16328 Set the grabbing frame rate. Default value is "ntsc", corresponding
16329 to a frame rate of "30000/1001".
16330 show_region
16332 Show grabbed region on screen.
16333 If show_region is specified with 1, then the grabbing region will
16335 be indicated on screen. With this option, it is easy to know what
16336 is being grabbed if only a portion of the screen is grabbed.
16337 region_border
16339 Set the region border thickness if -show_region 1 is used. Range
16340 is 1 to 128 and default is 3 (XCB-based x11grab only).
16341 For example:
16343 ffmpeg -f x11grab -show_region 1 -framerate 25 -video_size cif -i :0.0+10,20 out.mpg
16345 With follow_mouse:
16347 ffmpeg -f x11grab -follow_mouse centered -show_region 1 -framerate 25 -video_size cif -i :0.0 out.mpg
16349 window_id
16351 Grab this window, instead of the whole screen. Default value is 0,
16352 which maps to the whole screen (root window).
16353 The id of a window can be found using the xwininfo program,
16355 possibly with options -tree and -root.
16356 If the window is later enlarged, the new area is not recorded.
16358 Video ends when the window is closed, unmapped (i.e., iconified) or
16359 shrunk beyond the video size (which defaults to the initial window
16360 size).
16361 This option disables options follow_mouse and select_region.
16363 video_size
16365 Set the video frame size. Default is the full desktop or window.
16366 grab_x
16368 grab_y
16369 Set the grabbing region coordinates. They are expressed as offset
16370 from the top left corner of the X11 window and correspond to the
16371 x_offset and y_offset parameters in the device name. The default
16372 value for both options is 0.
16373
16375 Output devices are configured elements in FFmpeg that can write
16376 multimedia data to an output device attached to your system.
16377 When you configure your FFmpeg build, all the supported output devices
16379 are enabled by default. You can list all available ones using the
16380 configure option "--list-outdevs".
16381 You can disable all the output devices using the configure option
16383 "--disable-outdevs", and selectively enable an output device using the
16384 option "--enable-outdev=OUTDEV", or you can disable a particular input
16385 device using the option "--disable-outdev=OUTDEV".
16386 The option "-devices" of the ff* tools will display the list of enabled
16388 output devices.
16389 A description of the currently available output devices follows.
16391 alsa
16393 ALSA (Advanced Linux Sound Architecture) output device.
16394 Examples
16396 • Play a file on default ALSA device:
16398 ffmpeg -i INPUT -f alsa default
16400 • Play a file on soundcard 1, audio device 7:
16402 ffmpeg -i INPUT -f alsa hw:1,7
16404 AudioToolbox
16406 AudioToolbox output device.
16407 Allows native output to CoreAudio devices on OSX.
16409 The output filename can be empty (or "-") to refer to the default
16411 system output device or a number that refers to the device index as
16412 shown using: "-list_devices true".
16413 Alternatively, the audio input device can be chosen by index using the
16415 B<-audio_device_index E<lt>INDEXE<gt>>
16417 , overriding any device name or index given in the input filename.
16419 All available devices can be enumerated by using -list_devices true,
16421 listing all device names, UIDs and corresponding indices.
16422 Options
16424 AudioToolbox supports the following options:
16426 -audio_device_index <INDEX>
16428 Specify the audio device by its index. Overrides anything given in
16429 the output filename.
16430 Examples
16432 • Print the list of supported devices and output a sine wave to the
16434 default device:
16435 $ ffmpeg -f lavfi -i sine=r=44100 -f audiotoolbox -list_devices true -
16437 • Output a sine wave to the device with the index 2, overriding any
16439 output filename:
16440 $ ffmpeg -f lavfi -i sine=r=44100 -f audiotoolbox -audio_device_index 2 -
16442 caca
16444 CACA output device.
16445 This output device allows one to show a video stream in CACA window.
16447 Only one CACA window is allowed per application, so you can have only
16448 one instance of this output device in an application.
16449 To enable this output device you need to configure FFmpeg with
16451 "--enable-libcaca". libcaca is a graphics library that outputs text
16452 instead of pixels.
16453 For more information about libcaca, check:
16455 <http://caca.zoy.org/wiki/libcaca>
16456 Options
16458 window_title
16460 Set the CACA window title, if not specified default to the filename
16461 specified for the output device.
16462 window_size
16464 Set the CACA window size, can be a string of the form widthxheight
16465 or a video size abbreviation. If not specified it defaults to the
16466 size of the input video.
16467 driver
16469 Set display driver.
16470 algorithm
16472 Set dithering algorithm. Dithering is necessary because the picture
16473 being rendered has usually far more colours than the available
16474 palette. The accepted values are listed with "-list_dither
16475 algorithms".
16476 antialias
16478 Set antialias method. Antialiasing smoothens the rendered image and
16479 avoids the commonly seen staircase effect. The accepted values are
16480 listed with "-list_dither antialiases".
16481 charset
16483 Set which characters are going to be used when rendering text. The
16484 accepted values are listed with "-list_dither charsets".
16485 color
16487 Set color to be used when rendering text. The accepted values are
16488 listed with "-list_dither colors".
16489 list_drivers
16491 If set to true, print a list of available drivers and exit.
16492 list_dither
16494 List available dither options related to the argument. The
16495 argument must be one of "algorithms", "antialiases", "charsets",
16496 "colors".
16497 Examples
16499 • The following command shows the ffmpeg output is an CACA window,
16501 forcing its size to 80x25:
16502 ffmpeg -i INPUT -c:v rawvideo -pix_fmt rgb24 -window_size 80x25 -f caca -
16504 • Show the list of available drivers and exit:
16506 ffmpeg -i INPUT -pix_fmt rgb24 -f caca -list_drivers true -
16508 • Show the list of available dither colors and exit:
16510 ffmpeg -i INPUT -pix_fmt rgb24 -f caca -list_dither colors -
16512 decklink
16514 The decklink output device provides playback capabilities for
16515 Blackmagic DeckLink devices.
16516 To enable this output device, you need the Blackmagic DeckLink SDK and
16518 you need to configure with the appropriate "--extra-cflags" and
16519 "--extra-ldflags". On Windows, you need to run the IDL files through
16520 widl.
16521 DeckLink is very picky about the formats it supports. Pixel format is
16523 always uyvy422, framerate, field order and video size must be
16524 determined for your device with -list_formats 1. Audio sample rate is
16525 always 48 kHz.
16526 Options
16528 list_devices
16530 If set to true, print a list of devices and exit. Defaults to
16531 false. This option is deprecated, please use the "-sinks" option of
16532 ffmpeg to list the available output devices.
16533 list_formats
16535 If set to true, print a list of supported formats and exit.
16536 Defaults to false.
16537 preroll
16539 Amount of time to preroll video in seconds. Defaults to 0.5.
16540 duplex_mode
16542 Sets the decklink device duplex/profile mode. Must be unset, half,
16543 full, one_sub_device_full, one_sub_device_half,
16544 two_sub_device_full, four_sub_device_half Defaults to unset.
16545 Note: DeckLink SDK 11.0 have replaced the duplex property by a
16547 profile property. For the DeckLink Duo 2 and DeckLink Quad 2, a
16548 profile is shared between any 2 sub-devices that utilize the same
16549 connectors. For the DeckLink 8K Pro, a profile is shared between
16550 all 4 sub-devices. So DeckLink 8K Pro support four profiles.
16551 Valid profile modes for DeckLink 8K Pro(with DeckLink SDK >= 11.0):
16553 one_sub_device_full, one_sub_device_half, two_sub_device_full,
16554 four_sub_device_half
16555 Valid profile modes for DeckLink Quad 2 and DeckLink Duo 2: half,
16557 full
16558 timing_offset
16560 Sets the genlock timing pixel offset on the used output. Defaults
16561 to unset.
16562 link
16564 Sets the SDI video link configuration on the used output. Must be
16565 unset, single link SDI, dual link SDI or quad link SDI. Defaults
16566 to unset.
16567 sqd Enable Square Division Quad Split mode for Quad-link SDI output.
16569 Must be unset, true or false. Defaults to unset.
16570 level_a
16572 Enable SMPTE Level A mode on the used output. Must be unset, true
16573 or false. Defaults to unset.
16574 Examples
16576 • List output devices:
16578 ffmpeg -sinks decklink
16580 • List supported formats:
16582 ffmpeg -i test.avi -f decklink -list_formats 1 'DeckLink Mini Monitor'
16584 • Play video clip:
16586 ffmpeg -i test.avi -f decklink -pix_fmt uyvy422 'DeckLink Mini Monitor'
16588 • Play video clip with non-standard framerate or video size:
16590 ffmpeg -i test.avi -f decklink -pix_fmt uyvy422 -s 720x486 -r 24000/1001 'DeckLink Mini Monitor'
16592 fbdev
16594 Linux framebuffer output device.
16595 The Linux framebuffer is a graphic hardware-independent abstraction
16597 layer to show graphics on a computer monitor, typically on the console.
16598 It is accessed through a file device node, usually /dev/fb0.
16599 For more detailed information read the file
16601 Documentation/fb/framebuffer.txt included in the Linux source tree.
16602 Options
16604 xoffset
16606 yoffset
16607 Set x/y coordinate of top left corner. Default is 0.
16608 Examples
16610 Play a file on framebuffer device /dev/fb0. Required pixel format
16612 depends on current framebuffer settings.
16613 ffmpeg -re -i INPUT -c:v rawvideo -pix_fmt bgra -f fbdev /dev/fb0
16615 See also <http://linux-fbdev.sourceforge.net/>, and fbset(1).
16617 opengl
16619 OpenGL output device.
16620 To enable this output device you need to configure FFmpeg with
16622 "--enable-opengl".
16623 This output device allows one to render to OpenGL context. Context may
16625 be provided by application or default SDL window is created.
16626 When device renders to external context, application must implement
16628 handlers for following messages: "AV_DEV_TO_APP_CREATE_WINDOW_BUFFER" -
16629 create OpenGL context on current thread.
16630 "AV_DEV_TO_APP_PREPARE_WINDOW_BUFFER" - make OpenGL context current.
16631 "AV_DEV_TO_APP_DISPLAY_WINDOW_BUFFER" - swap buffers.
16632 "AV_DEV_TO_APP_DESTROY_WINDOW_BUFFER" - destroy OpenGL context.
16633 Application is also required to inform a device about current
16634 resolution by sending "AV_APP_TO_DEV_WINDOW_SIZE" message.
16635 Options
16637 background
16639 Set background color. Black is a default.
16640 no_window
16642 Disables default SDL window when set to non-zero value.
16643 Application must provide OpenGL context and both "window_size_cb"
16644 and "window_swap_buffers_cb" callbacks when set.
16645 window_title
16647 Set the SDL window title, if not specified default to the filename
16648 specified for the output device. Ignored when no_window is set.
16649 window_size
16651 Set preferred window size, can be a string of the form widthxheight
16652 or a video size abbreviation. If not specified it defaults to the
16653 size of the input video, downscaled according to the aspect ratio.
16654 Mostly usable when no_window is not set.
16655 Examples
16657 Play a file on SDL window using OpenGL rendering:
16659 ffmpeg -i INPUT -f opengl "window title"
16661 oss
16663 OSS (Open Sound System) output device.
16664 pulse
16666 PulseAudio output device.
16667 To enable this output device you need to configure FFmpeg with
16669 "--enable-libpulse".
16670 More information about PulseAudio can be found on
16672 <http://www.pulseaudio.org>
16673 Options
16675 server
16677 Connect to a specific PulseAudio server, specified by an IP
16678 address. Default server is used when not provided.
16679 name
16681 Specify the application name PulseAudio will use when showing
16682 active clients, by default it is the "LIBAVFORMAT_IDENT" string.
16683 stream_name
16685 Specify the stream name PulseAudio will use when showing active
16686 streams, by default it is set to the specified output name.
16687 device
16689 Specify the device to use. Default device is used when not
16690 provided. List of output devices can be obtained with command
16691 pactl list sinks.
16692 buffer_size
16694 buffer_duration
16695 Control the size and duration of the PulseAudio buffer. A small
16696 buffer gives more control, but requires more frequent updates.
16697 buffer_size specifies size in bytes while buffer_duration specifies
16699 duration in milliseconds.
16700 When both options are provided then the highest value is used
16702 (duration is recalculated to bytes using stream parameters). If
16703 they are set to 0 (which is default), the device will use the
16704 default PulseAudio duration value. By default PulseAudio set buffer
16705 duration to around 2 seconds.
16706 prebuf
16708 Specify pre-buffering size in bytes. The server does not start with
16709 playback before at least prebuf bytes are available in the buffer.
16710 By default this option is initialized to the same value as
16711 buffer_size or buffer_duration (whichever is bigger).
16712 minreq
16714 Specify minimum request size in bytes. The server does not request
16715 less than minreq bytes from the client, instead waits until the
16716 buffer is free enough to request more bytes at once. It is
16717 recommended to not set this option, which will initialize this to a
16718 value that is deemed sensible by the server.
16719 Examples
16721 Play a file on default device on default server:
16723 ffmpeg -i INPUT -f pulse "stream name"
16725 sdl
16727 SDL (Simple DirectMedia Layer) output device.
16728 "sdl2" can be used as alias for "sdl".
16730 This output device allows one to show a video stream in an SDL window.
16732 Only one SDL window is allowed per application, so you can have only
16733 one instance of this output device in an application.
16734 To enable this output device you need libsdl installed on your system
16736 when configuring your build.
16737 For more information about SDL, check: <http://www.libsdl.org/>
16739 Options
16741 window_title
16743 Set the SDL window title, if not specified default to the filename
16744 specified for the output device.
16745 icon_title
16747 Set the name of the iconified SDL window, if not specified it is
16748 set to the same value of window_title.
16749 window_size
16751 Set the SDL window size, can be a string of the form widthxheight
16752 or a video size abbreviation. If not specified it defaults to the
16753 size of the input video, downscaled according to the aspect ratio.
16754 window_x
16756 window_y
16757 Set the position of the window on the screen.
16758 window_fullscreen
16760 Set fullscreen mode when non-zero value is provided. Default value
16761 is zero.
16762 window_enable_quit
16764 Enable quit action (using window button or keyboard key) when non-
16765 zero value is provided. Default value is 1 (enable quit action)
16766 Interactive commands
16768 The window created by the device can be controlled through the
16770 following interactive commands.
16771 q, ESC
16773 Quit the device immediately.
16774 Examples
16776 The following command shows the ffmpeg output is an SDL window, forcing
16778 its size to the qcif format:
16779 ffmpeg -i INPUT -c:v rawvideo -pix_fmt yuv420p -window_size qcif -f sdl "SDL output"
16781 sndio
16783 sndio audio output device.
16784 v4l2
16786 Video4Linux2 output device.
16787 xv
16789 XV (XVideo) output device.
16790 This output device allows one to show a video stream in a X Window
16792 System window.
16793 Options
16795 display_name
16797 Specify the hardware display name, which determines the display and
16798 communications domain to be used.
16799 The display name or DISPLAY environment variable can be a string in
16801 the format hostname[:number[.screen_number]].
16802 hostname specifies the name of the host machine on which the
16804 display is physically attached. number specifies the number of the
16805 display server on that host machine. screen_number specifies the
16806 screen to be used on that server.
16807 If unspecified, it defaults to the value of the DISPLAY environment
16809 variable.
16810 For example, "dual-headed:0.1" would specify screen 1 of display 0
16812 on the machine named ``dual-headed''.
16813 Check the X11 specification for more detailed information about the
16815 display name format.
16816 window_id
16818 When set to non-zero value then device doesn't create new window,
16819 but uses existing one with provided window_id. By default this
16820 options is set to zero and device creates its own window.
16821 window_size
16823 Set the created window size, can be a string of the form
16824 widthxheight or a video size abbreviation. If not specified it
16825 defaults to the size of the input video. Ignored when window_id is
16826 set.
16827 window_x
16829 window_y
16830 Set the X and Y window offsets for the created window. They are
16831 both set to 0 by default. The values may be ignored by the window
16832 manager. Ignored when window_id is set.
16833 window_title
16835 Set the window title, if not specified default to the filename
16836 specified for the output device. Ignored when window_id is set.
16837 For more information about XVideo see <http://www.x.org/>.
16839 Examples
16841 • Decode, display and encode video input with ffmpeg at the same
16843 time:
16844 ffmpeg -i INPUT OUTPUT -f xv display
16846 • Decode and display the input video to multiple X11 windows:
16848 ffmpeg -i INPUT -f xv normal -vf negate -f xv negated
16850
16852 The audio resampler supports the following named options.
16853 Options may be set by specifying -option value in the FFmpeg tools,
16855 option=value for the aresample filter, by setting the value explicitly
16856 in the "SwrContext" options or using the libavutil/opt.h API for
16857 programmatic use.
16858 ich, in_channel_count
16860 Set the number of input channels. Default value is 0. Setting this
16861 value is not mandatory if the corresponding channel layout
16862 in_channel_layout is set.
16863 och, out_channel_count
16865 Set the number of output channels. Default value is 0. Setting this
16866 value is not mandatory if the corresponding channel layout
16867 out_channel_layout is set.
16868 uch, used_channel_count
16870 Set the number of used input channels. Default value is 0. This
16871 option is only used for special remapping.
16872 isr, in_sample_rate
16874 Set the input sample rate. Default value is 0.
16875 osr, out_sample_rate
16877 Set the output sample rate. Default value is 0.
16878 isf, in_sample_fmt
16880 Specify the input sample format. It is set by default to "none".
16881 osf, out_sample_fmt
16883 Specify the output sample format. It is set by default to "none".
16884 tsf, internal_sample_fmt
16886 Set the internal sample format. Default value is "none". This will
16887 automatically be chosen when it is not explicitly set.
16888 icl, in_channel_layout
16890 ocl, out_channel_layout
16891 Set the input/output channel layout.
16892 See the Channel Layout section in the ffmpeg-utils(1) manual for
16894 the required syntax.
16895 clev, center_mix_level
16897 Set the center mix level. It is a value expressed in deciBel, and
16898 must be in the interval [-32,32].
16899 slev, surround_mix_level
16901 Set the surround mix level. It is a value expressed in deciBel, and
16902 must be in the interval [-32,32].
16903 lfe_mix_level
16905 Set LFE mix into non LFE level. It is used when there is a LFE
16906 input but no LFE output. It is a value expressed in deciBel, and
16907 must be in the interval [-32,32].
16908 rmvol, rematrix_volume
16910 Set rematrix volume. Default value is 1.0.
16911 rematrix_maxval
16913 Set maximum output value for rematrixing. This can be used to
16914 prevent clipping vs. preventing volume reduction. A value of 1.0
16915 prevents clipping.
16916 flags, swr_flags
16918 Set flags used by the converter. Default value is 0.
16919 It supports the following individual flags:
16921 res force resampling, this flag forces resampling to be used even
16923 when the input and output sample rates match.
16924 dither_scale
16926 Set the dither scale. Default value is 1.
16927 dither_method
16929 Set dither method. Default value is 0.
16930 Supported values:
16932 rectangular
16934 select rectangular dither
16935 triangular
16937 select triangular dither
16938 triangular_hp
16940 select triangular dither with high pass
16941 lipshitz
16943 select Lipshitz noise shaping dither.
16944 shibata
16946 select Shibata noise shaping dither.
16947 low_shibata
16949 select low Shibata noise shaping dither.
16950 high_shibata
16952 select high Shibata noise shaping dither.
16953 f_weighted
16955 select f-weighted noise shaping dither
16956 modified_e_weighted
16958 select modified-e-weighted noise shaping dither
16959 improved_e_weighted
16961 select improved-e-weighted noise shaping dither
16962 resampler
16964 Set resampling engine. Default value is swr.
16965 Supported values:
16967 swr select the native SW Resampler; filter options precision and
16969 cheby are not applicable in this case.
16970 soxr
16972 select the SoX Resampler (where available); compensation, and
16973 filter options filter_size, phase_shift, exact_rational,
16974 filter_type & kaiser_beta, are not applicable in this case.
16975 filter_size
16977 For swr only, set resampling filter size, default value is 32.
16978 phase_shift
16980 For swr only, set resampling phase shift, default value is 10, and
16981 must be in the interval [0,30].
16982 linear_interp
16984 Use linear interpolation when enabled (the default). Disable it if
16985 you want to preserve speed instead of quality when exact_rational
16986 fails.
16987 exact_rational
16989 For swr only, when enabled, try to use exact phase_count based on
16990 input and output sample rate. However, if it is larger than "1 <<
16991 phase_shift", the phase_count will be "1 << phase_shift" as
16992 fallback. Default is enabled.
16993 cutoff
16995 Set cutoff frequency (swr: 6dB point; soxr: 0dB point) ratio; must
16996 be a float value between 0 and 1. Default value is 0.97 with swr,
16997 and 0.91 with soxr (which, with a sample-rate of 44100, preserves
16998 the entire audio band to 20kHz).
16999 precision
17001 For soxr only, the precision in bits to which the resampled signal
17002 will be calculated. The default value of 20 (which, with suitable
17003 dithering, is appropriate for a destination bit-depth of 16) gives
17004 SoX's 'High Quality'; a value of 28 gives SoX's 'Very High
17005 Quality'.
17006 cheby
17008 For soxr only, selects passband rolloff none (Chebyshev) & higher-
17009 precision approximation for 'irrational' ratios. Default value is
17010 0.
17011 async
17013 For swr only, simple 1 parameter audio sync to timestamps using
17014 stretching, squeezing, filling and trimming. Setting this to 1 will
17015 enable filling and trimming, larger values represent the maximum
17016 amount in samples that the data may be stretched or squeezed for
17017 each second. Default value is 0, thus no compensation is applied
17018 to make the samples match the audio timestamps.
17019 first_pts
17021 For swr only, assume the first pts should be this value. The time
17022 unit is 1 / sample rate. This allows for padding/trimming at the
17023 start of stream. By default, no assumption is made about the first
17024 frame's expected pts, so no padding or trimming is done. For
17025 example, this could be set to 0 to pad the beginning with silence
17026 if an audio stream starts after the video stream or to trim any
17027 samples with a negative pts due to encoder delay.
17028 min_comp
17030 For swr only, set the minimum difference between timestamps and
17031 audio data (in seconds) to trigger stretching/squeezing/filling or
17032 trimming of the data to make it match the timestamps. The default
17033 is that stretching/squeezing/filling and trimming is disabled
17034 (min_comp = "FLT_MAX").
17035 min_hard_comp
17037 For swr only, set the minimum difference between timestamps and
17038 audio data (in seconds) to trigger adding/dropping samples to make
17039 it match the timestamps. This option effectively is a threshold to
17040 select between hard (trim/fill) and soft (squeeze/stretch)
17041 compensation. Note that all compensation is by default disabled
17042 through min_comp. The default is 0.1.
17043 comp_duration
17045 For swr only, set duration (in seconds) over which data is
17046 stretched/squeezed to make it match the timestamps. Must be a non-
17047 negative double float value, default value is 1.0.
17048 max_soft_comp
17050 For swr only, set maximum factor by which data is
17051 stretched/squeezed to make it match the timestamps. Must be a non-
17052 negative double float value, default value is 0.
17053 matrix_encoding
17055 Select matrixed stereo encoding.
17056 It accepts the following values:
17058 none
17060 select none
17061 dolby
17063 select Dolby
17064 dplii
17066 select Dolby Pro Logic II
17067 Default value is "none".
17069 filter_type
17071 For swr only, select resampling filter type. This only affects
17072 resampling operations.
17073 It accepts the following values:
17075 cubic
17077 select cubic
17078 blackman_nuttall
17080 select Blackman Nuttall windowed sinc
17081 kaiser
17083 select Kaiser windowed sinc
17084 kaiser_beta
17086 For swr only, set Kaiser window beta value. Must be a double float
17087 value in the interval [2,16], default value is 9.
17088 output_sample_bits
17090 For swr only, set number of used output sample bits for dithering.
17091 Must be an integer in the interval [0,64], default value is 0,
17092 which means it's not used.
17093
17095 The video scaler supports the following named options.
17096 Options may be set by specifying -option value in the FFmpeg tools,
17098 with a few API-only exceptions noted below. For programmatic use, they
17099 can be set explicitly in the "SwsContext" options or through the
17100 libavutil/opt.h API.
17101 sws_flags
17103 Set the scaler flags. This is also used to set the scaling
17104 algorithm. Only a single algorithm should be selected. Default
17105 value is bicubic.
17106 It accepts the following values:
17108 fast_bilinear
17110 Select fast bilinear scaling algorithm.
17111 bilinear
17113 Select bilinear scaling algorithm.
17114 bicubic
17116 Select bicubic scaling algorithm.
17117 experimental
17119 Select experimental scaling algorithm.
17120 neighbor
17122 Select nearest neighbor rescaling algorithm.
17123 area
17125 Select averaging area rescaling algorithm.
17126 bicublin
17128 Select bicubic scaling algorithm for the luma component,
17129 bilinear for chroma components.
17130 gauss
17132 Select Gaussian rescaling algorithm.
17133 sinc
17135 Select sinc rescaling algorithm.
17136 lanczos
17138 Select Lanczos rescaling algorithm. The default width (alpha)
17139 is 3 and can be changed by setting "param0".
17140 spline
17142 Select natural bicubic spline rescaling algorithm.
17143 print_info
17145 Enable printing/debug logging.
17146 accurate_rnd
17148 Enable accurate rounding.
17149 full_chroma_int
17151 Enable full chroma interpolation.
17152 full_chroma_inp
17154 Select full chroma input.
17155 bitexact
17157 Enable bitexact output.
17158 srcw (API only)
17160 Set source width.
17161 srch (API only)
17163 Set source height.
17164 dstw (API only)
17166 Set destination width.
17167 dsth (API only)
17169 Set destination height.
17170 src_format (API only)
17172 Set source pixel format (must be expressed as an integer).
17173 dst_format (API only)
17175 Set destination pixel format (must be expressed as an integer).
17176 src_range (boolean)
17178 If value is set to 1, indicates source is full range. Default value
17179 is 0, which indicates source is limited range.
17180 dst_range (boolean)
17182 If value is set to 1, enable full range for destination. Default
17183 value is 0, which enables limited range.
17184 param0, param1
17186 Set scaling algorithm parameters. The specified values are specific
17187 of some scaling algorithms and ignored by others. The specified
17188 values are floating point number values.
17189 sws_dither
17191 Set the dithering algorithm. Accepts one of the following values.
17192 Default value is auto.
17193 auto
17195 automatic choice
17196 none
17198 no dithering
17199 bayer
17201 bayer dither
17202 ed error diffusion dither
17204 a_dither
17206 arithmetic dither, based using addition
17207 x_dither
17209 arithmetic dither, based using xor (more random/less apparent
17210 patterning that a_dither).
17211 alphablend
17213 Set the alpha blending to use when the input has alpha but the
17214 output does not. Default value is none.
17215 uniform_color
17217 Blend onto a uniform background color
17218 checkerboard
17220 Blend onto a checkerboard
17221 none
17223 No blending
17224
17226 Filtering in FFmpeg is enabled through the libavfilter library.
17227 In libavfilter, a filter can have multiple inputs and multiple outputs.
17229 To illustrate the sorts of things that are possible, we consider the
17230 following filtergraph.
17231 [main]
17233 input --> split ---------------------> overlay --> output
17234 | ^
17235 |[tmp] [flip]|
17236 +-----> crop --> vflip -------+
17237 This filtergraph splits the input stream in two streams, then sends one
17239 stream through the crop filter and the vflip filter, before merging it
17240 back with the other stream by overlaying it on top. You can use the
17241 following command to achieve this:
17242 ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
17244 The result will be that the top half of the video is mirrored onto the
17246 bottom half of the output video.
17247 Filters in the same linear chain are separated by commas, and distinct
17249 linear chains of filters are separated by semicolons. In our example,
17250 crop,vflip are in one linear chain, split and overlay are separately in
17251 another. The points where the linear chains join are labelled by names
17252 enclosed in square brackets. In the example, the split filter generates
17253 two outputs that are associated to the labels [main] and [tmp].
17254 The stream sent to the second output of split, labelled as [tmp], is
17256 processed through the crop filter, which crops away the lower half part
17257 of the video, and then vertically flipped. The overlay filter takes in
17258 input the first unchanged output of the split filter (which was
17259 labelled as [main]), and overlay on its lower half the output generated
17260 by the crop,vflip filterchain.
17261 Some filters take in input a list of parameters: they are specified
17263 after the filter name and an equal sign, and are separated from each
17264 other by a colon.
17265 There exist so-called source filters that do not have an audio/video
17267 input, and sink filters that will not have audio/video output.
17268
17270 The graph2dot program included in the FFmpeg tools directory can be
17271 used to parse a filtergraph description and issue a corresponding
17272 textual representation in the dot language.
17273 Invoke the command:
17275 graph2dot -h
17277 to see how to use graph2dot.
17279 You can then pass the dot description to the dot program (from the
17281 graphviz suite of programs) and obtain a graphical representation of
17282 the filtergraph.
17283 For example the sequence of commands:
17285 echo <GRAPH_DESCRIPTION> | \
17287 tools/graph2dot -o graph.tmp && \
17288 dot -Tpng graph.tmp -o graph.png && \
17289 display graph.png
17290 can be used to create and display an image representing the graph
17292 described by the GRAPH_DESCRIPTION string. Note that this string must
17293 be a complete self-contained graph, with its inputs and outputs
17294 explicitly defined. For example if your command line is of the form:
17295 ffmpeg -i infile -vf scale=640:360 outfile
17297 your GRAPH_DESCRIPTION string will need to be of the form:
17299 nullsrc,scale=640:360,nullsink
17301 you may also need to set the nullsrc parameters and add a format filter
17303 in order to simulate a specific input file.
17304
17306 A filtergraph is a directed graph of connected filters. It can contain
17307 cycles, and there can be multiple links between a pair of filters. Each
17308 link has one input pad on one side connecting it to one filter from
17309 which it takes its input, and one output pad on the other side
17310 connecting it to one filter accepting its output.
17311 Each filter in a filtergraph is an instance of a filter class
17313 registered in the application, which defines the features and the
17314 number of input and output pads of the filter.
17315 A filter with no input pads is called a "source", and a filter with no
17317 output pads is called a "sink".
17318 Filtergraph syntax
17320 A filtergraph has a textual representation, which is recognized by the
17321 -filter/-vf/-af and -filter_complex options in ffmpeg and -vf/-af in
17322 ffplay, and by the "avfilter_graph_parse_ptr()" function defined in
17323 libavfilter/avfilter.h.
17324 A filterchain consists of a sequence of connected filters, each one
17326 connected to the previous one in the sequence. A filterchain is
17327 represented by a list of ","-separated filter descriptions.
17328 A filtergraph consists of a sequence of filterchains. A sequence of
17330 filterchains is represented by a list of ";"-separated filterchain
17331 descriptions.
17332 A filter is represented by a string of the form:
17334 [in_link_1]...[in_link_N]filter_name@id=arguments[out_link_1]...[out_link_M]
17335 filter_name is the name of the filter class of which the described
17337 filter is an instance of, and has to be the name of one of the filter
17338 classes registered in the program optionally followed by "@id". The
17339 name of the filter class is optionally followed by a string
17340 "=arguments".
17341 arguments is a string which contains the parameters used to initialize
17343 the filter instance. It may have one of two forms:
17344 • A ':'-separated list of key=value pairs.
17346 • A ':'-separated list of value. In this case, the keys are assumed
17348 to be the option names in the order they are declared. E.g. the
17349 "fade" filter declares three options in this order -- type,
17350 start_frame and nb_frames. Then the parameter list in:0:30 means
17351 that the value in is assigned to the option type, 0 to start_frame
17352 and 30 to nb_frames.
17353 • A ':'-separated list of mixed direct value and long key=value
17355 pairs. The direct value must precede the key=value pairs, and
17356 follow the same constraints order of the previous point. The
17357 following key=value pairs can be set in any preferred order.
17358 If the option value itself is a list of items (e.g. the "format" filter
17360 takes a list of pixel formats), the items in the list are usually
17361 separated by |.
17362 The list of arguments can be quoted using the character ' as initial
17364 and ending mark, and the character \ for escaping the characters within
17365 the quoted text; otherwise the argument string is considered terminated
17366 when the next special character (belonging to the set []=;,) is
17367 encountered.
17368 The name and arguments of the filter are optionally preceded and
17370 followed by a list of link labels. A link label allows one to name a
17371 link and associate it to a filter output or input pad. The preceding
17372 labels in_link_1 ... in_link_N, are associated to the filter input
17373 pads, the following labels out_link_1 ... out_link_M, are associated to
17374 the output pads.
17375 When two link labels with the same name are found in the filtergraph, a
17377 link between the corresponding input and output pad is created.
17378 If an output pad is not labelled, it is linked by default to the first
17380 unlabelled input pad of the next filter in the filterchain. For
17381 example in the filterchain
17382 nullsrc, split[L1], [L2]overlay, nullsink
17384 the split filter instance has two output pads, and the overlay filter
17386 instance two input pads. The first output pad of split is labelled
17387 "L1", the first input pad of overlay is labelled "L2", and the second
17388 output pad of split is linked to the second input pad of overlay, which
17389 are both unlabelled.
17390 In a filter description, if the input label of the first filter is not
17392 specified, "in" is assumed; if the output label of the last filter is
17393 not specified, "out" is assumed.
17394 In a complete filterchain all the unlabelled filter input and output
17396 pads must be connected. A filtergraph is considered valid if all the
17397 filter input and output pads of all the filterchains are connected.
17398 Libavfilter will automatically insert scale filters where format
17400 conversion is required. It is possible to specify swscale flags for
17401 those automatically inserted scalers by prepending "sws_flags=flags;"
17402 to the filtergraph description.
17403 Here is a BNF description of the filtergraph syntax:
17405 <NAME> ::= sequence of alphanumeric characters and '_'
17407 <FILTER_NAME> ::= <NAME>["@"<NAME>]
17408 <LINKLABEL> ::= "[" <NAME> "]"
17409 <LINKLABELS> ::= <LINKLABEL> [<LINKLABELS>]
17410 <FILTER_ARGUMENTS> ::= sequence of chars (possibly quoted)
17411 <FILTER> ::= [<LINKLABELS>] <FILTER_NAME> ["=" <FILTER_ARGUMENTS>] [<LINKLABELS>]
17412 <FILTERCHAIN> ::= <FILTER> [,<FILTERCHAIN>]
17413 <FILTERGRAPH> ::= [sws_flags=<flags>;] <FILTERCHAIN> [;<FILTERGRAPH>]
17414 Notes on filtergraph escaping
17416 Filtergraph description composition entails several levels of escaping.
17417 See the "Quoting and escaping" section in the ffmpeg-utils(1) manual
17418 for more information about the employed escaping procedure.
17419 A first level escaping affects the content of each filter option value,
17421 which may contain the special character ":" used to separate values, or
17422 one of the escaping characters "\'".
17423 A second level escaping affects the whole filter description, which may
17425 contain the escaping characters "\'" or the special characters "[],;"
17426 used by the filtergraph description.
17427 Finally, when you specify a filtergraph on a shell commandline, you
17429 need to perform a third level escaping for the shell special characters
17430 contained within it.
17431 For example, consider the following string to be embedded in the
17433 drawtext filter description text value:
17434 this is a 'string': may contain one, or more, special characters
17436 This string contains the "'" special escaping character, and the ":"
17438 special character, so it needs to be escaped in this way:
17439 text=this is a \'string\'\: may contain one, or more, special characters
17441 A second level of escaping is required when embedding the filter
17443 description in a filtergraph description, in order to escape all the
17444 filtergraph special characters. Thus the example above becomes:
17445 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
17447 (note that in addition to the "\'" escaping special characters, also
17449 "," needs to be escaped).
17450 Finally an additional level of escaping is needed when writing the
17452 filtergraph description in a shell command, which depends on the
17453 escaping rules of the adopted shell. For example, assuming that "\" is
17454 special and needs to be escaped with another "\", the previous string
17455 will finally result in:
17456 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
17458
17460 Some filters support a generic enable option. For the filters
17461 supporting timeline editing, this option can be set to an expression
17462 which is evaluated before sending a frame to the filter. If the
17463 evaluation is non-zero, the filter will be enabled, otherwise the frame
17464 will be sent unchanged to the next filter in the filtergraph.
17465 The expression accepts the following values:
17467 t timestamp expressed in seconds, NAN if the input timestamp is
17469 unknown
17470 n sequential number of the input frame, starting from 0
17472 pos the position in the file of the input frame, NAN if unknown
17474 w
17476 h width and height of the input frame if video
17477 Additionally, these filters support an enable command that can be used
17479 to re-define the expression.
17480 Like any other filtering option, the enable option follows the same
17482 rules.
17483 For example, to enable a blur filter (smartblur) from 10 seconds to 3
17485 minutes, and a curves filter starting at 3 seconds:
17486 smartblur = enable='between(t,10,3*60)',
17488 curves = enable='gte(t,3)' : preset=cross_process
17489 See "ffmpeg -filters" to view which filters have timeline support.
17491
17493 Some options can be changed during the operation of the filter using a
17494 command. These options are marked 'T' on the output of ffmpeg -h
17495 filter=<name of filter>. The name of the command is the name of the
17496 option and the argument is the new value.
17497
17499 Some filters with several inputs support a common set of options.
17500 These options can only be set by name, not with the short notation.
17501 eof_action
17503 The action to take when EOF is encountered on the secondary input;
17504 it accepts one of the following values:
17505 repeat
17507 Repeat the last frame (the default).
17508 endall
17510 End both streams.
17511 pass
17513 Pass the main input through.
17514 shortest
17516 If set to 1, force the output to terminate when the shortest input
17517 terminates. Default value is 0.
17518 repeatlast
17520 If set to 1, force the filter to extend the last frame of secondary
17521 streams until the end of the primary stream. A value of 0 disables
17522 this behavior. Default value is 1.
17523
17525 When you configure your FFmpeg build, you can disable any of the
17526 existing filters using "--disable-filters". The configure output will
17527 show the audio filters included in your build.
17528 Below is a description of the currently available audio filters.
17530 acompressor
17532 A compressor is mainly used to reduce the dynamic range of a signal.
17533 Especially modern music is mostly compressed at a high ratio to improve
17534 the overall loudness. It's done to get the highest attention of a
17535 listener, "fatten" the sound and bring more "power" to the track. If a
17536 signal is compressed too much it may sound dull or "dead" afterwards or
17537 it may start to "pump" (which could be a powerful effect but can also
17538 destroy a track completely). The right compression is the key to reach
17539 a professional sound and is the high art of mixing and mastering.
17540 Because of its complex settings it may take a long time to get the
17541 right feeling for this kind of effect.
17542 Compression is done by detecting the volume above a chosen level
17544 "threshold" and dividing it by the factor set with "ratio". So if you
17545 set the threshold to -12dB and your signal reaches -6dB a ratio of 2:1
17546 will result in a signal at -9dB. Because an exact manipulation of the
17547 signal would cause distortion of the waveform the reduction can be
17548 levelled over the time. This is done by setting "Attack" and "Release".
17549 "attack" determines how long the signal has to rise above the threshold
17550 before any reduction will occur and "release" sets the time the signal
17551 has to fall below the threshold to reduce the reduction again. Shorter
17552 signals than the chosen attack time will be left untouched. The
17553 overall reduction of the signal can be made up afterwards with the
17554 "makeup" setting. So compressing the peaks of a signal about 6dB and
17555 raising the makeup to this level results in a signal twice as loud than
17556 the source. To gain a softer entry in the compression the "knee"
17557 flattens the hard edge at the threshold in the range of the chosen
17558 decibels.
17559 The filter accepts the following options:
17561 level_in
17563 Set input gain. Default is 1. Range is between 0.015625 and 64.
17564 mode
17566 Set mode of compressor operation. Can be "upward" or "downward".
17567 Default is "downward".
17568 threshold
17570 If a signal of stream rises above this level it will affect the
17571 gain reduction. By default it is 0.125. Range is between
17572 0.00097563 and 1.
17573 ratio
17575 Set a ratio by which the signal is reduced. 1:2 means that if the
17576 level rose 4dB above the threshold, it will be only 2dB above after
17577 the reduction. Default is 2. Range is between 1 and 20.
17578 attack
17580 Amount of milliseconds the signal has to rise above the threshold
17581 before gain reduction starts. Default is 20. Range is between 0.01
17582 and 2000.
17583 release
17585 Amount of milliseconds the signal has to fall below the threshold
17586 before reduction is decreased again. Default is 250. Range is
17587 between 0.01 and 9000.
17588 makeup
17590 Set the amount by how much signal will be amplified after
17591 processing. Default is 1. Range is from 1 to 64.
17592 knee
17594 Curve the sharp knee around the threshold to enter gain reduction
17595 more softly. Default is 2.82843. Range is between 1 and 8.
17596 link
17598 Choose if the "average" level between all channels of input stream
17599 or the louder("maximum") channel of input stream affects the
17600 reduction. Default is "average".
17601 detection
17603 Should the exact signal be taken in case of "peak" or an RMS one in
17604 case of "rms". Default is "rms" which is mostly smoother.
17605 mix How much to use compressed signal in output. Default is 1. Range
17607 is between 0 and 1.
17608 Commands
17610 This filter supports the all above options as commands.
17612 acontrast
17614 Simple audio dynamic range compression/expansion filter.
17615 The filter accepts the following options:
17617 contrast
17619 Set contrast. Default is 33. Allowed range is between 0 and 100.
17620 acopy
17622 Copy the input audio source unchanged to the output. This is mainly
17623 useful for testing purposes.
17624 acrossfade
17626 Apply cross fade from one input audio stream to another input audio
17627 stream. The cross fade is applied for specified duration near the end
17628 of first stream.
17629 The filter accepts the following options:
17631 nb_samples, ns
17633 Specify the number of samples for which the cross fade effect has
17634 to last. At the end of the cross fade effect the first input audio
17635 will be completely silent. Default is 44100.
17636 duration, d
17638 Specify the duration of the cross fade effect. See the Time
17639 duration section in the ffmpeg-utils(1) manual for the accepted
17640 syntax. By default the duration is determined by nb_samples. If
17641 set this option is used instead of nb_samples.
17642 overlap, o
17644 Should first stream end overlap with second stream start. Default
17645 is enabled.
17646 curve1
17648 Set curve for cross fade transition for first stream.
17649 curve2
17651 Set curve for cross fade transition for second stream.
17652 For description of available curve types see afade filter
17654 description.
17655 Examples
17657 • Cross fade from one input to another:
17659 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
17661 • Cross fade from one input to another but without overlapping:
17663 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
17665 acrossover
17667 Split audio stream into several bands.
17668 This filter splits audio stream into two or more frequency ranges.
17670 Summing all streams back will give flat output.
17671 The filter accepts the following options:
17673 split
17675 Set split frequencies. Those must be positive and increasing.
17676 order
17678 Set filter order for each band split. This controls filter roll-off
17679 or steepness of filter transfer function. Available values are:
17680 2nd 12 dB per octave.
17682 4th 24 dB per octave.
17684 6th 36 dB per octave.
17686 8th 48 dB per octave.
17688 10th
17690 60 dB per octave.
17691 12th
17693 72 dB per octave.
17694 14th
17696 84 dB per octave.
17697 16th
17699 96 dB per octave.
17700 18th
17702 108 dB per octave.
17703 20th
17705 120 dB per octave.
17706 Default is 4th.
17708 level
17710 Set input gain level. Allowed range is from 0 to 1. Default value
17711 is 1.
17712 gains
17714 Set output gain for each band. Default value is 1 for all bands.
17715 Examples
17717 • Split input audio stream into two bands (low and high) with split
17719 frequency of 1500 Hz, each band will be in separate stream:
17720 ffmpeg -i in.flac -filter_complex 'acrossover=split=1500[LOW][HIGH]' -map '[LOW]' low.wav -map '[HIGH]' high.wav
17722 • Same as above, but with higher filter order:
17724 ffmpeg -i in.flac -filter_complex 'acrossover=split=1500:order=8th[LOW][HIGH]' -map '[LOW]' low.wav -map '[HIGH]' high.wav
17726 • Same as above, but also with additional middle band (frequencies
17728 between 1500 and 8000):
17729 ffmpeg -i in.flac -filter_complex 'acrossover=split=1500 8000:order=8th[LOW][MID][HIGH]' -map '[LOW]' low.wav -map '[MID]' mid.wav -map '[HIGH]' high.wav
17731 acrusher
17733 Reduce audio bit resolution.
17734 This filter is bit crusher with enhanced functionality. A bit crusher
17736 is used to audibly reduce number of bits an audio signal is sampled
17737 with. This doesn't change the bit depth at all, it just produces the
17738 effect. Material reduced in bit depth sounds more harsh and "digital".
17739 This filter is able to even round to continuous values instead of
17740 discrete bit depths. Additionally it has a D/C offset which results in
17741 different crushing of the lower and the upper half of the signal. An
17742 Anti-Aliasing setting is able to produce "softer" crushing sounds.
17743 Another feature of this filter is the logarithmic mode. This setting
17745 switches from linear distances between bits to logarithmic ones. The
17746 result is a much more "natural" sounding crusher which doesn't gate low
17747 signals for example. The human ear has a logarithmic perception, so
17748 this kind of crushing is much more pleasant. Logarithmic crushing is
17749 also able to get anti-aliased.
17750 The filter accepts the following options:
17752 level_in
17754 Set level in.
17755 level_out
17757 Set level out.
17758 bits
17760 Set bit reduction.
17761 mix Set mixing amount.
17763 mode
17765 Can be linear: "lin" or logarithmic: "log".
17766 dc Set DC.
17768 aa Set anti-aliasing.
17770 samples
17772 Set sample reduction.
17773 lfo Enable LFO. By default disabled.
17775 lforange
17777 Set LFO range.
17778 lforate
17780 Set LFO rate.
17781 Commands
17783 This filter supports the all above options as commands.
17785 acue
17787 Delay audio filtering until a given wallclock timestamp. See the cue
17788 filter.
17789 adeclick
17791 Remove impulsive noise from input audio.
17792 Samples detected as impulsive noise are replaced by interpolated
17794 samples using autoregressive modelling.
17795 window, w
17797 Set window size, in milliseconds. Allowed range is from 10 to 100.
17798 Default value is 55 milliseconds. This sets size of window which
17799 will be processed at once.
17800 overlap, o
17802 Set window overlap, in percentage of window size. Allowed range is
17803 from 50 to 95. Default value is 75 percent. Setting this to a very
17804 high value increases impulsive noise removal but makes whole
17805 process much slower.
17806 arorder, a
17808 Set autoregression order, in percentage of window size. Allowed
17809 range is from 0 to 25. Default value is 2 percent. This option also
17810 controls quality of interpolated samples using neighbour good
17811 samples.
17812 threshold, t
17814 Set threshold value. Allowed range is from 1 to 100. Default value
17815 is 2. This controls the strength of impulsive noise which is going
17816 to be removed. The lower value, the more samples will be detected
17817 as impulsive noise.
17818 burst, b
17820 Set burst fusion, in percentage of window size. Allowed range is 0
17821 to 10. Default value is 2. If any two samples detected as noise
17822 are spaced less than this value then any sample between those two
17823 samples will be also detected as noise.
17824 method, m
17826 Set overlap method.
17827 It accepts the following values:
17829 add, a
17831 Select overlap-add method. Even not interpolated samples are
17832 slightly changed with this method.
17833 save, s
17835 Select overlap-save method. Not interpolated samples remain
17836 unchanged.
17837 Default value is "a".
17839 adeclip
17841 Remove clipped samples from input audio.
17842 Samples detected as clipped are replaced by interpolated samples using
17844 autoregressive modelling.
17845 window, w
17847 Set window size, in milliseconds. Allowed range is from 10 to 100.
17848 Default value is 55 milliseconds. This sets size of window which
17849 will be processed at once.
17850 overlap, o
17852 Set window overlap, in percentage of window size. Allowed range is
17853 from 50 to 95. Default value is 75 percent.
17854 arorder, a
17856 Set autoregression order, in percentage of window size. Allowed
17857 range is from 0 to 25. Default value is 8 percent. This option also
17858 controls quality of interpolated samples using neighbour good
17859 samples.
17860 threshold, t
17862 Set threshold value. Allowed range is from 1 to 100. Default value
17863 is 10. Higher values make clip detection less aggressive.
17864 hsize, n
17866 Set size of histogram used to detect clips. Allowed range is from
17867 100 to 9999. Default value is 1000. Higher values make clip
17868 detection less aggressive.
17869 method, m
17871 Set overlap method.
17872 It accepts the following values:
17874 add, a
17876 Select overlap-add method. Even not interpolated samples are
17877 slightly changed with this method.
17878 save, s
17880 Select overlap-save method. Not interpolated samples remain
17881 unchanged.
17882 Default value is "a".
17884 adecorrelate
17886 Apply decorrelation to input audio stream.
17887 The filter accepts the following options:
17889 stages
17891 Set decorrelation stages of filtering. Allowed range is from 1 to
17892 16. Default value is 6.
17893 seed
17895 Set random seed used for setting delay in samples across channels.
17896 adelay
17898 Delay one or more audio channels.
17899 Samples in delayed channel are filled with silence.
17901 The filter accepts the following option:
17903 delays
17905 Set list of delays in milliseconds for each channel separated by
17906 '|'. Unused delays will be silently ignored. If number of given
17907 delays is smaller than number of channels all remaining channels
17908 will not be delayed. If you want to delay exact number of samples,
17909 append 'S' to number. If you want instead to delay in seconds,
17910 append 's' to number.
17911 all Use last set delay for all remaining channels. By default is
17913 disabled. This option if enabled changes how option "delays" is
17914 interpreted.
17915 Examples
17917 • Delay first channel by 1.5 seconds, the third channel by 0.5
17919 seconds and leave the second channel (and any other channels that
17920 may be present) unchanged.
17921 adelay=1500|0|500
17923 • Delay second channel by 500 samples, the third channel by 700
17925 samples and leave the first channel (and any other channels that
17926 may be present) unchanged.
17927 adelay=0|500S|700S
17929 • Delay all channels by same number of samples:
17931 adelay=delays=64S:all=1
17933 adenorm
17935 Remedy denormals in audio by adding extremely low-level noise.
17936 This filter shall be placed before any filter that can produce
17938 denormals.
17939 A description of the accepted parameters follows.
17941 level
17943 Set level of added noise in dB. Default is "-351". Allowed range
17944 is from -451 to -90.
17945 type
17947 Set type of added noise.
17948 dc Add DC signal.
17950 ac Add AC signal.
17952 square
17954 Add square signal.
17955 pulse
17957 Add pulse signal.
17958 Default is "dc".
17960 Commands
17962 This filter supports the all above options as commands.
17964 aderivative, aintegral
17966 Compute derivative/integral of audio stream.
17967 Applying both filters one after another produces original audio.
17969 adynamicequalizer
17971 Apply dynamic equalization to input audio stream.
17972 A description of the accepted options follows.
17974 threshold
17976 Set the detection threshold used to trigger equalization.
17977 Threshold detection is using bandpass filter. Default value is 0.
17978 Allowed range is from 0 to 100.
17979 dfrequency
17981 Set the detection frequency in Hz used for bandpass filter used to
17982 trigger equalization. Default value is 1000 Hz. Allowed range is
17983 between 2 and 1000000 Hz.
17984 dqfactor
17986 Set the detection resonance factor for bandpass filter used to
17987 trigger equalization. Default value is 1. Allowed range is from
17988 0.001 to 1000.
17989 tfrequency
17991 Set the target frequency of equalization filter. Default value is
17992 1000 Hz. Allowed range is between 2 and 1000000 Hz.
17993 tqfactor
17995 Set the target resonance factor for target equalization filter.
17996 Default value is 1. Allowed range is from 0.001 to 1000.
17997 attack
17999 Set the amount of milliseconds the signal from detection has to
18000 rise above the detection threshold before equalization starts.
18001 Default is 20. Allowed range is between 1 and 2000.
18002 release
18004 Set the amount of milliseconds the signal from detection has to
18005 fall below the detection threshold before equalization ends.
18006 Default is 200. Allowed range is between 1 and 2000.
18007 knee
18009 Curve the sharp knee around the detection threshold to calculate
18010 equalization gain more softly. Default is 1. Allowed range is
18011 between 0 and 8.
18012 ratio
18014 Set the ratio by which the equalization gain is raised. Default is
18015 1. Allowed range is between 1 and 20.
18016 makeup
18018 Set the makeup offset in dB by which the equalization gain is
18019 raised. Default is 0. Allowed range is between 0 and 30.
18020 range
18022 Set the max allowed cut/boost amount in dB. Default is 0. Allowed
18023 range is from 0 to 200.
18024 slew
18026 Set the slew factor. Default is 1. Allowed range is from 1 to 200.
18027 mode
18029 Set the mode of filter operation, can be one of the following:
18030 listen
18032 Output only isolated bandpass signal.
18033 cut Cut frequencies above detection threshold.
18035 boost
18037 Boost frequencies bellow detection threshold.
18038 Default mode is cut.
18040 Commands
18042 This filter supports the all above options as commands.
18044 adynamicsmooth
18046 Apply dynamic smoothing to input audio stream.
18047 A description of the accepted options follows.
18049 sensitivity
18051 Set an amount of sensitivity to frequency fluctations. Default is
18052 2. Allowed range is from 0 to 1e+06.
18053 basefreq
18055 Set a base frequency for smoothing. Default value is 22050.
18056 Allowed range is from 2 to 1e+06.
18057 Commands
18059 This filter supports the all above options as commands.
18061 aecho
18063 Apply echoing to the input audio.
18064 Echoes are reflected sound and can occur naturally amongst mountains
18066 (and sometimes large buildings) when talking or shouting; digital echo
18067 effects emulate this behaviour and are often used to help fill out the
18068 sound of a single instrument or vocal. The time difference between the
18069 original signal and the reflection is the "delay", and the loudness of
18070 the reflected signal is the "decay". Multiple echoes can have
18071 different delays and decays.
18072 A description of the accepted parameters follows.
18074 in_gain
18076 Set input gain of reflected signal. Default is 0.6.
18077 out_gain
18079 Set output gain of reflected signal. Default is 0.3.
18080 delays
18082 Set list of time intervals in milliseconds between original signal
18083 and reflections separated by '|'. Allowed range for each "delay" is
18084 "(0 - 90000.0]". Default is 1000.
18085 decays
18087 Set list of loudness of reflected signals separated by '|'.
18088 Allowed range for each "decay" is "(0 - 1.0]". Default is 0.5.
18089 Examples
18091 • Make it sound as if there are twice as many instruments as are
18093 actually playing:
18094 aecho=0.8:0.88:60:0.4
18096 • If delay is very short, then it sounds like a (metallic) robot
18098 playing music:
18099 aecho=0.8:0.88:6:0.4
18101 • A longer delay will sound like an open air concert in the
18103 mountains:
18104 aecho=0.8:0.9:1000:0.3
18106 • Same as above but with one more mountain:
18108 aecho=0.8:0.9:1000|1800:0.3|0.25
18110 aemphasis
18112 Audio emphasis filter creates or restores material directly taken from
18113 LPs or emphased CDs with different filter curves. E.g. to store music
18114 on vinyl the signal has to be altered by a filter first to even out the
18115 disadvantages of this recording medium. Once the material is played
18116 back the inverse filter has to be applied to restore the distortion of
18117 the frequency response.
18118 The filter accepts the following options:
18120 level_in
18122 Set input gain.
18123 level_out
18125 Set output gain.
18126 mode
18128 Set filter mode. For restoring material use "reproduction" mode,
18129 otherwise use "production" mode. Default is "reproduction" mode.
18130 type
18132 Set filter type. Selects medium. Can be one of the following:
18133 col select Columbia.
18135 emi select EMI.
18137 bsi select BSI (78RPM).
18139 riaa
18141 select RIAA.
18142 cd select Compact Disc (CD).
18144 50fm
18146 select 50Xs (FM).
18147 75fm
18149 select 75Xs (FM).
18150 50kf
18152 select 50Xs (FM-KF).
18153 75kf
18155 select 75Xs (FM-KF).
18156 Commands
18158 This filter supports the all above options as commands.
18160 aeval
18162 Modify an audio signal according to the specified expressions.
18163 This filter accepts one or more expressions (one for each channel),
18165 which are evaluated and used to modify a corresponding audio signal.
18166 It accepts the following parameters:
18168 exprs
18170 Set the '|'-separated expressions list for each separate channel.
18171 If the number of input channels is greater than the number of
18172 expressions, the last specified expression is used for the
18173 remaining output channels.
18174 channel_layout, c
18176 Set output channel layout. If not specified, the channel layout is
18177 specified by the number of expressions. If set to same, it will use
18178 by default the same input channel layout.
18179 Each expression in exprs can contain the following constants and
18181 functions:
18182 ch channel number of the current expression
18184 n number of the evaluated sample, starting from 0
18186 s sample rate
18188 t time of the evaluated sample expressed in seconds
18190 nb_in_channels
18192 nb_out_channels
18193 input and output number of channels
18194 val(CH)
18196 the value of input channel with number CH
18197 Note: this filter is slow. For faster processing you should use a
18199 dedicated filter.
18200 Examples
18202 • Half volume:
18204 aeval=val(ch)/2:c=same
18206 • Invert phase of the second channel:
18208 aeval=val(0)|-val(1)
18210 aexciter
18212 An exciter is used to produce high sound that is not present in the
18213 original signal. This is done by creating harmonic distortions of the
18214 signal which are restricted in range and added to the original signal.
18215 An Exciter raises the upper end of an audio signal without simply
18216 raising the higher frequencies like an equalizer would do to create a
18217 more "crisp" or "brilliant" sound.
18218 The filter accepts the following options:
18220 level_in
18222 Set input level prior processing of signal. Allowed range is from
18223 0 to 64. Default value is 1.
18224 level_out
18226 Set output level after processing of signal. Allowed range is from
18227 0 to 64. Default value is 1.
18228 amount
18230 Set the amount of harmonics added to original signal. Allowed
18231 range is from 0 to 64. Default value is 1.
18232 drive
18234 Set the amount of newly created harmonics. Allowed range is from
18235 0.1 to 10. Default value is 8.5.
18236 blend
18238 Set the octave of newly created harmonics. Allowed range is from
18239 -10 to 10. Default value is 0.
18240 freq
18242 Set the lower frequency limit of producing harmonics in Hz.
18243 Allowed range is from 2000 to 12000 Hz. Default is 7500 Hz.
18244 ceil
18246 Set the upper frequency limit of producing harmonics. Allowed
18247 range is from 9999 to 20000 Hz. If value is lower than 10000 Hz no
18248 limit is applied.
18249 listen
18251 Mute the original signal and output only added harmonics. By
18252 default is disabled.
18253 Commands
18255 This filter supports the all above options as commands.
18257 afade
18259 Apply fade-in/out effect to input audio.
18260 A description of the accepted parameters follows.
18262 type, t
18264 Specify the effect type, can be either "in" for fade-in, or "out"
18265 for a fade-out effect. Default is "in".
18266 start_sample, ss
18268 Specify the number of the start sample for starting to apply the
18269 fade effect. Default is 0.
18270 nb_samples, ns
18272 Specify the number of samples for which the fade effect has to
18273 last. At the end of the fade-in effect the output audio will have
18274 the same volume as the input audio, at the end of the fade-out
18275 transition the output audio will be silence. Default is 44100.
18276 start_time, st
18278 Specify the start time of the fade effect. Default is 0. The value
18279 must be specified as a time duration; see the Time duration section
18280 in the ffmpeg-utils(1) manual for the accepted syntax. If set this
18281 option is used instead of start_sample.
18282 duration, d
18284 Specify the duration of the fade effect. See the Time duration
18285 section in the ffmpeg-utils(1) manual for the accepted syntax. At
18286 the end of the fade-in effect the output audio will have the same
18287 volume as the input audio, at the end of the fade-out transition
18288 the output audio will be silence. By default the duration is
18289 determined by nb_samples. If set this option is used instead of
18290 nb_samples.
18291 curve
18293 Set curve for fade transition.
18294 It accepts the following values:
18296 tri select triangular, linear slope (default)
18298 qsin
18300 select quarter of sine wave
18301 hsin
18303 select half of sine wave
18304 esin
18306 select exponential sine wave
18307 log select logarithmic
18309 ipar
18311 select inverted parabola
18312 qua select quadratic
18314 cub select cubic
18316 squ select square root
18318 cbr select cubic root
18320 par select parabola
18322 exp select exponential
18324 iqsin
18326 select inverted quarter of sine wave
18327 ihsin
18329 select inverted half of sine wave
18330 dese
18332 select double-exponential seat
18333 desi
18335 select double-exponential sigmoid
18336 losi
18338 select logistic sigmoid
18339 sinc
18341 select sine cardinal function
18342 isinc
18344 select inverted sine cardinal function
18345 nofade
18347 no fade applied
18348 Commands
18350 This filter supports the all above options as commands.
18352 Examples
18354 • Fade in first 15 seconds of audio:
18356 afade=t=in:ss=0:d=15
18358 • Fade out last 25 seconds of a 900 seconds audio:
18360 afade=t=out:st=875:d=25
18362 afftdn
18364 Denoise audio samples with FFT.
18365 A description of the accepted parameters follows.
18367 nr Set the noise reduction in dB, allowed range is 0.01 to 97.
18369 Default value is 12 dB.
18370 nf Set the noise floor in dB, allowed range is -80 to -20. Default
18372 value is -50 dB.
18373 nt Set the noise type.
18375 It accepts the following values:
18377 w Select white noise.
18379 v Select vinyl noise.
18381 s Select shellac noise.
18383 c Select custom noise, defined in "bn" option.
18385 Default value is white noise.
18387 bn Set custom band noise for every one of 15 bands. Bands are
18389 separated by ' ' or '|'.
18390 rf Set the residual floor in dB, allowed range is -80 to -20. Default
18392 value is -38 dB.
18393 tn Enable noise tracking. By default is disabled. With this enabled,
18395 noise floor is automatically adjusted.
18396 tr Enable residual tracking. By default is disabled.
18398 om Set the output mode.
18400 It accepts the following values:
18402 i Pass input unchanged.
18404 o Pass noise filtered out.
18406 n Pass only noise.
18408 Default value is o.
18410 Commands
18412 This filter supports the following commands:
18414 sample_noise, sn
18416 Start or stop measuring noise profile. Syntax for the command is :
18417 "start" or "stop" string. After measuring noise profile is stopped
18418 it will be automatically applied in filtering.
18419 noise_reduction, nr
18421 Change noise reduction. Argument is single float number. Syntax
18422 for the command is : "noise_reduction"
18423 noise_floor, nf
18425 Change noise floor. Argument is single float number. Syntax for
18426 the command is : "noise_floor"
18427 output_mode, om
18429 Change output mode operation. Syntax for the command is : "i", "o"
18430 or "n" string.
18431 afftfilt
18433 Apply arbitrary expressions to samples in frequency domain.
18434 real
18436 Set frequency domain real expression for each separate channel
18437 separated by '|'. Default is "re". If the number of input channels
18438 is greater than the number of expressions, the last specified
18439 expression is used for the remaining output channels.
18440 imag
18442 Set frequency domain imaginary expression for each separate channel
18443 separated by '|'. Default is "im".
18444 Each expression in real and imag can contain the following
18446 constants and functions:
18447 sr sample rate
18449 b current frequency bin number
18451 nb number of available bins
18453 ch channel number of the current expression
18455 chs number of channels
18457 pts current frame pts
18459 re current real part of frequency bin of current channel
18461 im current imaginary part of frequency bin of current channel
18463 real(b, ch)
18465 Return the value of real part of frequency bin at location
18466 (bin,channel)
18467 imag(b, ch)
18469 Return the value of imaginary part of frequency bin at location
18470 (bin,channel)
18471 win_size
18473 Set window size. Allowed range is from 16 to 131072. Default is
18474 4096
18475 win_func
18477 Set window function.
18478 It accepts the following values:
18480 rect
18482 bartlett
18483 hann, hanning
18484 hamming
18485 blackman
18486 welch
18487 flattop
18488 bharris
18489 bnuttall
18490 bhann
18491 sine
18492 nuttall
18493 lanczos
18494 gauss
18495 tukey
18496 dolph
18497 cauchy
18498 parzen
18499 poisson
18500 bohman
18501 Default is "hann".
18503 overlap
18505 Set window overlap. If set to 1, the recommended overlap for
18506 selected window function will be picked. Default is 0.75.
18507 Examples
18509 • Leave almost only low frequencies in audio:
18511 afftfilt="'real=re * (1-clip((b/nb)*b,0,1))':imag='im * (1-clip((b/nb)*b,0,1))'"
18513 • Apply robotize effect:
18515 afftfilt="real='hypot(re,im)*sin(0)':imag='hypot(re,im)*cos(0)':win_size=512:overlap=0.75"
18517 • Apply whisper effect:
18519 afftfilt="real='hypot(re,im)*cos((random(0)*2-1)*2*3.14)':imag='hypot(re,im)*sin((random(1)*2-1)*2*3.14)':win_size=128:overlap=0.8"
18521 afir
18523 Apply an arbitrary Finite Impulse Response filter.
18524 This filter is designed for applying long FIR filters, up to 60 seconds
18526 long.
18527 It can be used as component for digital crossover filters, room
18529 equalization, cross talk cancellation, wavefield synthesis,
18530 auralization, ambiophonics, ambisonics and spatialization.
18531 This filter uses the streams higher than first one as FIR coefficients.
18533 If the non-first stream holds a single channel, it will be used for all
18534 input channels in the first stream, otherwise the number of channels in
18535 the non-first stream must be same as the number of channels in the
18536 first stream.
18537 It accepts the following parameters:
18539 dry Set dry gain. This sets input gain.
18541 wet Set wet gain. This sets final output gain.
18543 length
18545 Set Impulse Response filter length. Default is 1, which means whole
18546 IR is processed.
18547 gtype
18549 Enable applying gain measured from power of IR.
18550 Set which approach to use for auto gain measurement.
18552 none
18554 Do not apply any gain.
18555 peak
18557 select peak gain, very conservative approach. This is default
18558 value.
18559 dc select DC gain, limited application.
18561 gn select gain to noise approach, this is most popular one.
18563 irgain
18565 Set gain to be applied to IR coefficients before filtering.
18566 Allowed range is 0 to 1. This gain is applied after any gain
18567 applied with gtype option.
18568 irfmt
18570 Set format of IR stream. Can be "mono" or "input". Default is
18571 "input".
18572 maxir
18574 Set max allowed Impulse Response filter duration in seconds.
18575 Default is 30 seconds. Allowed range is 0.1 to 60 seconds.
18576 response
18578 Show IR frequency response, magnitude(magenta), phase(green) and
18579 group delay(yellow) in additional video stream. By default it is
18580 disabled.
18581 channel
18583 Set for which IR channel to display frequency response. By default
18584 is first channel displayed. This option is used only when response
18585 is enabled.
18586 size
18588 Set video stream size. This option is used only when response is
18589 enabled.
18590 rate
18592 Set video stream frame rate. This option is used only when response
18593 is enabled.
18594 minp
18596 Set minimal partition size used for convolution. Default is 8192.
18597 Allowed range is from 1 to 32768. Lower values decreases latency
18598 at cost of higher CPU usage.
18599 maxp
18601 Set maximal partition size used for convolution. Default is 8192.
18602 Allowed range is from 8 to 32768. Lower values may increase CPU
18603 usage.
18604 nbirs
18606 Set number of input impulse responses streams which will be
18607 switchable at runtime. Allowed range is from 1 to 32. Default is
18608 1.
18609 ir Set IR stream which will be used for convolution, starting from 0,
18611 should always be lower than supplied value by "nbirs" option.
18612 Default is 0. This option can be changed at runtime via commands.
18613 Examples
18615 • Apply reverb to stream using mono IR file as second input, complete
18617 command using ffmpeg:
18618 ffmpeg -i input.wav -i middle_tunnel_1way_mono.wav -lavfi afir output.wav
18620 aformat
18622 Set output format constraints for the input audio. The framework will
18623 negotiate the most appropriate format to minimize conversions.
18624 It accepts the following parameters:
18626 sample_fmts, f
18628 A '|'-separated list of requested sample formats.
18629 sample_rates, r
18631 A '|'-separated list of requested sample rates.
18632 channel_layouts, cl
18634 A '|'-separated list of requested channel layouts.
18635 See the Channel Layout section in the ffmpeg-utils(1) manual for
18637 the required syntax.
18638 If a parameter is omitted, all values are allowed.
18640 Force the output to either unsigned 8-bit or signed 16-bit stereo
18642 aformat=sample_fmts=u8|s16:channel_layouts=stereo
18644 afreqshift
18646 Apply frequency shift to input audio samples.
18647 The filter accepts the following options:
18649 shift
18651 Specify frequency shift. Allowed range is -INT_MAX to INT_MAX.
18652 Default value is 0.0.
18653 level
18655 Set output gain applied to final output. Allowed range is from 0.0
18656 to 1.0. Default value is 1.0.
18657 order
18659 Set filter order used for filtering. Allowed range is from 1 to 16.
18660 Default value is 8.
18661 Commands
18663 This filter supports the all above options as commands.
18665 afwtdn
18667 Reduce broadband noise from input samples using Wavelets.
18668 A description of the accepted options follows.
18670 sigma
18672 Set the noise sigma, allowed range is from 0 to 1. Default value
18673 is 0. This option controls strength of denoising applied to input
18674 samples. Most useful way to set this option is via decibels, eg.
18675 -45dB.
18676 levels
18678 Set the number of wavelet levels of decomposition. Allowed range
18679 is from 1 to 12. Default value is 10. Setting this too low make
18680 denoising performance very poor.
18681 wavet
18683 Set wavelet type for decomposition of input frame. They are sorted
18684 by number of coefficients, from lowest to highest. More
18685 coefficients means worse filtering speed, but overall better
18686 quality. Available wavelets are:
18687 sym2
18689 sym4
18690 rbior68
18691 deb10
18692 sym10
18693 coif5
18694 bl3
18695 percent
18696 Set percent of full denoising. Allowed range is from 0 to 100
18697 percent. Default value is 85 percent or partial denoising.
18698 profile
18700 If enabled, first input frame will be used as noise profile. If
18701 first frame samples contain non-noise performance will be very
18702 poor.
18703 adaptive
18705 If enabled, input frames are analyzed for presence of noise. If
18706 noise is detected with high possibility then input frame profile
18707 will be used for processing following frames, until new noise frame
18708 is detected.
18709 samples
18711 Set size of single frame in number of samples. Allowed range is
18712 from 512 to 65536. Default frame size is 8192 samples.
18713 softness
18715 Set softness applied inside thresholding function. Allowed range is
18716 from 0 to 10. Default softness is 1.
18717 Commands
18719 This filter supports the all above options as commands.
18721 agate
18723 A gate is mainly used to reduce lower parts of a signal. This kind of
18724 signal processing reduces disturbing noise between useful signals.
18725 Gating is done by detecting the volume below a chosen level threshold
18727 and dividing it by the factor set with ratio. The bottom of the noise
18728 floor is set via range. Because an exact manipulation of the signal
18729 would cause distortion of the waveform the reduction can be levelled
18730 over time. This is done by setting attack and release.
18731 attack determines how long the signal has to fall below the threshold
18733 before any reduction will occur and release sets the time the signal
18734 has to rise above the threshold to reduce the reduction again. Shorter
18735 signals than the chosen attack time will be left untouched.
18736 level_in
18738 Set input level before filtering. Default is 1. Allowed range is
18739 from 0.015625 to 64.
18740 mode
18742 Set the mode of operation. Can be "upward" or "downward". Default
18743 is "downward". If set to "upward" mode, higher parts of signal will
18744 be amplified, expanding dynamic range in upward direction.
18745 Otherwise, in case of "downward" lower parts of signal will be
18746 reduced.
18747 range
18749 Set the level of gain reduction when the signal is below the
18750 threshold. Default is 0.06125. Allowed range is from 0 to 1.
18751 Setting this to 0 disables reduction and then filter behaves like
18752 expander.
18753 threshold
18755 If a signal rises above this level the gain reduction is released.
18756 Default is 0.125. Allowed range is from 0 to 1.
18757 ratio
18759 Set a ratio by which the signal is reduced. Default is 2. Allowed
18760 range is from 1 to 9000.
18761 attack
18763 Amount of milliseconds the signal has to rise above the threshold
18764 before gain reduction stops. Default is 20 milliseconds. Allowed
18765 range is from 0.01 to 9000.
18766 release
18768 Amount of milliseconds the signal has to fall below the threshold
18769 before the reduction is increased again. Default is 250
18770 milliseconds. Allowed range is from 0.01 to 9000.
18771 makeup
18773 Set amount of amplification of signal after processing. Default is
18774 1. Allowed range is from 1 to 64.
18775 knee
18777 Curve the sharp knee around the threshold to enter gain reduction
18778 more softly. Default is 2.828427125. Allowed range is from 1 to 8.
18779 detection
18781 Choose if exact signal should be taken for detection or an RMS like
18782 one. Default is "rms". Can be "peak" or "rms".
18783 link
18785 Choose if the average level between all channels or the louder
18786 channel affects the reduction. Default is "average". Can be
18787 "average" or "maximum".
18788 Commands
18790 This filter supports the all above options as commands.
18792 aiir
18794 Apply an arbitrary Infinite Impulse Response filter.
18795 It accepts the following parameters:
18797 zeros, z
18799 Set B/numerator/zeros/reflection coefficients.
18800 poles, p
18802 Set A/denominator/poles/ladder coefficients.
18803 gains, k
18805 Set channels gains.
18806 dry_gain
18808 Set input gain.
18809 wet_gain
18811 Set output gain.
18812 format, f
18814 Set coefficients format.
18815 ll lattice-ladder function
18817 sf analog transfer function
18819 tf digital transfer function
18821 zp Z-plane zeros/poles, cartesian (default)
18823 pr Z-plane zeros/poles, polar radians
18825 pd Z-plane zeros/poles, polar degrees
18827 sp S-plane zeros/poles
18829 process, r
18831 Set type of processing.
18832 d direct processing
18834 s serial processing
18836 p parallel processing
18838 precision, e
18840 Set filtering precision.
18841 dbl double-precision floating-point (default)
18843 flt single-precision floating-point
18845 i32 32-bit integers
18847 i16 16-bit integers
18849 normalize, n
18851 Normalize filter coefficients, by default is enabled. Enabling it
18852 will normalize magnitude response at DC to 0dB.
18853 mix How much to use filtered signal in output. Default is 1. Range is
18855 between 0 and 1.
18856 response
18858 Show IR frequency response, magnitude(magenta), phase(green) and
18859 group delay(yellow) in additional video stream. By default it is
18860 disabled.
18861 channel
18863 Set for which IR channel to display frequency response. By default
18864 is first channel displayed. This option is used only when response
18865 is enabled.
18866 size
18868 Set video stream size. This option is used only when response is
18869 enabled.
18870 Coefficients in "tf" and "sf" format are separated by spaces and are in
18872 ascending order.
18873 Coefficients in "zp" format are separated by spaces and order of
18875 coefficients doesn't matter. Coefficients in "zp" format are complex
18876 numbers with i imaginary unit.
18877 Different coefficients and gains can be provided for every channel, in
18879 such case use '|' to separate coefficients or gains. Last provided
18880 coefficients will be used for all remaining channels.
18881 Examples
18883 • Apply 2 pole elliptic notch at around 5000Hz for 48000 Hz sample
18885 rate:
18886 aiir=k=1:z=7.957584807809675810E-1 -2.575128568908332300 3.674839853930788710 -2.57512875289799137 7.957586296317130880E-1:p=1 -2.86950072432325953 3.63022088054647218 -2.28075678147272232 6.361362326477423500E-1:f=tf:r=d
18888 • Same as above but in "zp" format:
18890 aiir=k=0.79575848078096756:z=0.80918701+0.58773007i 0.80918701-0.58773007i 0.80884700+0.58784055i 0.80884700-0.58784055i:p=0.63892345+0.59951235i 0.63892345-0.59951235i 0.79582691+0.44198673i 0.79582691-0.44198673i:f=zp:r=s
18892 • Apply 3-rd order analog normalized Butterworth low-pass filter,
18894 using analog transfer function format:
18895 aiir=z=1.3057 0 0 0:p=1.3057 2.3892 2.1860 1:f=sf:r=d
18897 alimiter
18899 The limiter prevents an input signal from rising over a desired
18900 threshold. This limiter uses lookahead technology to prevent your
18901 signal from distorting. It means that there is a small delay after the
18902 signal is processed. Keep in mind that the delay it produces is the
18903 attack time you set.
18904 The filter accepts the following options:
18906 level_in
18908 Set input gain. Default is 1.
18909 level_out
18911 Set output gain. Default is 1.
18912 limit
18914 Don't let signals above this level pass the limiter. Default is 1.
18915 attack
18917 The limiter will reach its attenuation level in this amount of time
18918 in milliseconds. Default is 5 milliseconds.
18919 release
18921 Come back from limiting to attenuation 1.0 in this amount of
18922 milliseconds. Default is 50 milliseconds.
18923 asc When gain reduction is always needed ASC takes care of releasing to
18925 an average reduction level rather than reaching a reduction of 0 in
18926 the release time.
18927 asc_level
18929 Select how much the release time is affected by ASC, 0 means nearly
18930 no changes in release time while 1 produces higher release times.
18931 level
18933 Auto level output signal. Default is enabled. This normalizes
18934 audio back to 0dB if enabled.
18935 Depending on picked setting it is recommended to upsample input 2x or
18937 4x times with aresample before applying this filter.
18938 allpass
18940 Apply a two-pole all-pass filter with central frequency (in Hz)
18941 frequency, and filter-width width. An all-pass filter changes the
18942 audio's frequency to phase relationship without changing its frequency
18943 to amplitude relationship.
18944 The filter accepts the following options:
18946 frequency, f
18948 Set frequency in Hz.
18949 width_type, t
18951 Set method to specify band-width of filter.
18952 h Hz
18954 q Q-Factor
18956 o octave
18958 s slope
18960 k kHz
18962 width, w
18964 Specify the band-width of a filter in width_type units.
18965 mix, m
18967 How much to use filtered signal in output. Default is 1. Range is
18968 between 0 and 1.
18969 channels, c
18971 Specify which channels to filter, by default all available are
18972 filtered.
18973 normalize, n
18975 Normalize biquad coefficients, by default is disabled. Enabling it
18976 will normalize magnitude response at DC to 0dB.
18977 order, o
18979 Set the filter order, can be 1 or 2. Default is 2.
18980 transform, a
18982 Set transform type of IIR filter.
18983 di
18985 dii
18986 tdii
18987 latt
18988 svf
18989 precision, r
18990 Set precison of filtering.
18991 auto
18993 Pick automatic sample format depending on surround filters.
18994 s16 Always use signed 16-bit.
18996 s32 Always use signed 32-bit.
18998 f32 Always use float 32-bit.
19000 f64 Always use float 64-bit.
19002 Commands
19004 This filter supports the following commands:
19006 frequency, f
19008 Change allpass frequency. Syntax for the command is : "frequency"
19009 width_type, t
19011 Change allpass width_type. Syntax for the command is :
19012 "width_type"
19013 width, w
19015 Change allpass width. Syntax for the command is : "width"
19016 mix, m
19018 Change allpass mix. Syntax for the command is : "mix"
19019 aloop
19021 Loop audio samples.
19022 The filter accepts the following options:
19024 loop
19026 Set the number of loops. Setting this value to -1 will result in
19027 infinite loops. Default is 0.
19028 size
19030 Set maximal number of samples. Default is 0.
19031 start
19033 Set first sample of loop. Default is 0.
19034 amerge
19036 Merge two or more audio streams into a single multi-channel stream.
19037 The filter accepts the following options:
19039 inputs
19041 Set the number of inputs. Default is 2.
19042 If the channel layouts of the inputs are disjoint, and therefore
19044 compatible, the channel layout of the output will be set accordingly
19045 and the channels will be reordered as necessary. If the channel layouts
19046 of the inputs are not disjoint, the output will have all the channels
19047 of the first input then all the channels of the second input, in that
19048 order, and the channel layout of the output will be the default value
19049 corresponding to the total number of channels.
19050 For example, if the first input is in 2.1 (FL+FR+LF) and the second
19052 input is FC+BL+BR, then the output will be in 5.1, with the channels in
19053 the following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of
19054 the first input, b1 is the first channel of the second input).
19055 On the other hand, if both input are in stereo, the output channels
19057 will be in the default order: a1, a2, b1, b2, and the channel layout
19058 will be arbitrarily set to 4.0, which may or may not be the expected
19059 value.
19060 All inputs must have the same sample rate, and format.
19062 If inputs do not have the same duration, the output will stop with the
19064 shortest.
19065 Examples
19067 • Merge two mono files into a stereo stream:
19069 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
19071 • Multiple merges assuming 1 video stream and 6 audio streams in
19073 input.mkv:
19074 ffmpeg -i input.mkv -filter_complex "[0:1][0:2][0:3][0:4][0:5][0:6] amerge=inputs=6" -c:a pcm_s16le output.mkv
19076 amix
19078 Mixes multiple audio inputs into a single output.
19079 Note that this filter only supports float samples (the amerge and pan
19081 audio filters support many formats). If the amix input has integer
19082 samples then aresample will be automatically inserted to perform the
19083 conversion to float samples.
19084 For example
19086 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
19088 will mix 3 input audio streams to a single output with the same
19090 duration as the first input and a dropout transition time of 3 seconds.
19091 It accepts the following parameters:
19093 inputs
19095 The number of inputs. If unspecified, it defaults to 2.
19096 duration
19098 How to determine the end-of-stream.
19099 longest
19101 The duration of the longest input. (default)
19102 shortest
19104 The duration of the shortest input.
19105 first
19107 The duration of the first input.
19108 dropout_transition
19110 The transition time, in seconds, for volume renormalization when an
19111 input stream ends. The default value is 2 seconds.
19112 weights
19114 Specify weight of each input audio stream as sequence. Each weight
19115 is separated by space. By default all inputs have same weight.
19116 normalize
19118 Always scale inputs instead of only doing summation of samples.
19119 Beware of heavy clipping if inputs are not normalized prior or
19120 after filtering by this filter if this option is disabled. By
19121 default is enabled.
19122 Commands
19124 This filter supports the following commands:
19126 weights
19128 normalize
19129 Syntax is same as option with same name.
19130 amultiply
19132 Multiply first audio stream with second audio stream and store result
19133 in output audio stream. Multiplication is done by multiplying each
19134 sample from first stream with sample at same position from second
19135 stream.
19136 With this element-wise multiplication one can create amplitude fades
19138 and amplitude modulations.
19139 anequalizer
19141 High-order parametric multiband equalizer for each channel.
19142 It accepts the following parameters:
19144 params
19146 This option string is in format: "cchn f=cf w=w g=g t=f | ..."
19147 Each equalizer band is separated by '|'.
19148 chn Set channel number to which equalization will be applied. If
19150 input doesn't have that channel the entry is ignored.
19151 f Set central frequency for band. If input doesn't have that
19153 frequency the entry is ignored.
19154 w Set band width in Hertz.
19156 g Set band gain in dB.
19158 t Set filter type for band, optional, can be:
19160 0 Butterworth, this is default.
19162 1 Chebyshev type 1.
19164 2 Chebyshev type 2.
19166 curves
19168 With this option activated frequency response of anequalizer is
19169 displayed in video stream.
19170 size
19172 Set video stream size. Only useful if curves option is activated.
19173 mgain
19175 Set max gain that will be displayed. Only useful if curves option
19176 is activated. Setting this to a reasonable value makes it possible
19177 to display gain which is derived from neighbour bands which are too
19178 close to each other and thus produce higher gain when both are
19179 activated.
19180 fscale
19182 Set frequency scale used to draw frequency response in video
19183 output. Can be linear or logarithmic. Default is logarithmic.
19184 colors
19186 Set color for each channel curve which is going to be displayed in
19187 video stream. This is list of color names separated by space or by
19188 '|'. Unrecognised or missing colors will be replaced by white
19189 color.
19190 Examples
19192 • Lower gain by 10 of central frequency 200Hz and width 100 Hz for
19194 first 2 channels using Chebyshev type 1 filter:
19195 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
19197 Commands
19199 This filter supports the following commands:
19201 change
19203 Alter existing filter parameters. Syntax for the commands is :
19204 "fN|f=freq|w=width|g=gain"
19205 fN is existing filter number, starting from 0, if no such filter is
19207 available error is returned. freq set new frequency parameter.
19208 width set new width parameter in Hertz. gain set new gain
19209 parameter in dB.
19210 Full filter invocation with asendcmd may look like this:
19212 asendcmd=c='4.0 anequalizer change
19213 0|f=200|w=50|g=1',anequalizer=...
19214 anlmdn
19216 Reduce broadband noise in audio samples using Non-Local Means
19217 algorithm.
19218 Each sample is adjusted by looking for other samples with similar
19220 contexts. This context similarity is defined by comparing their
19221 surrounding patches of size p. Patches are searched in an area of r
19222 around the sample.
19223 The filter accepts the following options:
19225 strength, s
19227 Set denoising strength. Allowed range is from 0.00001 to 10.
19228 Default value is 0.00001.
19229 patch, p
19231 Set patch radius duration. Allowed range is from 1 to 100
19232 milliseconds. Default value is 2 milliseconds.
19233 research, r
19235 Set research radius duration. Allowed range is from 2 to 300
19236 milliseconds. Default value is 6 milliseconds.
19237 output, o
19239 Set the output mode.
19240 It accepts the following values:
19242 i Pass input unchanged.
19244 o Pass noise filtered out.
19246 n Pass only noise.
19248 Default value is o.
19250 smooth, m
19252 Set smooth factor. Default value is 11. Allowed range is from 1 to
19253 15.
19254 Commands
19256 This filter supports the all above options as commands.
19258 anlmf, anlms
19260 Apply Normalized Least-Mean-(Squares|Fourth) algorithm to the first
19261 audio stream using the second audio stream.
19262 This adaptive filter is used to mimic a desired filter by finding the
19264 filter coefficients that relate to producing the least mean square of
19265 the error signal (difference between the desired, 2nd input audio
19266 stream and the actual signal, the 1st input audio stream).
19267 A description of the accepted options follows.
19269 order
19271 Set filter order.
19272 mu Set filter mu.
19274 eps Set the filter eps.
19276 leakage
19278 Set the filter leakage.
19279 out_mode
19281 It accepts the following values:
19282 i Pass the 1st input.
19284 d Pass the 2nd input.
19286 o Pass filtered samples.
19288 n Pass difference between desired and filtered samples.
19290 Default value is o.
19292 Examples
19294 • One of many usages of this filter is noise reduction, input audio
19296 is filtered with same samples that are delayed by fixed amount, one
19297 such example for stereo audio is:
19298 asplit[a][b],[a]adelay=32S|32S[a],[b][a]anlms=order=128:leakage=0.0005:mu=.5:out_mode=o
19300 Commands
19302 This filter supports the same commands as options, excluding option
19304 "order".
19305 anull
19307 Pass the audio source unchanged to the output.
19308 apad
19310 Pad the end of an audio stream with silence.
19311 This can be used together with ffmpeg -shortest to extend audio streams
19313 to the same length as the video stream.
19314 A description of the accepted options follows.
19316 packet_size
19318 Set silence packet size. Default value is 4096.
19319 pad_len
19321 Set the number of samples of silence to add to the end. After the
19322 value is reached, the stream is terminated. This option is mutually
19323 exclusive with whole_len.
19324 whole_len
19326 Set the minimum total number of samples in the output audio stream.
19327 If the value is longer than the input audio length, silence is
19328 added to the end, until the value is reached. This option is
19329 mutually exclusive with pad_len.
19330 pad_dur
19332 Specify the duration of samples of silence to add. See the Time
19333 duration section in the ffmpeg-utils(1) manual for the accepted
19334 syntax. Used only if set to non-negative value.
19335 whole_dur
19337 Specify the minimum total duration in the output audio stream. See
19338 the Time duration section in the ffmpeg-utils(1) manual for the
19339 accepted syntax. Used only if set to non-negative value. If the
19340 value is longer than the input audio length, silence is added to
19341 the end, until the value is reached. This option is mutually
19342 exclusive with pad_dur
19343 If neither the pad_len nor the whole_len nor pad_dur nor whole_dur
19345 option is set, the filter will add silence to the end of the input
19346 stream indefinitely.
19347 Note that for ffmpeg 4.4 and earlier a zero pad_dur or whole_dur also
19349 caused the filter to add silence indefinitely.
19350 Examples
19352 • Add 1024 samples of silence to the end of the input:
19354 apad=pad_len=1024
19356 • Make sure the audio output will contain at least 10000 samples, pad
19358 the input with silence if required:
19359 apad=whole_len=10000
19361 • Use ffmpeg to pad the audio input with silence, so that the video
19363 stream will always result the shortest and will be converted until
19364 the end in the output file when using the shortest option:
19365 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
19367 aphaser
19369 Add a phasing effect to the input audio.
19370 A phaser filter creates series of peaks and troughs in the frequency
19372 spectrum. The position of the peaks and troughs are modulated so that
19373 they vary over time, creating a sweeping effect.
19374 A description of the accepted parameters follows.
19376 in_gain
19378 Set input gain. Default is 0.4.
19379 out_gain
19381 Set output gain. Default is 0.74
19382 delay
19384 Set delay in milliseconds. Default is 3.0.
19385 decay
19387 Set decay. Default is 0.4.
19388 speed
19390 Set modulation speed in Hz. Default is 0.5.
19391 type
19393 Set modulation type. Default is triangular.
19394 It accepts the following values:
19396 triangular, t
19398 sinusoidal, s
19399 aphaseshift
19401 Apply phase shift to input audio samples.
19402 The filter accepts the following options:
19404 shift
19406 Specify phase shift. Allowed range is from -1.0 to 1.0. Default
19407 value is 0.0.
19408 level
19410 Set output gain applied to final output. Allowed range is from 0.0
19411 to 1.0. Default value is 1.0.
19412 order
19414 Set filter order used for filtering. Allowed range is from 1 to 16.
19415 Default value is 8.
19416 Commands
19418 This filter supports the all above options as commands.
19420 apsyclip
19422 Apply Psychoacoustic clipper to input audio stream.
19423 The filter accepts the following options:
19425 level_in
19427 Set input gain. By default it is 1. Range is [0.015625 - 64].
19428 level_out
19430 Set output gain. By default it is 1. Range is [0.015625 - 64].
19431 clip
19433 Set the clipping start value. Default value is 0dBFS or 1.
19434 diff
19436 Output only difference samples, useful to hear introduced
19437 distortions. By default is disabled.
19438 adaptive
19440 Set strenght of adaptive distortion applied. Default value is 0.5.
19441 Allowed range is from 0 to 1.
19442 iterations
19444 Set number of iterations of psychoacoustic clipper. Allowed range
19445 is from 1 to 20. Default value is 10.
19446 level
19448 Auto level output signal. Default is disabled. This normalizes
19449 audio back to 0dBFS if enabled.
19450 Commands
19452 This filter supports the all above options as commands.
19454 apulsator
19456 Audio pulsator is something between an autopanner and a tremolo. But
19457 it can produce funny stereo effects as well. Pulsator changes the
19458 volume of the left and right channel based on a LFO (low frequency
19459 oscillator) with different waveforms and shifted phases. This filter
19460 have the ability to define an offset between left and right channel. An
19461 offset of 0 means that both LFO shapes match each other. The left and
19462 right channel are altered equally - a conventional tremolo. An offset
19463 of 50% means that the shape of the right channel is exactly shifted in
19464 phase (or moved backwards about half of the frequency) - pulsator acts
19465 as an autopanner. At 1 both curves match again. Every setting in
19466 between moves the phase shift gapless between all stages and produces
19467 some "bypassing" sounds with sine and triangle waveforms. The more you
19468 set the offset near 1 (starting from the 0.5) the faster the signal
19469 passes from the left to the right speaker.
19470 The filter accepts the following options:
19472 level_in
19474 Set input gain. By default it is 1. Range is [0.015625 - 64].
19475 level_out
19477 Set output gain. By default it is 1. Range is [0.015625 - 64].
19478 mode
19480 Set waveform shape the LFO will use. Can be one of: sine, triangle,
19481 square, sawup or sawdown. Default is sine.
19482 amount
19484 Set modulation. Define how much of original signal is affected by
19485 the LFO.
19486 offset_l
19488 Set left channel offset. Default is 0. Allowed range is [0 - 1].
19489 offset_r
19491 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
19492 width
19494 Set pulse width. Default is 1. Allowed range is [0 - 2].
19495 timing
19497 Set possible timing mode. Can be one of: bpm, ms or hz. Default is
19498 hz.
19499 bpm Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if
19501 timing is set to bpm.
19502 ms Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if
19504 timing is set to ms.
19505 hz Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100].
19507 Only used if timing is set to hz.
19508 aresample
19510 Resample the input audio to the specified parameters, using the
19511 libswresample library. If none are specified then the filter will
19512 automatically convert between its input and output.
19513 This filter is also able to stretch/squeeze the audio data to make it
19515 match the timestamps or to inject silence / cut out audio to make it
19516 match the timestamps, do a combination of both or do neither.
19517 The filter accepts the syntax [sample_rate:]resampler_options, where
19519 sample_rate expresses a sample rate and resampler_options is a list of
19520 key=value pairs, separated by ":". See the "Resampler Options" section
19521 in the ffmpeg-resampler(1) manual for the complete list of supported
19522 options.
19523 Examples
19525 • Resample the input audio to 44100Hz:
19527 aresample=44100
19529 • Stretch/squeeze samples to the given timestamps, with a maximum of
19531 1000 samples per second compensation:
19532 aresample=async=1000
19534 areverse
19536 Reverse an audio clip.
19537 Warning: This filter requires memory to buffer the entire clip, so
19539 trimming is suggested.
19540 Examples
19542 • Take the first 5 seconds of a clip, and reverse it.
19544 atrim=end=5,areverse
19546 arnndn
19548 Reduce noise from speech using Recurrent Neural Networks.
19549 This filter accepts the following options:
19551 model, m
19553 Set train model file to load. This option is always required.
19554 mix Set how much to mix filtered samples into final output. Allowed
19556 range is from -1 to 1. Default value is 1. Negative values are
19557 special, they set how much to keep filtered noise in the final
19558 filter output. Set this option to -1 to hear actual noise removed
19559 from input signal.
19560 Commands
19562 This filter supports the all above options as commands.
19564 asdr
19566 Measure Audio Signal-to-Distortion Ratio.
19567 This filter takes two audio streams for input, and outputs first audio
19569 stream. Results are in dB per channel at end of either input.
19570 asetnsamples
19572 Set the number of samples per each output audio frame.
19573 The last output packet may contain a different number of samples, as
19575 the filter will flush all the remaining samples when the input audio
19576 signals its end.
19577 The filter accepts the following options:
19579 nb_out_samples, n
19581 Set the number of frames per each output audio frame. The number is
19582 intended as the number of samples per each channel. Default value
19583 is 1024.
19584 pad, p
19586 If set to 1, the filter will pad the last audio frame with zeroes,
19587 so that the last frame will contain the same number of samples as
19588 the previous ones. Default value is 1.
19589 For example, to set the number of per-frame samples to 1234 and disable
19591 padding for the last frame, use:
19592 asetnsamples=n=1234:p=0
19594 asetrate
19596 Set the sample rate without altering the PCM data. This will result in
19597 a change of speed and pitch.
19598 The filter accepts the following options:
19600 sample_rate, r
19602 Set the output sample rate. Default is 44100 Hz.
19603 ashowinfo
19605 Show a line containing various information for each input audio frame.
19606 The input audio is not modified.
19607 The shown line contains a sequence of key/value pairs of the form
19609 key:value.
19610 The following values are shown in the output:
19612 n The (sequential) number of the input frame, starting from 0.
19614 pts The presentation timestamp of the input frame, in time base units;
19616 the time base depends on the filter input pad, and is usually
19617 1/sample_rate.
19618 pts_time
19620 The presentation timestamp of the input frame in seconds.
19621 pos position of the frame in the input stream, -1 if this information
19623 in unavailable and/or meaningless (for example in case of synthetic
19624 audio)
19625 fmt The sample format.
19627 chlayout
19629 The channel layout.
19630 rate
19632 The sample rate for the audio frame.
19633 nb_samples
19635 The number of samples (per channel) in the frame.
19636 checksum
19638 The Adler-32 checksum (printed in hexadecimal) of the audio data.
19639 For planar audio, the data is treated as if all the planes were
19640 concatenated.
19641 plane_checksums
19643 A list of Adler-32 checksums for each data plane.
19644 asoftclip
19646 Apply audio soft clipping.
19647 Soft clipping is a type of distortion effect where the amplitude of a
19649 signal is saturated along a smooth curve, rather than the abrupt shape
19650 of hard-clipping.
19651 This filter accepts the following options:
19653 type
19655 Set type of soft-clipping.
19656 It accepts the following values:
19658 hard
19660 tanh
19661 atan
19662 cubic
19663 exp
19664 alg
19665 quintic
19666 sin
19667 erf
19668 threshold
19669 Set threshold from where to start clipping. Default value is 0dB or
19670 1.
19671 output
19673 Set gain applied to output. Default value is 0dB or 1.
19674 param
19676 Set additional parameter which controls sigmoid function.
19677 oversample
19679 Set oversampling factor.
19680 Commands
19682 This filter supports the all above options as commands.
19684 aspectralstats
19686 Display frequency domain statistical information about the audio
19687 channels. Statistics are calculated and stored as metadata for each
19688 audio channel and for each audio frame.
19689 It accepts the following option:
19691 win_size
19693 Set the window length in samples. Default value is 2048. Allowed
19694 range is from 32 to 65536.
19695 win_func
19697 Set window function.
19698 It accepts the following values:
19700 rect
19702 bartlett
19703 hann, hanning
19704 hamming
19705 blackman
19706 welch
19707 flattop
19708 bharris
19709 bnuttall
19710 bhann
19711 sine
19712 nuttall
19713 lanczos
19714 gauss
19715 tukey
19716 dolph
19717 cauchy
19718 parzen
19719 poisson
19720 bohman
19721 Default is "hann".
19723 overlap
19725 Set window overlap. Allowed range is from 0 to 1. Default value is
19726 0.5.
19727 A list of each metadata key follows:
19729 mean
19731 variance
19732 centroid
19733 spread
19734 skewness
19735 kurtosis
19736 entropy
19737 flatness
19738 crest
19739 flux
19740 slope
19741 decrease
19742 rolloff
19743 asr
19745 Automatic Speech Recognition
19746 This filter uses PocketSphinx for speech recognition. To enable
19748 compilation of this filter, you need to configure FFmpeg with
19749 "--enable-pocketsphinx".
19750 It accepts the following options:
19752 rate
19754 Set sampling rate of input audio. Defaults is 16000. This need to
19755 match speech models, otherwise one will get poor results.
19756 hmm Set dictionary containing acoustic model files.
19758 dict
19760 Set pronunciation dictionary.
19761 lm Set language model file.
19763 lmctl
19765 Set language model set.
19766 lmname
19768 Set which language model to use.
19769 logfn
19771 Set output for log messages.
19772 The filter exports recognized speech as the frame metadata
19774 "lavfi.asr.text".
19775 astats
19777 Display time domain statistical information about the audio channels.
19778 Statistics are calculated and displayed for each audio channel and,
19779 where applicable, an overall figure is also given.
19780 It accepts the following option:
19782 length
19784 Short window length in seconds, used for peak and trough RMS
19785 measurement. Default is 0.05 (50 milliseconds). Allowed range is
19786 "[0 - 10]".
19787 metadata
19789 Set metadata injection. All the metadata keys are prefixed with
19790 "lavfi.astats.X", where "X" is channel number starting from 1 or
19791 string "Overall". Default is disabled.
19792 Available keys for each channel are: DC_offset Min_level Max_level
19794 Min_difference Max_difference Mean_difference RMS_difference
19795 Peak_level RMS_peak RMS_trough Crest_factor Flat_factor Peak_count
19796 Noise_floor Noise_floor_count Entropy Bit_depth Dynamic_range
19797 Zero_crossings Zero_crossings_rate Number_of_NaNs Number_of_Infs
19798 Number_of_denormals
19799 and for Overall: DC_offset Min_level Max_level Min_difference
19801 Max_difference Mean_difference RMS_difference Peak_level RMS_level
19802 RMS_peak RMS_trough Flat_factor Peak_count Noise_floor
19803 Noise_floor_count Entropy Bit_depth Number_of_samples
19804 Number_of_NaNs Number_of_Infs Number_of_denormals
19805 For example full key look like this "lavfi.astats.1.DC_offset" or
19807 this "lavfi.astats.Overall.Peak_count".
19808 For description what each key means read below.
19810 reset
19812 Set the number of frames over which cumulative stats are calculated
19813 before being reset Default is disabled.
19814 measure_perchannel
19816 Select the parameters which are measured per channel. The metadata
19817 keys can be used as flags, default is all which measures
19818 everything. none disables all per channel measurement.
19819 measure_overall
19821 Select the parameters which are measured overall. The metadata keys
19822 can be used as flags, default is all which measures everything.
19823 none disables all overall measurement.
19824 A description of each shown parameter follows:
19826 DC offset
19828 Mean amplitude displacement from zero.
19829 Min level
19831 Minimal sample level.
19832 Max level
19834 Maximal sample level.
19835 Min difference
19837 Minimal difference between two consecutive samples.
19838 Max difference
19840 Maximal difference between two consecutive samples.
19841 Mean difference
19843 Mean difference between two consecutive samples. The average of
19844 each difference between two consecutive samples.
19845 RMS difference
19847 Root Mean Square difference between two consecutive samples.
19848 Peak level dB
19850 RMS level dB
19851 Standard peak and RMS level measured in dBFS.
19852 RMS peak dB
19854 RMS trough dB
19855 Peak and trough values for RMS level measured over a short window.
19856 Crest factor
19858 Standard ratio of peak to RMS level (note: not in dB).
19859 Flat factor
19861 Flatness (i.e. consecutive samples with the same value) of the
19862 signal at its peak levels (i.e. either Min level or Max level).
19863 Peak count
19865 Number of occasions (not the number of samples) that the signal
19866 attained either Min level or Max level.
19867 Noise floor dB
19869 Minimum local peak measured in dBFS over a short window.
19870 Noise floor count
19872 Number of occasions (not the number of samples) that the signal
19873 attained Noise floor.
19874 Entropy
19876 Entropy measured across whole audio. Entropy of value near 1.0 is
19877 typically measured for white noise.
19878 Bit depth
19880 Overall bit depth of audio. Number of bits used for each sample.
19881 Dynamic range
19883 Measured dynamic range of audio in dB.
19884 Zero crossings
19886 Number of points where the waveform crosses the zero level axis.
19887 Zero crossings rate
19889 Rate of Zero crossings and number of audio samples.
19890 asubboost
19892 Boost subwoofer frequencies.
19893 The filter accepts the following options:
19895 dry Set dry gain, how much of original signal is kept. Allowed range is
19897 from 0 to 1. Default value is 0.7.
19898 wet Set wet gain, how much of filtered signal is kept. Allowed range is
19900 from 0 to 1. Default value is 0.7.
19901 decay
19903 Set delay line decay gain value. Allowed range is from 0 to 1.
19904 Default value is 0.7.
19905 feedback
19907 Set delay line feedback gain value. Allowed range is from 0 to 1.
19908 Default value is 0.9.
19909 cutoff
19911 Set cutoff frequency in Hertz. Allowed range is 50 to 900. Default
19912 value is 100.
19913 slope
19915 Set slope amount for cutoff frequency. Allowed range is 0.0001 to
19916 1. Default value is 0.5.
19917 delay
19919 Set delay. Allowed range is from 1 to 100. Default value is 20.
19920 Commands
19922 This filter supports the all above options as commands.
19924 asubcut
19926 Cut subwoofer frequencies.
19927 This filter allows to set custom, steeper roll off than highpass
19929 filter, and thus is able to more attenuate frequency content in stop-
19930 band.
19931 The filter accepts the following options:
19933 cutoff
19935 Set cutoff frequency in Hertz. Allowed range is 2 to 200. Default
19936 value is 20.
19937 order
19939 Set filter order. Available values are from 3 to 20. Default value
19940 is 10.
19941 level
19943 Set input gain level. Allowed range is from 0 to 1. Default value
19944 is 1.
19945 Commands
19947 This filter supports the all above options as commands.
19949 asupercut
19951 Cut super frequencies.
19952 The filter accepts the following options:
19954 cutoff
19956 Set cutoff frequency in Hertz. Allowed range is 20000 to 192000.
19957 Default value is 20000.
19958 order
19960 Set filter order. Available values are from 3 to 20. Default value
19961 is 10.
19962 level
19964 Set input gain level. Allowed range is from 0 to 1. Default value
19965 is 1.
19966 Commands
19968 This filter supports the all above options as commands.
19970 asuperpass
19972 Apply high order Butterworth band-pass filter.
19973 The filter accepts the following options:
19975 centerf
19977 Set center frequency in Hertz. Allowed range is 2 to 999999.
19978 Default value is 1000.
19979 order
19981 Set filter order. Available values are from 4 to 20. Default value
19982 is 4.
19983 qfactor
19985 Set Q-factor. Allowed range is from 0.01 to 100. Default value is
19986 1.
19987 level
19989 Set input gain level. Allowed range is from 0 to 2. Default value
19990 is 1.
19991 Commands
19993 This filter supports the all above options as commands.
19995 asuperstop
19997 Apply high order Butterworth band-stop filter.
19998 The filter accepts the following options:
20000 centerf
20002 Set center frequency in Hertz. Allowed range is 2 to 999999.
20003 Default value is 1000.
20004 order
20006 Set filter order. Available values are from 4 to 20. Default value
20007 is 4.
20008 qfactor
20010 Set Q-factor. Allowed range is from 0.01 to 100. Default value is
20011 1.
20012 level
20014 Set input gain level. Allowed range is from 0 to 2. Default value
20015 is 1.
20016 Commands
20018 This filter supports the all above options as commands.
20020 atempo
20022 Adjust audio tempo.
20023 The filter accepts exactly one parameter, the audio tempo. If not
20025 specified then the filter will assume nominal 1.0 tempo. Tempo must be
20026 in the [0.5, 100.0] range.
20027 Note that tempo greater than 2 will skip some samples rather than blend
20029 them in. If for any reason this is a concern it is always possible to
20030 daisy-chain several instances of atempo to achieve the desired product
20031 tempo.
20032 Examples
20034 • Slow down audio to 80% tempo:
20036 atempo=0.8
20038 • To speed up audio to 300% tempo:
20040 atempo=3
20042 • To speed up audio to 300% tempo by daisy-chaining two atempo
20044 instances:
20045 atempo=sqrt(3),atempo=sqrt(3)
20047 Commands
20049 This filter supports the following commands:
20051 tempo
20053 Change filter tempo scale factor. Syntax for the command is :
20054 "tempo"
20055 atilt
20057 Apply spectral tilt filter to audio stream.
20058 This filter apply any spectral roll-off slope over any specified
20060 frequency band.
20061 The filter accepts the following options:
20063 freq
20065 Set central frequency of tilt in Hz. Default is 10000 Hz.
20066 slope
20068 Set slope direction of tilt. Default is 0. Allowed range is from -1
20069 to 1.
20070 width
20072 Set width of tilt. Default is 1000. Allowed range is from 100 to
20073 10000.
20074 order
20076 Set order of tilt filter.
20077 level
20079 Set input volume level. Allowed range is from 0 to 4. Defalt is 1.
20080 Commands
20082 This filter supports the all above options as commands.
20084 atrim
20086 Trim the input so that the output contains one continuous subpart of
20087 the input.
20088 It accepts the following parameters:
20090 start
20092 Timestamp (in seconds) of the start of the section to keep. I.e.
20093 the audio sample with the timestamp start will be the first sample
20094 in the output.
20095 end Specify time of the first audio sample that will be dropped, i.e.
20097 the audio sample immediately preceding the one with the timestamp
20098 end will be the last sample in the output.
20099 start_pts
20101 Same as start, except this option sets the start timestamp in
20102 samples instead of seconds.
20103 end_pts
20105 Same as end, except this option sets the end timestamp in samples
20106 instead of seconds.
20107 duration
20109 The maximum duration of the output in seconds.
20110 start_sample
20112 The number of the first sample that should be output.
20113 end_sample
20115 The number of the first sample that should be dropped.
20116 start, end, and duration are expressed as time duration specifications;
20118 see the Time duration section in the ffmpeg-utils(1) manual.
20119 Note that the first two sets of the start/end options and the duration
20121 option look at the frame timestamp, while the _sample options simply
20122 count the samples that pass through the filter. So start/end_pts and
20123 start/end_sample will give different results when the timestamps are
20124 wrong, inexact or do not start at zero. Also note that this filter does
20125 not modify the timestamps. If you wish to have the output timestamps
20126 start at zero, insert the asetpts filter after the atrim filter.
20127 If multiple start or end options are set, this filter tries to be
20129 greedy and keep all samples that match at least one of the specified
20130 constraints. To keep only the part that matches all the constraints at
20131 once, chain multiple atrim filters.
20132 The defaults are such that all the input is kept. So it is possible to
20134 set e.g. just the end values to keep everything before the specified
20135 time.
20136 Examples:
20138 • Drop everything except the second minute of input:
20140 ffmpeg -i INPUT -af atrim=60:120
20142 • Keep only the first 1000 samples:
20144 ffmpeg -i INPUT -af atrim=end_sample=1000
20146 axcorrelate
20148 Calculate normalized windowed cross-correlation between two input audio
20149 streams.
20150 Resulted samples are always between -1 and 1 inclusive. If result is 1
20152 it means two input samples are highly correlated in that selected
20153 segment. Result 0 means they are not correlated at all. If result is
20154 -1 it means two input samples are out of phase, which means they cancel
20155 each other.
20156 The filter accepts the following options:
20158 size
20160 Set size of segment over which cross-correlation is calculated.
20161 Default is 256. Allowed range is from 2 to 131072.
20162 algo
20164 Set algorithm for cross-correlation. Can be "slow" or "fast".
20165 Default is "slow". Fast algorithm assumes mean values over any
20166 given segment are always zero and thus need much less calculations
20167 to make. This is generally not true, but is valid for typical
20168 audio streams.
20169 Examples
20171 • Calculate correlation between channels in stereo audio stream:
20173 ffmpeg -i stereo.wav -af channelsplit,axcorrelate=size=1024:algo=fast correlation.wav
20175 bandpass
20177 Apply a two-pole Butterworth band-pass filter with central frequency
20178 frequency, and (3dB-point) band-width width. The csg option selects a
20179 constant skirt gain (peak gain = Q) instead of the default: constant
20180 0dB peak gain. The filter roll off at 6dB per octave (20dB per
20181 decade).
20182 The filter accepts the following options:
20184 frequency, f
20186 Set the filter's central frequency. Default is 3000.
20187 csg Constant skirt gain if set to 1. Defaults to 0.
20189 width_type, t
20191 Set method to specify band-width of filter.
20192 h Hz
20194 q Q-Factor
20196 o octave
20198 s slope
20200 k kHz
20202 width, w
20204 Specify the band-width of a filter in width_type units.
20205 mix, m
20207 How much to use filtered signal in output. Default is 1. Range is
20208 between 0 and 1.
20209 channels, c
20211 Specify which channels to filter, by default all available are
20212 filtered.
20213 normalize, n
20215 Normalize biquad coefficients, by default is disabled. Enabling it
20216 will normalize magnitude response at DC to 0dB.
20217 transform, a
20219 Set transform type of IIR filter.
20220 di
20222 dii
20223 tdii
20224 latt
20225 svf
20226 precision, r
20227 Set precison of filtering.
20228 auto
20230 Pick automatic sample format depending on surround filters.
20231 s16 Always use signed 16-bit.
20233 s32 Always use signed 32-bit.
20235 f32 Always use float 32-bit.
20237 f64 Always use float 64-bit.
20239 Commands
20241 This filter supports the following commands:
20243 frequency, f
20245 Change bandpass frequency. Syntax for the command is : "frequency"
20246 width_type, t
20248 Change bandpass width_type. Syntax for the command is :
20249 "width_type"
20250 width, w
20252 Change bandpass width. Syntax for the command is : "width"
20253 mix, m
20255 Change bandpass mix. Syntax for the command is : "mix"
20256 bandreject
20258 Apply a two-pole Butterworth band-reject filter with central frequency
20259 frequency, and (3dB-point) band-width width. The filter roll off at
20260 6dB per octave (20dB per decade).
20261 The filter accepts the following options:
20263 frequency, f
20265 Set the filter's central frequency. Default is 3000.
20266 width_type, t
20268 Set method to specify band-width of filter.
20269 h Hz
20271 q Q-Factor
20273 o octave
20275 s slope
20277 k kHz
20279 width, w
20281 Specify the band-width of a filter in width_type units.
20282 mix, m
20284 How much to use filtered signal in output. Default is 1. Range is
20285 between 0 and 1.
20286 channels, c
20288 Specify which channels to filter, by default all available are
20289 filtered.
20290 normalize, n
20292 Normalize biquad coefficients, by default is disabled. Enabling it
20293 will normalize magnitude response at DC to 0dB.
20294 transform, a
20296 Set transform type of IIR filter.
20297 di
20299 dii
20300 tdii
20301 latt
20302 svf
20303 precision, r
20304 Set precison of filtering.
20305 auto
20307 Pick automatic sample format depending on surround filters.
20308 s16 Always use signed 16-bit.
20310 s32 Always use signed 32-bit.
20312 f32 Always use float 32-bit.
20314 f64 Always use float 64-bit.
20316 Commands
20318 This filter supports the following commands:
20320 frequency, f
20322 Change bandreject frequency. Syntax for the command is :
20323 "frequency"
20324 width_type, t
20326 Change bandreject width_type. Syntax for the command is :
20327 "width_type"
20328 width, w
20330 Change bandreject width. Syntax for the command is : "width"
20331 mix, m
20333 Change bandreject mix. Syntax for the command is : "mix"
20334 bass, lowshelf
20336 Boost or cut the bass (lower) frequencies of the audio using a two-pole
20337 shelving filter with a response similar to that of a standard hi-fi's
20338 tone-controls. This is also known as shelving equalisation (EQ).
20339 The filter accepts the following options:
20341 gain, g
20343 Give the gain at 0 Hz. Its useful range is about -20 (for a large
20344 cut) to +20 (for a large boost). Beware of clipping when using a
20345 positive gain.
20346 frequency, f
20348 Set the filter's central frequency and so can be used to extend or
20349 reduce the frequency range to be boosted or cut. The default value
20350 is 100 Hz.
20351 width_type, t
20353 Set method to specify band-width of filter.
20354 h Hz
20356 q Q-Factor
20358 o octave
20360 s slope
20362 k kHz
20364 width, w
20366 Determine how steep is the filter's shelf transition.
20367 poles, p
20369 Set number of poles. Default is 2.
20370 mix, m
20372 How much to use filtered signal in output. Default is 1. Range is
20373 between 0 and 1.
20374 channels, c
20376 Specify which channels to filter, by default all available are
20377 filtered.
20378 normalize, n
20380 Normalize biquad coefficients, by default is disabled. Enabling it
20381 will normalize magnitude response at DC to 0dB.
20382 transform, a
20384 Set transform type of IIR filter.
20385 di
20387 dii
20388 tdii
20389 latt
20390 svf
20391 precision, r
20392 Set precison of filtering.
20393 auto
20395 Pick automatic sample format depending on surround filters.
20396 s16 Always use signed 16-bit.
20398 s32 Always use signed 32-bit.
20400 f32 Always use float 32-bit.
20402 f64 Always use float 64-bit.
20404 Commands
20406 This filter supports the following commands:
20408 frequency, f
20410 Change bass frequency. Syntax for the command is : "frequency"
20411 width_type, t
20413 Change bass width_type. Syntax for the command is : "width_type"
20414 width, w
20416 Change bass width. Syntax for the command is : "width"
20417 gain, g
20419 Change bass gain. Syntax for the command is : "gain"
20420 mix, m
20422 Change bass mix. Syntax for the command is : "mix"
20423 biquad
20425 Apply a biquad IIR filter with the given coefficients. Where b0, b1,
20426 b2 and a0, a1, a2 are the numerator and denominator coefficients
20427 respectively. and channels, c specify which channels to filter, by
20428 default all available are filtered.
20429 Commands
20431 This filter supports the following commands:
20433 a0
20435 a1
20436 a2
20437 b0
20438 b1
20439 b2 Change biquad parameter. Syntax for the command is : "value"
20440 mix, m
20442 How much to use filtered signal in output. Default is 1. Range is
20443 between 0 and 1.
20444 channels, c
20446 Specify which channels to filter, by default all available are
20447 filtered.
20448 normalize, n
20450 Normalize biquad coefficients, by default is disabled. Enabling it
20451 will normalize magnitude response at DC to 0dB.
20452 transform, a
20454 Set transform type of IIR filter.
20455 di
20457 dii
20458 tdii
20459 latt
20460 svf
20461 precision, r
20462 Set precison of filtering.
20463 auto
20465 Pick automatic sample format depending on surround filters.
20466 s16 Always use signed 16-bit.
20468 s32 Always use signed 32-bit.
20470 f32 Always use float 32-bit.
20472 f64 Always use float 64-bit.
20474 bs2b
20476 Bauer stereo to binaural transformation, which improves headphone
20477 listening of stereo audio records.
20478 To enable compilation of this filter you need to configure FFmpeg with
20480 "--enable-libbs2b".
20481 It accepts the following parameters:
20483 profile
20485 Pre-defined crossfeed level.
20486 default
20488 Default level (fcut=700, feed=50).
20489 cmoy
20491 Chu Moy circuit (fcut=700, feed=60).
20492 jmeier
20494 Jan Meier circuit (fcut=650, feed=95).
20495 fcut
20497 Cut frequency (in Hz).
20498 feed
20500 Feed level (in Hz).
20501 channelmap
20503 Remap input channels to new locations.
20504 It accepts the following parameters:
20506 map Map channels from input to output. The argument is a '|'-separated
20508 list of mappings, each in the "in_channel-out_channel" or
20509 in_channel form. in_channel can be either the name of the input
20510 channel (e.g. FL for front left) or its index in the input channel
20511 layout. out_channel is the name of the output channel or its index
20512 in the output channel layout. If out_channel is not given then it
20513 is implicitly an index, starting with zero and increasing by one
20514 for each mapping.
20515 channel_layout
20517 The channel layout of the output stream.
20518 If no mapping is present, the filter will implicitly map input channels
20520 to output channels, preserving indices.
20521 Examples
20523 • For example, assuming a 5.1+downmix input MOV file,
20525 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
20527 will create an output WAV file tagged as stereo from the downmix
20529 channels of the input.
20530 • To fix a 5.1 WAV improperly encoded in AAC's native channel order
20532 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
20534 channelsplit
20536 Split each channel from an input audio stream into a separate output
20537 stream.
20538 It accepts the following parameters:
20540 channel_layout
20542 The channel layout of the input stream. The default is "stereo".
20543 channels
20545 A channel layout describing the channels to be extracted as
20546 separate output streams or "all" to extract each input channel as a
20547 separate stream. The default is "all".
20548 Choosing channels not present in channel layout in the input will
20550 result in an error.
20551 Examples
20553 • For example, assuming a stereo input MP3 file,
20555 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
20557 will create an output Matroska file with two audio streams, one
20559 containing only the left channel and the other the right channel.
20560 • Split a 5.1 WAV file into per-channel files:
20562 ffmpeg -i in.wav -filter_complex
20564 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
20565 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
20566 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
20567 side_right.wav
20568 • Extract only LFE from a 5.1 WAV file:
20570 ffmpeg -i in.wav -filter_complex 'channelsplit=channel_layout=5.1:channels=LFE[LFE]'
20572 -map '[LFE]' lfe.wav
20573 chorus
20575 Add a chorus effect to the audio.
20576 Can make a single vocal sound like a chorus, but can also be applied to
20578 instrumentation.
20579 Chorus resembles an echo effect with a short delay, but whereas with
20581 echo the delay is constant, with chorus, it is varied using using
20582 sinusoidal or triangular modulation. The modulation depth defines the
20583 range the modulated delay is played before or after the delay. Hence
20584 the delayed sound will sound slower or faster, that is the delayed
20585 sound tuned around the original one, like in a chorus where some vocals
20586 are slightly off key.
20587 It accepts the following parameters:
20589 in_gain
20591 Set input gain. Default is 0.4.
20592 out_gain
20594 Set output gain. Default is 0.4.
20595 delays
20597 Set delays. A typical delay is around 40ms to 60ms.
20598 decays
20600 Set decays.
20601 speeds
20603 Set speeds.
20604 depths
20606 Set depths.
20607 Examples
20609 • A single delay:
20611 chorus=0.7:0.9:55:0.4:0.25:2
20613 • Two delays:
20615 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
20617 • Fuller sounding chorus with three delays:
20619 chorus=0.5:0.9:50|60|40:0.4|0.32|0.3:0.25|0.4|0.3:2|2.3|1.3
20621 compand
20623 Compress or expand the audio's dynamic range.
20624 It accepts the following parameters:
20626 attacks
20628 decays
20629 A list of times in seconds for each channel over which the
20630 instantaneous level of the input signal is averaged to determine
20631 its volume. attacks refers to increase of volume and decays refers
20632 to decrease of volume. For most situations, the attack time
20633 (response to the audio getting louder) should be shorter than the
20634 decay time, because the human ear is more sensitive to sudden loud
20635 audio than sudden soft audio. A typical value for attack is 0.3
20636 seconds and a typical value for decay is 0.8 seconds. If specified
20637 number of attacks & decays is lower than number of channels, the
20638 last set attack/decay will be used for all remaining channels.
20639 points
20641 A list of points for the transfer function, specified in dB
20642 relative to the maximum possible signal amplitude. Each key points
20643 list must be defined using the following syntax:
20644 "x0/y0|x1/y1|x2/y2|...." or "x0/y0 x1/y1 x2/y2 ...."
20645 The input values must be in strictly increasing order but the
20647 transfer function does not have to be monotonically rising. The
20648 point "0/0" is assumed but may be overridden (by "0/out-dBn").
20649 Typical values for the transfer function are "-70/-70|-60/-20|1/0".
20650 soft-knee
20652 Set the curve radius in dB for all joints. It defaults to 0.01.
20653 gain
20655 Set the additional gain in dB to be applied at all points on the
20656 transfer function. This allows for easy adjustment of the overall
20657 gain. It defaults to 0.
20658 volume
20660 Set an initial volume, in dB, to be assumed for each channel when
20661 filtering starts. This permits the user to supply a nominal level
20662 initially, so that, for example, a very large gain is not applied
20663 to initial signal levels before the companding has begun to
20664 operate. A typical value for audio which is initially quiet is -90
20665 dB. It defaults to 0.
20666 delay
20668 Set a delay, in seconds. The input audio is analyzed immediately,
20669 but audio is delayed before being fed to the volume adjuster.
20670 Specifying a delay approximately equal to the attack/decay times
20671 allows the filter to effectively operate in predictive rather than
20672 reactive mode. It defaults to 0.
20673 Examples
20675 • Make music with both quiet and loud passages suitable for listening
20677 to in a noisy environment:
20678 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
20680 Another example for audio with whisper and explosion parts:
20682 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
20684 • A noise gate for when the noise is at a lower level than the
20686 signal:
20687 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
20689 • Here is another noise gate, this time for when the noise is at a
20691 higher level than the signal (making it, in some ways, similar to
20692 squelch):
20693 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
20695 • 2:1 compression starting at -6dB:
20697 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
20699 • 2:1 compression starting at -9dB:
20701 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
20703 • 2:1 compression starting at -12dB:
20705 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
20707 • 2:1 compression starting at -18dB:
20709 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
20711 • 3:1 compression starting at -15dB:
20713 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
20715 • Compressor/Gate:
20717 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
20719 • Expander:
20721 compand=attacks=0:points=-80/-169|-54/-80|-49.5/-64.6|-41.1/-41.1|-25.8/-15|-10.8/-4.5|0/0|20/8.3
20723 • Hard limiter at -6dB:
20725 compand=attacks=0:points=-80/-80|-6/-6|20/-6
20727 • Hard limiter at -12dB:
20729 compand=attacks=0:points=-80/-80|-12/-12|20/-12
20731 • Hard noise gate at -35 dB:
20733 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
20735 • Soft limiter:
20737 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
20739 compensationdelay
20741 Compensation Delay Line is a metric based delay to compensate differing
20742 positions of microphones or speakers.
20743 For example, you have recorded guitar with two microphones placed in
20745 different locations. Because the front of sound wave has fixed speed in
20746 normal conditions, the phasing of microphones can vary and depends on
20747 their location and interposition. The best sound mix can be achieved
20748 when these microphones are in phase (synchronized). Note that a
20749 distance of ~30 cm between microphones makes one microphone capture the
20750 signal in antiphase to the other microphone. That makes the final mix
20751 sound moody. This filter helps to solve phasing problems by adding
20752 different delays to each microphone track and make them synchronized.
20753 The best result can be reached when you take one track as base and
20755 synchronize other tracks one by one with it. Remember that
20756 synchronization/delay tolerance depends on sample rate, too. Higher
20757 sample rates will give more tolerance.
20758 The filter accepts the following parameters:
20760 mm Set millimeters distance. This is compensation distance for fine
20762 tuning. Default is 0.
20763 cm Set cm distance. This is compensation distance for tightening
20765 distance setup. Default is 0.
20766 m Set meters distance. This is compensation distance for hard
20768 distance setup. Default is 0.
20769 dry Set dry amount. Amount of unprocessed (dry) signal. Default is 0.
20771 wet Set wet amount. Amount of processed (wet) signal. Default is 1.
20773 temp
20775 Set temperature in degrees Celsius. This is the temperature of the
20776 environment. Default is 20.
20777 crossfeed
20779 Apply headphone crossfeed filter.
20780 Crossfeed is the process of blending the left and right channels of
20782 stereo audio recording. It is mainly used to reduce extreme stereo
20783 separation of low frequencies.
20784 The intent is to produce more speaker like sound to the listener.
20786 The filter accepts the following options:
20788 strength
20790 Set strength of crossfeed. Default is 0.2. Allowed range is from 0
20791 to 1. This sets gain of low shelf filter for side part of stereo
20792 image. Default is -6dB. Max allowed is -30db when strength is set
20793 to 1.
20794 range
20796 Set soundstage wideness. Default is 0.5. Allowed range is from 0 to
20797 1. This sets cut off frequency of low shelf filter. Default is cut
20798 off near 1550 Hz. With range set to 1 cut off frequency is set to
20799 2100 Hz.
20800 slope
20802 Set curve slope of low shelf filter. Default is 0.5. Allowed range
20803 is from 0.01 to 1.
20804 level_in
20806 Set input gain. Default is 0.9.
20807 level_out
20809 Set output gain. Default is 1.
20810 Commands
20812 This filter supports the all above options as commands.
20814 crystalizer
20816 Simple algorithm for audio noise sharpening.
20817 This filter linearly increases differences betweeen each audio sample.
20819 The filter accepts the following options:
20821 i Sets the intensity of effect (default: 2.0). Must be in range
20823 between -10.0 to 0 (unchanged sound) to 10.0 (maximum effect). To
20824 inverse filtering use negative value.
20825 c Enable clipping. By default is enabled.
20827 Commands
20829 This filter supports the all above options as commands.
20831 dcshift
20833 Apply a DC shift to the audio.
20834 This can be useful to remove a DC offset (caused perhaps by a hardware
20836 problem in the recording chain) from the audio. The effect of a DC
20837 offset is reduced headroom and hence volume. The astats filter can be
20838 used to determine if a signal has a DC offset.
20839 shift
20841 Set the DC shift, allowed range is [-1, 1]. It indicates the amount
20842 to shift the audio.
20843 limitergain
20845 Optional. It should have a value much less than 1 (e.g. 0.05 or
20846 0.02) and is used to prevent clipping.
20847 deesser
20849 Apply de-essing to the audio samples.
20850 i Set intensity for triggering de-essing. Allowed range is from 0 to
20852 1. Default is 0.
20853 m Set amount of ducking on treble part of sound. Allowed range is
20855 from 0 to 1. Default is 0.5.
20856 f How much of original frequency content to keep when de-essing.
20858 Allowed range is from 0 to 1. Default is 0.5.
20859 s Set the output mode.
20861 It accepts the following values:
20863 i Pass input unchanged.
20865 o Pass ess filtered out.
20867 e Pass only ess.
20869 Default value is o.
20871 drmeter
20873 Measure audio dynamic range.
20874 DR values of 14 and higher is found in very dynamic material. DR of 8
20876 to 13 is found in transition material. And anything less that 8 have
20877 very poor dynamics and is very compressed.
20878 The filter accepts the following options:
20880 length
20882 Set window length in seconds used to split audio into segments of
20883 equal length. Default is 3 seconds.
20884 dynaudnorm
20886 Dynamic Audio Normalizer.
20887 This filter applies a certain amount of gain to the input audio in
20889 order to bring its peak magnitude to a target level (e.g. 0 dBFS).
20890 However, in contrast to more "simple" normalization algorithms, the
20891 Dynamic Audio Normalizer *dynamically* re-adjusts the gain factor to
20892 the input audio. This allows for applying extra gain to the "quiet"
20893 sections of the audio while avoiding distortions or clipping the "loud"
20894 sections. In other words: The Dynamic Audio Normalizer will "even out"
20895 the volume of quiet and loud sections, in the sense that the volume of
20896 each section is brought to the same target level. Note, however, that
20897 the Dynamic Audio Normalizer achieves this goal *without* applying
20898 "dynamic range compressing". It will retain 100% of the dynamic range
20899 *within* each section of the audio file.
20900 framelen, f
20902 Set the frame length in milliseconds. In range from 10 to 8000
20903 milliseconds. Default is 500 milliseconds. The Dynamic Audio
20904 Normalizer processes the input audio in small chunks, referred to
20905 as frames. This is required, because a peak magnitude has no
20906 meaning for just a single sample value. Instead, we need to
20907 determine the peak magnitude for a contiguous sequence of sample
20908 values. While a "standard" normalizer would simply use the peak
20909 magnitude of the complete file, the Dynamic Audio Normalizer
20910 determines the peak magnitude individually for each frame. The
20911 length of a frame is specified in milliseconds. By default, the
20912 Dynamic Audio Normalizer uses a frame length of 500 milliseconds,
20913 which has been found to give good results with most files. Note
20914 that the exact frame length, in number of samples, will be
20915 determined automatically, based on the sampling rate of the
20916 individual input audio file.
20917 gausssize, g
20919 Set the Gaussian filter window size. In range from 3 to 301, must
20920 be odd number. Default is 31. Probably the most important
20921 parameter of the Dynamic Audio Normalizer is the "window size" of
20922 the Gaussian smoothing filter. The filter's window size is
20923 specified in frames, centered around the current frame. For the
20924 sake of simplicity, this must be an odd number. Consequently, the
20925 default value of 31 takes into account the current frame, as well
20926 as the 15 preceding frames and the 15 subsequent frames. Using a
20927 larger window results in a stronger smoothing effect and thus in
20928 less gain variation, i.e. slower gain adaptation. Conversely, using
20929 a smaller window results in a weaker smoothing effect and thus in
20930 more gain variation, i.e. faster gain adaptation. In other words,
20931 the more you increase this value, the more the Dynamic Audio
20932 Normalizer will behave like a "traditional" normalization filter.
20933 On the contrary, the more you decrease this value, the more the
20934 Dynamic Audio Normalizer will behave like a dynamic range
20935 compressor.
20936 peak, p
20938 Set the target peak value. This specifies the highest permissible
20939 magnitude level for the normalized audio input. This filter will
20940 try to approach the target peak magnitude as closely as possible,
20941 but at the same time it also makes sure that the normalized signal
20942 will never exceed the peak magnitude. A frame's maximum local gain
20943 factor is imposed directly by the target peak magnitude. The
20944 default value is 0.95 and thus leaves a headroom of 5%*. It is not
20945 recommended to go above this value.
20946 maxgain, m
20948 Set the maximum gain factor. In range from 1.0 to 100.0. Default is
20949 10.0. The Dynamic Audio Normalizer determines the maximum possible
20950 (local) gain factor for each input frame, i.e. the maximum gain
20951 factor that does not result in clipping or distortion. The maximum
20952 gain factor is determined by the frame's highest magnitude sample.
20953 However, the Dynamic Audio Normalizer additionally bounds the
20954 frame's maximum gain factor by a predetermined (global) maximum
20955 gain factor. This is done in order to avoid excessive gain factors
20956 in "silent" or almost silent frames. By default, the maximum gain
20957 factor is 10.0, For most inputs the default value should be
20958 sufficient and it usually is not recommended to increase this
20959 value. Though, for input with an extremely low overall volume
20960 level, it may be necessary to allow even higher gain factors. Note,
20961 however, that the Dynamic Audio Normalizer does not simply apply a
20962 "hard" threshold (i.e. cut off values above the threshold).
20963 Instead, a "sigmoid" threshold function will be applied. This way,
20964 the gain factors will smoothly approach the threshold value, but
20965 never exceed that value.
20966 targetrms, r
20968 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 -
20969 disabled. By default, the Dynamic Audio Normalizer performs "peak"
20970 normalization. This means that the maximum local gain factor for
20971 each frame is defined (only) by the frame's highest magnitude
20972 sample. This way, the samples can be amplified as much as possible
20973 without exceeding the maximum signal level, i.e. without clipping.
20974 Optionally, however, the Dynamic Audio Normalizer can also take
20975 into account the frame's root mean square, abbreviated RMS. In
20976 electrical engineering, the RMS is commonly used to determine the
20977 power of a time-varying signal. It is therefore considered that the
20978 RMS is a better approximation of the "perceived loudness" than just
20979 looking at the signal's peak magnitude. Consequently, by adjusting
20980 all frames to a constant RMS value, a uniform "perceived loudness"
20981 can be established. If a target RMS value has been specified, a
20982 frame's local gain factor is defined as the factor that would
20983 result in exactly that RMS value. Note, however, that the maximum
20984 local gain factor is still restricted by the frame's highest
20985 magnitude sample, in order to prevent clipping.
20986 coupling, n
20988 Enable channels coupling. By default is enabled. By default, the
20989 Dynamic Audio Normalizer will amplify all channels by the same
20990 amount. This means the same gain factor will be applied to all
20991 channels, i.e. the maximum possible gain factor is determined by
20992 the "loudest" channel. However, in some recordings, it may happen
20993 that the volume of the different channels is uneven, e.g. one
20994 channel may be "quieter" than the other one(s). In this case, this
20995 option can be used to disable the channel coupling. This way, the
20996 gain factor will be determined independently for each channel,
20997 depending only on the individual channel's highest magnitude
20998 sample. This allows for harmonizing the volume of the different
20999 channels.
21000 correctdc, c
21002 Enable DC bias correction. By default is disabled. An audio signal
21003 (in the time domain) is a sequence of sample values. In the
21004 Dynamic Audio Normalizer these sample values are represented in the
21005 -1.0 to 1.0 range, regardless of the original input format.
21006 Normally, the audio signal, or "waveform", should be centered
21007 around the zero point. That means if we calculate the mean value
21008 of all samples in a file, or in a single frame, then the result
21009 should be 0.0 or at least very close to that value. If, however,
21010 there is a significant deviation of the mean value from 0.0, in
21011 either positive or negative direction, this is referred to as a DC
21012 bias or DC offset. Since a DC bias is clearly undesirable, the
21013 Dynamic Audio Normalizer provides optional DC bias correction.
21014 With DC bias correction enabled, the Dynamic Audio Normalizer will
21015 determine the mean value, or "DC correction" offset, of each input
21016 frame and subtract that value from all of the frame's sample values
21017 which ensures those samples are centered around 0.0 again. Also, in
21018 order to avoid "gaps" at the frame boundaries, the DC correction
21019 offset values will be interpolated smoothly between neighbouring
21020 frames.
21021 altboundary, b
21023 Enable alternative boundary mode. By default is disabled. The
21024 Dynamic Audio Normalizer takes into account a certain neighbourhood
21025 around each frame. This includes the preceding frames as well as
21026 the subsequent frames. However, for the "boundary" frames, located
21027 at the very beginning and at the very end of the audio file, not
21028 all neighbouring frames are available. In particular, for the first
21029 few frames in the audio file, the preceding frames are not known.
21030 And, similarly, for the last few frames in the audio file, the
21031 subsequent frames are not known. Thus, the question arises which
21032 gain factors should be assumed for the missing frames in the
21033 "boundary" region. The Dynamic Audio Normalizer implements two
21034 modes to deal with this situation. The default boundary mode
21035 assumes a gain factor of exactly 1.0 for the missing frames,
21036 resulting in a smooth "fade in" and "fade out" at the beginning and
21037 at the end of the input, respectively.
21038 compress, s
21040 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
21041 By default, the Dynamic Audio Normalizer does not apply
21042 "traditional" compression. This means that signal peaks will not be
21043 pruned and thus the full dynamic range will be retained within each
21044 local neighbourhood. However, in some cases it may be desirable to
21045 combine the Dynamic Audio Normalizer's normalization algorithm with
21046 a more "traditional" compression. For this purpose, the Dynamic
21047 Audio Normalizer provides an optional compression (thresholding)
21048 function. If (and only if) the compression feature is enabled, all
21049 input frames will be processed by a soft knee thresholding function
21050 prior to the actual normalization process. Put simply, the
21051 thresholding function is going to prune all samples whose magnitude
21052 exceeds a certain threshold value. However, the Dynamic Audio
21053 Normalizer does not simply apply a fixed threshold value. Instead,
21054 the threshold value will be adjusted for each individual frame. In
21055 general, smaller parameters result in stronger compression, and
21056 vice versa. Values below 3.0 are not recommended, because audible
21057 distortion may appear.
21058 threshold, t
21060 Set the target threshold value. This specifies the lowest
21061 permissible magnitude level for the audio input which will be
21062 normalized. If input frame volume is above this value frame will
21063 be normalized. Otherwise frame may not be normalized at all. The
21064 default value is set to 0, which means all input frames will be
21065 normalized. This option is mostly useful if digital noise is not
21066 wanted to be amplified.
21067 Commands
21069 This filter supports the all above options as commands.
21071 earwax
21073 Make audio easier to listen to on headphones.
21074 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
21076 so that when listened to on headphones the stereo image is moved from
21077 inside your head (standard for headphones) to outside and in front of
21078 the listener (standard for speakers).
21079 Ported from SoX.
21081 equalizer
21083 Apply a two-pole peaking equalisation (EQ) filter. With this filter,
21084 the signal-level at and around a selected frequency can be increased or
21085 decreased, whilst (unlike bandpass and bandreject filters) that at all
21086 other frequencies is unchanged.
21087 In order to produce complex equalisation curves, this filter can be
21089 given several times, each with a different central frequency.
21090 The filter accepts the following options:
21092 frequency, f
21094 Set the filter's central frequency in Hz.
21095 width_type, t
21097 Set method to specify band-width of filter.
21098 h Hz
21100 q Q-Factor
21102 o octave
21104 s slope
21106 k kHz
21108 width, w
21110 Specify the band-width of a filter in width_type units.
21111 gain, g
21113 Set the required gain or attenuation in dB. Beware of clipping
21114 when using a positive gain.
21115 mix, m
21117 How much to use filtered signal in output. Default is 1. Range is
21118 between 0 and 1.
21119 channels, c
21121 Specify which channels to filter, by default all available are
21122 filtered.
21123 normalize, n
21125 Normalize biquad coefficients, by default is disabled. Enabling it
21126 will normalize magnitude response at DC to 0dB.
21127 transform, a
21129 Set transform type of IIR filter.
21130 di
21132 dii
21133 tdii
21134 latt
21135 svf
21136 precision, r
21137 Set precison of filtering.
21138 auto
21140 Pick automatic sample format depending on surround filters.
21141 s16 Always use signed 16-bit.
21143 s32 Always use signed 32-bit.
21145 f32 Always use float 32-bit.
21147 f64 Always use float 64-bit.
21149 Examples
21151 • Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
21153 equalizer=f=1000:t=h:width=200:g=-10
21155 • Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz
21157 with Q 2:
21158 equalizer=f=1000:t=q:w=1:g=2,equalizer=f=100:t=q:w=2:g=-5
21160 Commands
21162 This filter supports the following commands:
21164 frequency, f
21166 Change equalizer frequency. Syntax for the command is :
21167 "frequency"
21168 width_type, t
21170 Change equalizer width_type. Syntax for the command is :
21171 "width_type"
21172 width, w
21174 Change equalizer width. Syntax for the command is : "width"
21175 gain, g
21177 Change equalizer gain. Syntax for the command is : "gain"
21178 mix, m
21180 Change equalizer mix. Syntax for the command is : "mix"
21181 extrastereo
21183 Linearly increases the difference between left and right channels which
21184 adds some sort of "live" effect to playback.
21185 The filter accepts the following options:
21187 m Sets the difference coefficient (default: 2.5). 0.0 means mono
21189 sound (average of both channels), with 1.0 sound will be unchanged,
21190 with -1.0 left and right channels will be swapped.
21191 c Enable clipping. By default is enabled.
21193 Commands
21195 This filter supports the all above options as commands.
21197 firequalizer
21199 Apply FIR Equalization using arbitrary frequency response.
21200 The filter accepts the following option:
21202 gain
21204 Set gain curve equation (in dB). The expression can contain
21205 variables:
21206 f the evaluated frequency
21208 sr sample rate
21210 ch channel number, set to 0 when multichannels evaluation is
21212 disabled
21213 chid
21215 channel id, see libavutil/channel_layout.h, set to the first
21216 channel id when multichannels evaluation is disabled
21217 chs number of channels
21219 chlayout
21221 channel_layout, see libavutil/channel_layout.h
21222 and functions:
21224 gain_interpolate(f)
21226 interpolate gain on frequency f based on gain_entry
21227 cubic_interpolate(f)
21229 same as gain_interpolate, but smoother
21230 This option is also available as command. Default is
21232 gain_interpolate(f).
21233 gain_entry
21235 Set gain entry for gain_interpolate function. The expression can
21236 contain functions:
21237 entry(f, g)
21239 store gain entry at frequency f with value g
21240 This option is also available as command.
21242 delay
21244 Set filter delay in seconds. Higher value means more accurate.
21245 Default is 0.01.
21246 accuracy
21248 Set filter accuracy in Hz. Lower value means more accurate.
21249 Default is 5.
21250 wfunc
21252 Set window function. Acceptable values are:
21253 rectangular
21255 rectangular window, useful when gain curve is already smooth
21256 hann
21258 hann window (default)
21259 hamming
21261 hamming window
21262 blackman
21264 blackman window
21265 nuttall3
21267 3-terms continuous 1st derivative nuttall window
21268 mnuttall3
21270 minimum 3-terms discontinuous nuttall window
21271 nuttall
21273 4-terms continuous 1st derivative nuttall window
21274 bnuttall
21276 minimum 4-terms discontinuous nuttall (blackman-nuttall) window
21277 bharris
21279 blackman-harris window
21280 tukey
21282 tukey window
21283 fixed
21285 If enabled, use fixed number of audio samples. This improves speed
21286 when filtering with large delay. Default is disabled.
21287 multi
21289 Enable multichannels evaluation on gain. Default is disabled.
21290 zero_phase
21292 Enable zero phase mode by subtracting timestamp to compensate
21293 delay. Default is disabled.
21294 scale
21296 Set scale used by gain. Acceptable values are:
21297 linlin
21299 linear frequency, linear gain
21300 linlog
21302 linear frequency, logarithmic (in dB) gain (default)
21303 loglin
21305 logarithmic (in octave scale where 20 Hz is 0) frequency,
21306 linear gain
21307 loglog
21309 logarithmic frequency, logarithmic gain
21310 dumpfile
21312 Set file for dumping, suitable for gnuplot.
21313 dumpscale
21315 Set scale for dumpfile. Acceptable values are same with scale
21316 option. Default is linlog.
21317 fft2
21319 Enable 2-channel convolution using complex FFT. This improves speed
21320 significantly. Default is disabled.
21321 min_phase
21323 Enable minimum phase impulse response. Default is disabled.
21324 Examples
21326 • lowpass at 1000 Hz:
21328 firequalizer=gain='if(lt(f,1000), 0, -INF)'
21330 • lowpass at 1000 Hz with gain_entry:
21332 firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
21334 • custom equalization:
21336 firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
21338 • higher delay with zero phase to compensate delay:
21340 firequalizer=delay=0.1:fixed=on:zero_phase=on
21342 • lowpass on left channel, highpass on right channel:
21344 firequalizer=gain='if(eq(chid,1), gain_interpolate(f), if(eq(chid,2), gain_interpolate(1e6+f), 0))'
21346 :gain_entry='entry(1000, 0); entry(1001,-INF); entry(1e6+1000,0)':multi=on
21347 flanger
21349 Apply a flanging effect to the audio.
21350 The filter accepts the following options:
21352 delay
21354 Set base delay in milliseconds. Range from 0 to 30. Default value
21355 is 0.
21356 depth
21358 Set added sweep delay in milliseconds. Range from 0 to 10. Default
21359 value is 2.
21360 regen
21362 Set percentage regeneration (delayed signal feedback). Range from
21363 -95 to 95. Default value is 0.
21364 width
21366 Set percentage of delayed signal mixed with original. Range from 0
21367 to 100. Default value is 71.
21368 speed
21370 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is
21371 0.5.
21372 shape
21374 Set swept wave shape, can be triangular or sinusoidal. Default
21375 value is sinusoidal.
21376 phase
21378 Set swept wave percentage-shift for multi channel. Range from 0 to
21379 100. Default value is 25.
21380 interp
21382 Set delay-line interpolation, linear or quadratic. Default is
21383 linear.
21384 haas
21386 Apply Haas effect to audio.
21387 Note that this makes most sense to apply on mono signals. With this
21389 filter applied to mono signals it give some directionality and
21390 stretches its stereo image.
21391 The filter accepts the following options:
21393 level_in
21395 Set input level. By default is 1, or 0dB
21396 level_out
21398 Set output level. By default is 1, or 0dB.
21399 side_gain
21401 Set gain applied to side part of signal. By default is 1.
21402 middle_source
21404 Set kind of middle source. Can be one of the following:
21405 left
21407 Pick left channel.
21408 right
21410 Pick right channel.
21411 mid Pick middle part signal of stereo image.
21413 side
21415 Pick side part signal of stereo image.
21416 middle_phase
21418 Change middle phase. By default is disabled.
21419 left_delay
21421 Set left channel delay. By default is 2.05 milliseconds.
21422 left_balance
21424 Set left channel balance. By default is -1.
21425 left_gain
21427 Set left channel gain. By default is 1.
21428 left_phase
21430 Change left phase. By default is disabled.
21431 right_delay
21433 Set right channel delay. By defaults is 2.12 milliseconds.
21434 right_balance
21436 Set right channel balance. By default is 1.
21437 right_gain
21439 Set right channel gain. By default is 1.
21440 right_phase
21442 Change right phase. By default is enabled.
21443 hdcd
21445 Decodes High Definition Compatible Digital (HDCD) data. A 16-bit PCM
21446 stream with embedded HDCD codes is expanded into a 20-bit PCM stream.
21447 The filter supports the Peak Extend and Low-level Gain Adjustment
21449 features of HDCD, and detects the Transient Filter flag.
21450 ffmpeg -i HDCD16.flac -af hdcd OUT24.flac
21452 When using the filter with wav, note the default encoding for wav is
21454 16-bit, so the resulting 20-bit stream will be truncated back to
21455 16-bit. Use something like -acodec pcm_s24le after the filter to get
21456 24-bit PCM output.
21457 ffmpeg -i HDCD16.wav -af hdcd OUT16.wav
21459 ffmpeg -i HDCD16.wav -af hdcd -c:a pcm_s24le OUT24.wav
21460 The filter accepts the following options:
21462 disable_autoconvert
21464 Disable any automatic format conversion or resampling in the filter
21465 graph.
21466 process_stereo
21468 Process the stereo channels together. If target_gain does not match
21469 between channels, consider it invalid and use the last valid
21470 target_gain.
21471 cdt_ms
21473 Set the code detect timer period in ms.
21474 force_pe
21476 Always extend peaks above -3dBFS even if PE isn't signaled.
21477 analyze_mode
21479 Replace audio with a solid tone and adjust the amplitude to signal
21480 some specific aspect of the decoding process. The output file can
21481 be loaded in an audio editor alongside the original to aid
21482 analysis.
21483 "analyze_mode=pe:force_pe=true" can be used to see all samples
21485 above the PE level.
21486 Modes are:
21488 0, off
21490 Disabled
21491 1, lle
21493 Gain adjustment level at each sample
21494 2, pe
21496 Samples where peak extend occurs
21497 3, cdt
21499 Samples where the code detect timer is active
21500 4, tgm
21502 Samples where the target gain does not match between channels
21503 headphone
21505 Apply head-related transfer functions (HRTFs) to create virtual
21506 loudspeakers around the user for binaural listening via headphones.
21507 The HRIRs are provided via additional streams, for each channel one
21508 stereo input stream is needed.
21509 The filter accepts the following options:
21511 map Set mapping of input streams for convolution. The argument is a
21513 '|'-separated list of channel names in order as they are given as
21514 additional stream inputs for filter. This also specify number of
21515 input streams. Number of input streams must be not less than number
21516 of channels in first stream plus one.
21517 gain
21519 Set gain applied to audio. Value is in dB. Default is 0.
21520 type
21522 Set processing type. Can be time or freq. time is processing audio
21523 in time domain which is slow. freq is processing audio in
21524 frequency domain which is fast. Default is freq.
21525 lfe Set custom gain for LFE channels. Value is in dB. Default is 0.
21527 size
21529 Set size of frame in number of samples which will be processed at
21530 once. Default value is 1024. Allowed range is from 1024 to 96000.
21531 hrir
21533 Set format of hrir stream. Default value is stereo. Alternative
21534 value is multich. If value is set to stereo, number of additional
21535 streams should be greater or equal to number of input channels in
21536 first input stream. Also each additional stream should have stereo
21537 number of channels. If value is set to multich, number of
21538 additional streams should be exactly one. Also number of input
21539 channels of additional stream should be equal or greater than twice
21540 number of channels of first input stream.
21541 Examples
21543 • Full example using wav files as coefficients with amovie filters
21545 for 7.1 downmix, each amovie filter use stereo file with IR
21546 coefficients as input. The files give coefficients for each
21547 position of virtual loudspeaker:
21548 ffmpeg -i input.wav
21550 -filter_complex "amovie=azi_270_ele_0_DFC.wav[sr];amovie=azi_90_ele_0_DFC.wav[sl];amovie=azi_225_ele_0_DFC.wav[br];amovie=azi_135_ele_0_DFC.wav[bl];amovie=azi_0_ele_0_DFC.wav,asplit[fc][lfe];amovie=azi_35_ele_0_DFC.wav[fl];amovie=azi_325_ele_0_DFC.wav[fr];[0:a][fl][fr][fc][lfe][bl][br][sl][sr]headphone=FL|FR|FC|LFE|BL|BR|SL|SR"
21551 output.wav
21552 • Full example using wav files as coefficients with amovie filters
21554 for 7.1 downmix, but now in multich hrir format.
21555 ffmpeg -i input.wav -filter_complex "amovie=minp.wav[hrirs];[0:a][hrirs]headphone=map=FL|FR|FC|LFE|BL|BR|SL|SR:hrir=multich"
21557 output.wav
21558 highpass
21560 Apply a high-pass filter with 3dB point frequency. The filter can be
21561 either single-pole, or double-pole (the default). The filter roll off
21562 at 6dB per pole per octave (20dB per pole per decade).
21563 The filter accepts the following options:
21565 frequency, f
21567 Set frequency in Hz. Default is 3000.
21568 poles, p
21570 Set number of poles. Default is 2.
21571 width_type, t
21573 Set method to specify band-width of filter.
21574 h Hz
21576 q Q-Factor
21578 o octave
21580 s slope
21582 k kHz
21584 width, w
21586 Specify the band-width of a filter in width_type units. Applies
21587 only to double-pole filter. The default is 0.707q and gives a
21588 Butterworth response.
21589 mix, m
21591 How much to use filtered signal in output. Default is 1. Range is
21592 between 0 and 1.
21593 channels, c
21595 Specify which channels to filter, by default all available are
21596 filtered.
21597 normalize, n
21599 Normalize biquad coefficients, by default is disabled. Enabling it
21600 will normalize magnitude response at DC to 0dB.
21601 transform, a
21603 Set transform type of IIR filter.
21604 di
21606 dii
21607 tdii
21608 latt
21609 svf
21610 precision, r
21611 Set precison of filtering.
21612 auto
21614 Pick automatic sample format depending on surround filters.
21615 s16 Always use signed 16-bit.
21617 s32 Always use signed 32-bit.
21619 f32 Always use float 32-bit.
21621 f64 Always use float 64-bit.
21623 Commands
21625 This filter supports the following commands:
21627 frequency, f
21629 Change highpass frequency. Syntax for the command is : "frequency"
21630 width_type, t
21632 Change highpass width_type. Syntax for the command is :
21633 "width_type"
21634 width, w
21636 Change highpass width. Syntax for the command is : "width"
21637 mix, m
21639 Change highpass mix. Syntax for the command is : "mix"
21640 join
21642 Join multiple input streams into one multi-channel stream.
21643 It accepts the following parameters:
21645 inputs
21647 The number of input streams. It defaults to 2.
21648 channel_layout
21650 The desired output channel layout. It defaults to stereo.
21651 map Map channels from inputs to output. The argument is a '|'-separated
21653 list of mappings, each in the "input_idx.in_channel-out_channel"
21654 form. input_idx is the 0-based index of the input stream.
21655 in_channel can be either the name of the input channel (e.g. FL for
21656 front left) or its index in the specified input stream. out_channel
21657 is the name of the output channel.
21658 The filter will attempt to guess the mappings when they are not
21660 specified explicitly. It does so by first trying to find an unused
21661 matching input channel and if that fails it picks the first unused
21662 input channel.
21663 Join 3 inputs (with properly set channel layouts):
21665 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
21667 Build a 5.1 output from 6 single-channel streams:
21669 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
21671 'join=inputs=6:channel_layout=5.1:map=0.0-FL|1.0-FR|2.0-FC|3.0-SL|4.0-SR|5.0-LFE'
21672 out
21673 ladspa
21675 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
21676 To enable compilation of this filter you need to configure FFmpeg with
21678 "--enable-ladspa".
21679 file, f
21681 Specifies the name of LADSPA plugin library to load. If the
21682 environment variable LADSPA_PATH is defined, the LADSPA plugin is
21683 searched in each one of the directories specified by the colon
21684 separated list in LADSPA_PATH, otherwise in the standard LADSPA
21685 paths, which are in this order: HOME/.ladspa/lib/,
21686 /usr/local/lib/ladspa/, /usr/lib/ladspa/.
21687 plugin, p
21689 Specifies the plugin within the library. Some libraries contain
21690 only one plugin, but others contain many of them. If this is not
21691 set filter will list all available plugins within the specified
21692 library.
21693 controls, c
21695 Set the '|' separated list of controls which are zero or more
21696 floating point values that determine the behavior of the loaded
21697 plugin (for example delay, threshold or gain). Controls need to be
21698 defined using the following syntax:
21699 c0=value0|c1=value1|c2=value2|..., where valuei is the value set on
21700 the i-th control. Alternatively they can be also defined using the
21701 following syntax: value0|value1|value2|..., where valuei is the
21702 value set on the i-th control. If controls is set to "help", all
21703 available controls and their valid ranges are printed.
21704 sample_rate, s
21706 Specify the sample rate, default to 44100. Only used if plugin have
21707 zero inputs.
21708 nb_samples, n
21710 Set the number of samples per channel per each output frame,
21711 default is 1024. Only used if plugin have zero inputs.
21712 duration, d
21714 Set the minimum duration of the sourced audio. See the Time
21715 duration section in the ffmpeg-utils(1) manual for the accepted
21716 syntax. Note that the resulting duration may be greater than the
21717 specified duration, as the generated audio is always cut at the end
21718 of a complete frame. If not specified, or the expressed duration
21719 is negative, the audio is supposed to be generated forever. Only
21720 used if plugin have zero inputs.
21721 latency, l
21723 Enable latency compensation, by default is disabled. Only used if
21724 plugin have inputs.
21725 Examples
21727 • List all available plugins within amp (LADSPA example plugin)
21729 library:
21730 ladspa=file=amp
21732 • List all available controls and their valid ranges for "vcf_notch"
21734 plugin from "VCF" library:
21735 ladspa=f=vcf:p=vcf_notch:c=help
21737 • Simulate low quality audio equipment using "Computer Music Toolkit"
21739 (CMT) plugin library:
21740 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
21742 • Add reverberation to the audio using TAP-plugins (Tom's Audio
21744 Processing plugins):
21745 ladspa=file=tap_reverb:tap_reverb
21747 • Generate white noise, with 0.2 amplitude:
21749 ladspa=file=cmt:noise_source_white:c=c0=.2
21751 • Generate 20 bpm clicks using plugin "C* Click - Metronome" from the
21753 "C* Audio Plugin Suite" (CAPS) library:
21754 ladspa=file=caps:Click:c=c1=20'
21756 • Apply "C* Eq10X2 - Stereo 10-band equaliser" effect:
21758 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
21760 • Increase volume by 20dB using fast lookahead limiter from Steve
21762 Harris "SWH Plugins" collection:
21763 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
21765 • Attenuate low frequencies using Multiband EQ from Steve Harris "SWH
21767 Plugins" collection:
21768 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
21770 • Reduce stereo image using "Narrower" from the "C* Audio Plugin
21772 Suite" (CAPS) library:
21773 ladspa=caps:Narrower
21775 • Another white noise, now using "C* Audio Plugin Suite" (CAPS)
21777 library:
21778 ladspa=caps:White:.2
21780 • Some fractal noise, using "C* Audio Plugin Suite" (CAPS) library:
21782 ladspa=caps:Fractal:c=c1=1
21784 • Dynamic volume normalization using "VLevel" plugin:
21786 ladspa=vlevel-ladspa:vlevel_mono
21788 Commands
21790 This filter supports the following commands:
21792 cN Modify the N-th control value.
21794 If the specified value is not valid, it is ignored and prior one is
21796 kept.
21797 loudnorm
21799 EBU R128 loudness normalization. Includes both dynamic and linear
21800 normalization modes. Support for both single pass (livestreams, files)
21801 and double pass (files) modes. This algorithm can target IL, LRA, and
21802 maximum true peak. In dynamic mode, to accurately detect true peaks,
21803 the audio stream will be upsampled to 192 kHz. Use the "-ar" option or
21804 "aresample" filter to explicitly set an output sample rate.
21805 The filter accepts the following options:
21807 I, i
21809 Set integrated loudness target. Range is -70.0 - -5.0. Default
21810 value is -24.0.
21811 LRA, lra
21813 Set loudness range target. Range is 1.0 - 20.0. Default value is
21814 7.0.
21815 TP, tp
21817 Set maximum true peak. Range is -9.0 - +0.0. Default value is
21818 -2.0.
21819 measured_I, measured_i
21821 Measured IL of input file. Range is -99.0 - +0.0.
21822 measured_LRA, measured_lra
21824 Measured LRA of input file. Range is 0.0 - 99.0.
21825 measured_TP, measured_tp
21827 Measured true peak of input file. Range is -99.0 - +99.0.
21828 measured_thresh
21830 Measured threshold of input file. Range is -99.0 - +0.0.
21831 offset
21833 Set offset gain. Gain is applied before the true-peak limiter.
21834 Range is -99.0 - +99.0. Default is +0.0.
21835 linear
21837 Normalize by linearly scaling the source audio. "measured_I",
21838 "measured_LRA", "measured_TP", and "measured_thresh" must all be
21839 specified. Target LRA shouldn't be lower than source LRA and the
21840 change in integrated loudness shouldn't result in a true peak which
21841 exceeds the target TP. If any of these conditions aren't met,
21842 normalization mode will revert to dynamic. Options are "true" or
21843 "false". Default is "true".
21844 dual_mono
21846 Treat mono input files as "dual-mono". If a mono file is intended
21847 for playback on a stereo system, its EBU R128 measurement will be
21848 perceptually incorrect. If set to "true", this option will
21849 compensate for this effect. Multi-channel input files are not
21850 affected by this option. Options are true or false. Default is
21851 false.
21852 print_format
21854 Set print format for stats. Options are summary, json, or none.
21855 Default value is none.
21856 lowpass
21858 Apply a low-pass filter with 3dB point frequency. The filter can be
21859 either single-pole or double-pole (the default). The filter roll off
21860 at 6dB per pole per octave (20dB per pole per decade).
21861 The filter accepts the following options:
21863 frequency, f
21865 Set frequency in Hz. Default is 500.
21866 poles, p
21868 Set number of poles. Default is 2.
21869 width_type, t
21871 Set method to specify band-width of filter.
21872 h Hz
21874 q Q-Factor
21876 o octave
21878 s slope
21880 k kHz
21882 width, w
21884 Specify the band-width of a filter in width_type units. Applies
21885 only to double-pole filter. The default is 0.707q and gives a
21886 Butterworth response.
21887 mix, m
21889 How much to use filtered signal in output. Default is 1. Range is
21890 between 0 and 1.
21891 channels, c
21893 Specify which channels to filter, by default all available are
21894 filtered.
21895 normalize, n
21897 Normalize biquad coefficients, by default is disabled. Enabling it
21898 will normalize magnitude response at DC to 0dB.
21899 transform, a
21901 Set transform type of IIR filter.
21902 di
21904 dii
21905 tdii
21906 latt
21907 svf
21908 precision, r
21909 Set precison of filtering.
21910 auto
21912 Pick automatic sample format depending on surround filters.
21913 s16 Always use signed 16-bit.
21915 s32 Always use signed 32-bit.
21917 f32 Always use float 32-bit.
21919 f64 Always use float 64-bit.
21921 Examples
21923 • Lowpass only LFE channel, it LFE is not present it does nothing:
21925 lowpass=c=LFE
21927 Commands
21929 This filter supports the following commands:
21931 frequency, f
21933 Change lowpass frequency. Syntax for the command is : "frequency"
21934 width_type, t
21936 Change lowpass width_type. Syntax for the command is :
21937 "width_type"
21938 width, w
21940 Change lowpass width. Syntax for the command is : "width"
21941 mix, m
21943 Change lowpass mix. Syntax for the command is : "mix"
21944 lv2
21946 Load a LV2 (LADSPA Version 2) plugin.
21947 To enable compilation of this filter you need to configure FFmpeg with
21949 "--enable-lv2".
21950 plugin, p
21952 Specifies the plugin URI. You may need to escape ':'.
21953 controls, c
21955 Set the '|' separated list of controls which are zero or more
21956 floating point values that determine the behavior of the loaded
21957 plugin (for example delay, threshold or gain). If controls is set
21958 to "help", all available controls and their valid ranges are
21959 printed.
21960 sample_rate, s
21962 Specify the sample rate, default to 44100. Only used if plugin have
21963 zero inputs.
21964 nb_samples, n
21966 Set the number of samples per channel per each output frame,
21967 default is 1024. Only used if plugin have zero inputs.
21968 duration, d
21970 Set the minimum duration of the sourced audio. See the Time
21971 duration section in the ffmpeg-utils(1) manual for the accepted
21972 syntax. Note that the resulting duration may be greater than the
21973 specified duration, as the generated audio is always cut at the end
21974 of a complete frame. If not specified, or the expressed duration
21975 is negative, the audio is supposed to be generated forever. Only
21976 used if plugin have zero inputs.
21977 Examples
21979 • Apply bass enhancer plugin from Calf:
21981 lv2=p=http\\\\://calf.sourceforge.net/plugins/BassEnhancer:c=amount=2
21983 • Apply vinyl plugin from Calf:
21985 lv2=p=http\\\\://calf.sourceforge.net/plugins/Vinyl:c=drone=0.2|aging=0.5
21987 • Apply bit crusher plugin from ArtyFX:
21989 lv2=p=http\\\\://www.openavproductions.com/artyfx#bitta:c=crush=0.3
21991 mcompand
21993 Multiband Compress or expand the audio's dynamic range.
21994 The input audio is divided into bands using 4th order Linkwitz-Riley
21996 IIRs. This is akin to the crossover of a loudspeaker, and results in
21997 flat frequency response when absent compander action.
21998 It accepts the following parameters:
22000 args
22002 This option syntax is: attack,decay,[attack,decay..] soft-knee
22003 points crossover_frequency [delay [initial_volume [gain]]] |
22004 attack,decay ... For explanation of each item refer to compand
22005 filter documentation.
22006 pan
22008 Mix channels with specific gain levels. The filter accepts the output
22009 channel layout followed by a set of channels definitions.
22010 This filter is also designed to efficiently remap the channels of an
22012 audio stream.
22013 The filter accepts parameters of the form: "l|outdef|outdef|..."
22015 l output channel layout or number of channels
22017 outdef
22019 output channel specification, of the form:
22020 "out_name=[gain*]in_name[(+-)[gain*]in_name...]"
22021 out_name
22023 output channel to define, either a channel name (FL, FR, etc.) or a
22024 channel number (c0, c1, etc.)
22025 gain
22027 multiplicative coefficient for the channel, 1 leaving the volume
22028 unchanged
22029 in_name
22031 input channel to use, see out_name for details; it is not possible
22032 to mix named and numbered input channels
22033 If the `=' in a channel specification is replaced by `<', then the
22035 gains for that specification will be renormalized so that the total is
22036 1, thus avoiding clipping noise.
22037 Mixing examples
22039 For example, if you want to down-mix from stereo to mono, but with a
22041 bigger factor for the left channel:
22042 pan=1c|c0=0.9*c0+0.1*c1
22044 A customized down-mix to stereo that works automatically for 3-, 4-, 5-
22046 and 7-channels surround:
22047 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
22049 Note that ffmpeg integrates a default down-mix (and up-mix) system that
22051 should be preferred (see "-ac" option) unless you have very specific
22052 needs.
22053 Remapping examples
22055 The channel remapping will be effective if, and only if:
22057 *<gain coefficients are zeroes or ones,>
22059 *<only one input per channel output,>
22060 If all these conditions are satisfied, the filter will notify the user
22062 ("Pure channel mapping detected"), and use an optimized and lossless
22063 method to do the remapping.
22064 For example, if you have a 5.1 source and want a stereo audio stream by
22066 dropping the extra channels:
22067 pan="stereo| c0=FL | c1=FR"
22069 Given the same source, you can also switch front left and front right
22071 channels and keep the input channel layout:
22072 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
22074 If the input is a stereo audio stream, you can mute the front left
22076 channel (and still keep the stereo channel layout) with:
22077 pan="stereo|c1=c1"
22079 Still with a stereo audio stream input, you can copy the right channel
22081 in both front left and right:
22082 pan="stereo| c0=FR | c1=FR"
22084 replaygain
22086 ReplayGain scanner filter. This filter takes an audio stream as an
22087 input and outputs it unchanged. At end of filtering it displays
22088 "track_gain" and "track_peak".
22089 resample
22091 Convert the audio sample format, sample rate and channel layout. It is
22092 not meant to be used directly.
22093 rubberband
22095 Apply time-stretching and pitch-shifting with librubberband.
22096 To enable compilation of this filter, you need to configure FFmpeg with
22098 "--enable-librubberband".
22099 The filter accepts the following options:
22101 tempo
22103 Set tempo scale factor.
22104 pitch
22106 Set pitch scale factor.
22107 transients
22109 Set transients detector. Possible values are:
22110 crisp
22112 mixed
22113 smooth
22114 detector
22115 Set detector. Possible values are:
22116 compound
22118 percussive
22119 soft
22120 phase
22121 Set phase. Possible values are:
22122 laminar
22124 independent
22125 window
22126 Set processing window size. Possible values are:
22127 standard
22129 short
22130 long
22131 smoothing
22132 Set smoothing. Possible values are:
22133 off
22135 on
22136 formant
22137 Enable formant preservation when shift pitching. Possible values
22138 are:
22139 shifted
22141 preserved
22142 pitchq
22143 Set pitch quality. Possible values are:
22144 quality
22146 speed
22147 consistency
22148 channels
22149 Set channels. Possible values are:
22150 apart
22152 together
22153 Commands
22155 This filter supports the following commands:
22157 tempo
22159 Change filter tempo scale factor. Syntax for the command is :
22160 "tempo"
22161 pitch
22163 Change filter pitch scale factor. Syntax for the command is :
22164 "pitch"
22165 sidechaincompress
22167 This filter acts like normal compressor but has the ability to compress
22168 detected signal using second input signal. It needs two input streams
22169 and returns one output stream. First input stream will be processed
22170 depending on second stream signal. The filtered signal then can be
22171 filtered with other filters in later stages of processing. See pan and
22172 amerge filter.
22173 The filter accepts the following options:
22175 level_in
22177 Set input gain. Default is 1. Range is between 0.015625 and 64.
22178 mode
22180 Set mode of compressor operation. Can be "upward" or "downward".
22181 Default is "downward".
22182 threshold
22184 If a signal of second stream raises above this level it will affect
22185 the gain reduction of first stream. By default is 0.125. Range is
22186 between 0.00097563 and 1.
22187 ratio
22189 Set a ratio about which the signal is reduced. 1:2 means that if
22190 the level raised 4dB above the threshold, it will be only 2dB above
22191 after the reduction. Default is 2. Range is between 1 and 20.
22192 attack
22194 Amount of milliseconds the signal has to rise above the threshold
22195 before gain reduction starts. Default is 20. Range is between 0.01
22196 and 2000.
22197 release
22199 Amount of milliseconds the signal has to fall below the threshold
22200 before reduction is decreased again. Default is 250. Range is
22201 between 0.01 and 9000.
22202 makeup
22204 Set the amount by how much signal will be amplified after
22205 processing. Default is 1. Range is from 1 to 64.
22206 knee
22208 Curve the sharp knee around the threshold to enter gain reduction
22209 more softly. Default is 2.82843. Range is between 1 and 8.
22210 link
22212 Choose if the "average" level between all channels of side-chain
22213 stream or the louder("maximum") channel of side-chain stream
22214 affects the reduction. Default is "average".
22215 detection
22217 Should the exact signal be taken in case of "peak" or an RMS one in
22218 case of "rms". Default is "rms" which is mainly smoother.
22219 level_sc
22221 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
22222 mix How much to use compressed signal in output. Default is 1. Range
22224 is between 0 and 1.
22225 Commands
22227 This filter supports the all above options as commands.
22229 Examples
22231 • Full ffmpeg example taking 2 audio inputs, 1st input to be
22233 compressed depending on the signal of 2nd input and later
22234 compressed signal to be merged with 2nd input:
22235 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
22237 sidechaingate
22239 A sidechain gate acts like a normal (wideband) gate but has the ability
22240 to filter the detected signal before sending it to the gain reduction
22241 stage. Normally a gate uses the full range signal to detect a level
22242 above the threshold. For example: If you cut all lower frequencies
22243 from your sidechain signal the gate will decrease the volume of your
22244 track only if not enough highs appear. With this technique you are able
22245 to reduce the resonation of a natural drum or remove "rumbling" of
22246 muted strokes from a heavily distorted guitar. It needs two input
22247 streams and returns one output stream. First input stream will be
22248 processed depending on second stream signal.
22249 The filter accepts the following options:
22251 level_in
22253 Set input level before filtering. Default is 1. Allowed range is
22254 from 0.015625 to 64.
22255 mode
22257 Set the mode of operation. Can be "upward" or "downward". Default
22258 is "downward". If set to "upward" mode, higher parts of signal will
22259 be amplified, expanding dynamic range in upward direction.
22260 Otherwise, in case of "downward" lower parts of signal will be
22261 reduced.
22262 range
22264 Set the level of gain reduction when the signal is below the
22265 threshold. Default is 0.06125. Allowed range is from 0 to 1.
22266 Setting this to 0 disables reduction and then filter behaves like
22267 expander.
22268 threshold
22270 If a signal rises above this level the gain reduction is released.
22271 Default is 0.125. Allowed range is from 0 to 1.
22272 ratio
22274 Set a ratio about which the signal is reduced. Default is 2.
22275 Allowed range is from 1 to 9000.
22276 attack
22278 Amount of milliseconds the signal has to rise above the threshold
22279 before gain reduction stops. Default is 20 milliseconds. Allowed
22280 range is from 0.01 to 9000.
22281 release
22283 Amount of milliseconds the signal has to fall below the threshold
22284 before the reduction is increased again. Default is 250
22285 milliseconds. Allowed range is from 0.01 to 9000.
22286 makeup
22288 Set amount of amplification of signal after processing. Default is
22289 1. Allowed range is from 1 to 64.
22290 knee
22292 Curve the sharp knee around the threshold to enter gain reduction
22293 more softly. Default is 2.828427125. Allowed range is from 1 to 8.
22294 detection
22296 Choose if exact signal should be taken for detection or an RMS like
22297 one. Default is rms. Can be peak or rms.
22298 link
22300 Choose if the average level between all channels or the louder
22301 channel affects the reduction. Default is average. Can be average
22302 or maximum.
22303 level_sc
22305 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
22306 Commands
22308 This filter supports the all above options as commands.
22310 silencedetect
22312 Detect silence in an audio stream.
22313 This filter logs a message when it detects that the input audio volume
22315 is less or equal to a noise tolerance value for a duration greater or
22316 equal to the minimum detected noise duration.
22317 The printed times and duration are expressed in seconds. The
22319 "lavfi.silence_start" or "lavfi.silence_start.X" metadata key is set on
22320 the first frame whose timestamp equals or exceeds the detection
22321 duration and it contains the timestamp of the first frame of the
22322 silence.
22323 The "lavfi.silence_duration" or "lavfi.silence_duration.X" and
22325 "lavfi.silence_end" or "lavfi.silence_end.X" metadata keys are set on
22326 the first frame after the silence. If mono is enabled, and each channel
22327 is evaluated separately, the ".X" suffixed keys are used, and "X"
22328 corresponds to the channel number.
22329 The filter accepts the following options:
22331 noise, n
22333 Set noise tolerance. Can be specified in dB (in case "dB" is
22334 appended to the specified value) or amplitude ratio. Default is
22335 -60dB, or 0.001.
22336 duration, d
22338 Set silence duration until notification (default is 2 seconds). See
22339 the Time duration section in the ffmpeg-utils(1) manual for the
22340 accepted syntax.
22341 mono, m
22343 Process each channel separately, instead of combined. By default is
22344 disabled.
22345 Examples
22347 • Detect 5 seconds of silence with -50dB noise tolerance:
22349 silencedetect=n=-50dB:d=5
22351 • Complete example with ffmpeg to detect silence with 0.0001 noise
22353 tolerance in silence.mp3:
22354 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
22356 silenceremove
22358 Remove silence from the beginning, middle or end of the audio.
22359 The filter accepts the following options:
22361 start_periods
22363 This value is used to indicate if audio should be trimmed at
22364 beginning of the audio. A value of zero indicates no silence should
22365 be trimmed from the beginning. When specifying a non-zero value, it
22366 trims audio up until it finds non-silence. Normally, when trimming
22367 silence from beginning of audio the start_periods will be 1 but it
22368 can be increased to higher values to trim all audio up to specific
22369 count of non-silence periods. Default value is 0.
22370 start_duration
22372 Specify the amount of time that non-silence must be detected before
22373 it stops trimming audio. By increasing the duration, bursts of
22374 noises can be treated as silence and trimmed off. Default value is
22375 0.
22376 start_threshold
22378 This indicates what sample value should be treated as silence. For
22379 digital audio, a value of 0 may be fine but for audio recorded from
22380 analog, you may wish to increase the value to account for
22381 background noise. Can be specified in dB (in case "dB" is appended
22382 to the specified value) or amplitude ratio. Default value is 0.
22383 start_silence
22385 Specify max duration of silence at beginning that will be kept
22386 after trimming. Default is 0, which is equal to trimming all
22387 samples detected as silence.
22388 start_mode
22390 Specify mode of detection of silence end in start of multi-channel
22391 audio. Can be any or all. Default is any. With any, any sample
22392 that is detected as non-silence will cause stopped trimming of
22393 silence. With all, only if all channels are detected as non-
22394 silence will cause stopped trimming of silence.
22395 stop_periods
22397 Set the count for trimming silence from the end of audio. To
22398 remove silence from the middle of a file, specify a stop_periods
22399 that is negative. This value is then treated as a positive value
22400 and is used to indicate the effect should restart processing as
22401 specified by start_periods, making it suitable for removing periods
22402 of silence in the middle of the audio. Default value is 0.
22403 stop_duration
22405 Specify a duration of silence that must exist before audio is not
22406 copied any more. By specifying a higher duration, silence that is
22407 wanted can be left in the audio. Default value is 0.
22408 stop_threshold
22410 This is the same as start_threshold but for trimming silence from
22411 the end of audio. Can be specified in dB (in case "dB" is appended
22412 to the specified value) or amplitude ratio. Default value is 0.
22413 stop_silence
22415 Specify max duration of silence at end that will be kept after
22416 trimming. Default is 0, which is equal to trimming all samples
22417 detected as silence.
22418 stop_mode
22420 Specify mode of detection of silence start in end of multi-channel
22421 audio. Can be any or all. Default is any. With any, any sample
22422 that is detected as non-silence will cause stopped trimming of
22423 silence. With all, only if all channels are detected as non-
22424 silence will cause stopped trimming of silence.
22425 detection
22427 Set how is silence detected. Can be "rms" or "peak". Second is
22428 faster and works better with digital silence which is exactly 0.
22429 Default value is "rms".
22430 window
22432 Set duration in number of seconds used to calculate size of window
22433 in number of samples for detecting silence. Default value is 0.02.
22434 Allowed range is from 0 to 10.
22435 Examples
22437 • The following example shows how this filter can be used to start a
22439 recording that does not contain the delay at the start which
22440 usually occurs between pressing the record button and the start of
22441 the performance:
22442 silenceremove=start_periods=1:start_duration=5:start_threshold=0.02
22444 • Trim all silence encountered from beginning to end where there is
22446 more than 1 second of silence in audio:
22447 silenceremove=stop_periods=-1:stop_duration=1:stop_threshold=-90dB
22449 • Trim all digital silence samples, using peak detection, from
22451 beginning to end where there is more than 0 samples of digital
22452 silence in audio and digital silence is detected in all channels at
22453 same positions in stream:
22454 silenceremove=window=0:detection=peak:stop_mode=all:start_mode=all:stop_periods=-1:stop_threshold=0
22456 sofalizer
22458 SOFAlizer uses head-related transfer functions (HRTFs) to create
22459 virtual loudspeakers around the user for binaural listening via
22460 headphones (audio formats up to 9 channels supported). The HRTFs are
22461 stored in SOFA files (see <http://www.sofacoustics.org/> for a
22462 database). SOFAlizer is developed at the Acoustics Research Institute
22463 (ARI) of the Austrian Academy of Sciences.
22464 To enable compilation of this filter you need to configure FFmpeg with
22466 "--enable-libmysofa".
22467 The filter accepts the following options:
22469 sofa
22471 Set the SOFA file used for rendering.
22472 gain
22474 Set gain applied to audio. Value is in dB. Default is 0.
22475 rotation
22477 Set rotation of virtual loudspeakers in deg. Default is 0.
22478 elevation
22480 Set elevation of virtual speakers in deg. Default is 0.
22481 radius
22483 Set distance in meters between loudspeakers and the listener with
22484 near-field HRTFs. Default is 1.
22485 type
22487 Set processing type. Can be time or freq. time is processing audio
22488 in time domain which is slow. freq is processing audio in
22489 frequency domain which is fast. Default is freq.
22490 speakers
22492 Set custom positions of virtual loudspeakers. Syntax for this
22493 option is: <CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...]. Each
22494 virtual loudspeaker is described with short channel name following
22495 with azimuth and elevation in degrees. Each virtual loudspeaker
22496 description is separated by '|'. For example to override front
22497 left and front right channel positions use: 'speakers=FL 45 15|FR
22498 345 15'. Descriptions with unrecognised channel names are ignored.
22499 lfegain
22501 Set custom gain for LFE channels. Value is in dB. Default is 0.
22502 framesize
22504 Set custom frame size in number of samples. Default is 1024.
22505 Allowed range is from 1024 to 96000. Only used if option type is
22506 set to freq.
22507 normalize
22509 Should all IRs be normalized upon importing SOFA file. By default
22510 is enabled.
22511 interpolate
22513 Should nearest IRs be interpolated with neighbor IRs if exact
22514 position does not match. By default is disabled.
22515 minphase
22517 Minphase all IRs upon loading of SOFA file. By default is disabled.
22518 anglestep
22520 Set neighbor search angle step. Only used if option interpolate is
22521 enabled.
22522 radstep
22524 Set neighbor search radius step. Only used if option interpolate is
22525 enabled.
22526 Examples
22528 • Using ClubFritz6 sofa file:
22530 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
22532 • Using ClubFritz12 sofa file and bigger radius with small rotation:
22534 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
22536 • Similar as above but with custom speaker positions for front left,
22538 front right, back left and back right and also with custom gain:
22539 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|BL 135|BR 225:gain=28"
22541 speechnorm
22543 Speech Normalizer.
22544 This filter expands or compresses each half-cycle of audio samples
22546 (local set of samples all above or all below zero and between two
22547 nearest zero crossings) depending on threshold value, so audio reaches
22548 target peak value under conditions controlled by below options.
22549 The filter accepts the following options:
22551 peak, p
22553 Set the expansion target peak value. This specifies the highest
22554 allowed absolute amplitude level for the normalized audio input.
22555 Default value is 0.95. Allowed range is from 0.0 to 1.0.
22556 expansion, e
22558 Set the maximum expansion factor. Allowed range is from 1.0 to
22559 50.0. Default value is 2.0. This option controls maximum local
22560 half-cycle of samples expansion. The maximum expansion would be
22561 such that local peak value reaches target peak value but never to
22562 surpass it and that ratio between new and previous peak value does
22563 not surpass this option value.
22564 compression, c
22566 Set the maximum compression factor. Allowed range is from 1.0 to
22567 50.0. Default value is 2.0. This option controls maximum local
22568 half-cycle of samples compression. This option is used only if
22569 threshold option is set to value greater than 0.0, then in such
22570 cases when local peak is lower or same as value set by threshold
22571 all samples belonging to that peak's half-cycle will be compressed
22572 by current compression factor.
22573 threshold, t
22575 Set the threshold value. Default value is 0.0. Allowed range is
22576 from 0.0 to 1.0. This option specifies which half-cycles of
22577 samples will be compressed and which will be expanded. Any half-
22578 cycle samples with their local peak value below or same as this
22579 option value will be compressed by current compression factor,
22580 otherwise, if greater than threshold value they will be expanded
22581 with expansion factor so that it could reach peak target value but
22582 never surpass it.
22583 raise, r
22585 Set the expansion raising amount per each half-cycle of samples.
22586 Default value is 0.001. Allowed range is from 0.0 to 1.0. This
22587 controls how fast expansion factor is raised per each new half-
22588 cycle until it reaches expansion value. Setting this options too
22589 high may lead to distortions.
22590 fall, f
22592 Set the compression raising amount per each half-cycle of samples.
22593 Default value is 0.001. Allowed range is from 0.0 to 1.0. This
22594 controls how fast compression factor is raised per each new half-
22595 cycle until it reaches compression value.
22596 channels, h
22598 Specify which channels to filter, by default all available channels
22599 are filtered.
22600 invert, i
22602 Enable inverted filtering, by default is disabled. This inverts
22603 interpretation of threshold option. When enabled any half-cycle of
22604 samples with their local peak value below or same as threshold
22605 option will be expanded otherwise it will be compressed.
22606 link, l
22608 Link channels when calculating gain applied to each filtered
22609 channel sample, by default is disabled. When disabled each
22610 filtered channel gain calculation is independent, otherwise when
22611 this option is enabled the minimum of all possible gains for each
22612 filtered channel is used.
22613 Commands
22615 This filter supports the all above options as commands.
22617 stereotools
22619 This filter has some handy utilities to manage stereo signals, for
22620 converting M/S stereo recordings to L/R signal while having control
22621 over the parameters or spreading the stereo image of master track.
22622 The filter accepts the following options:
22624 level_in
22626 Set input level before filtering for both channels. Defaults is 1.
22627 Allowed range is from 0.015625 to 64.
22628 level_out
22630 Set output level after filtering for both channels. Defaults is 1.
22631 Allowed range is from 0.015625 to 64.
22632 balance_in
22634 Set input balance between both channels. Default is 0. Allowed
22635 range is from -1 to 1.
22636 balance_out
22638 Set output balance between both channels. Default is 0. Allowed
22639 range is from -1 to 1.
22640 softclip
22642 Enable softclipping. Results in analog distortion instead of harsh
22643 digital 0dB clipping. Disabled by default.
22644 mutel
22646 Mute the left channel. Disabled by default.
22647 muter
22649 Mute the right channel. Disabled by default.
22650 phasel
22652 Change the phase of the left channel. Disabled by default.
22653 phaser
22655 Change the phase of the right channel. Disabled by default.
22656 mode
22658 Set stereo mode. Available values are:
22659 lr>lr
22661 Left/Right to Left/Right, this is default.
22662 lr>ms
22664 Left/Right to Mid/Side.
22665 ms>lr
22667 Mid/Side to Left/Right.
22668 lr>ll
22670 Left/Right to Left/Left.
22671 lr>rr
22673 Left/Right to Right/Right.
22674 lr>l+r
22676 Left/Right to Left + Right.
22677 lr>rl
22679 Left/Right to Right/Left.
22680 ms>ll
22682 Mid/Side to Left/Left.
22683 ms>rr
22685 Mid/Side to Right/Right.
22686 ms>rl
22688 Mid/Side to Right/Left.
22689 lr>l-r
22691 Left/Right to Left - Right.
22692 slev
22694 Set level of side signal. Default is 1. Allowed range is from
22695 0.015625 to 64.
22696 sbal
22698 Set balance of side signal. Default is 0. Allowed range is from -1
22699 to 1.
22700 mlev
22702 Set level of the middle signal. Default is 1. Allowed range is
22703 from 0.015625 to 64.
22704 mpan
22706 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
22707 base
22709 Set stereo base between mono and inversed channels. Default is 0.
22710 Allowed range is from -1 to 1.
22711 delay
22713 Set delay in milliseconds how much to delay left from right channel
22714 and vice versa. Default is 0. Allowed range is from -20 to 20.
22715 sclevel
22717 Set S/C level. Default is 1. Allowed range is from 1 to 100.
22718 phase
22720 Set the stereo phase in degrees. Default is 0. Allowed range is
22721 from 0 to 360.
22722 bmode_in, bmode_out
22724 Set balance mode for balance_in/balance_out option.
22725 Can be one of the following:
22727 balance
22729 Classic balance mode. Attenuate one channel at time. Gain is
22730 raised up to 1.
22731 amplitude
22733 Similar as classic mode above but gain is raised up to 2.
22734 power
22736 Equal power distribution, from -6dB to +6dB range.
22737 Commands
22739 This filter supports the all above options as commands.
22741 Examples
22743 • Apply karaoke like effect:
22745 stereotools=mlev=0.015625
22747 • Convert M/S signal to L/R:
22749 "stereotools=mode=ms>lr"
22751 stereowiden
22753 This filter enhance the stereo effect by suppressing signal common to
22754 both channels and by delaying the signal of left into right and vice
22755 versa, thereby widening the stereo effect.
22756 The filter accepts the following options:
22758 delay
22760 Time in milliseconds of the delay of left signal into right and
22761 vice versa. Default is 20 milliseconds.
22762 feedback
22764 Amount of gain in delayed signal into right and vice versa. Gives a
22765 delay effect of left signal in right output and vice versa which
22766 gives widening effect. Default is 0.3.
22767 crossfeed
22769 Cross feed of left into right with inverted phase. This helps in
22770 suppressing the mono. If the value is 1 it will cancel all the
22771 signal common to both channels. Default is 0.3.
22772 drymix
22774 Set level of input signal of original channel. Default is 0.8.
22775 Commands
22777 This filter supports the all above options except "delay" as commands.
22779 superequalizer
22781 Apply 18 band equalizer.
22782 The filter accepts the following options:
22784 1b Set 65Hz band gain.
22786 2b Set 92Hz band gain.
22788 3b Set 131Hz band gain.
22790 4b Set 185Hz band gain.
22792 5b Set 262Hz band gain.
22794 6b Set 370Hz band gain.
22796 7b Set 523Hz band gain.
22798 8b Set 740Hz band gain.
22800 9b Set 1047Hz band gain.
22802 10b Set 1480Hz band gain.
22804 11b Set 2093Hz band gain.
22806 12b Set 2960Hz band gain.
22808 13b Set 4186Hz band gain.
22810 14b Set 5920Hz band gain.
22812 15b Set 8372Hz band gain.
22814 16b Set 11840Hz band gain.
22816 17b Set 16744Hz band gain.
22818 18b Set 20000Hz band gain.
22820 surround
22822 Apply audio surround upmix filter.
22823 This filter allows to produce multichannel output from audio stream.
22825 The filter accepts the following options:
22827 chl_out
22829 Set output channel layout. By default, this is 5.1.
22830 See the Channel Layout section in the ffmpeg-utils(1) manual for
22832 the required syntax.
22833 chl_in
22835 Set input channel layout. By default, this is stereo.
22836 See the Channel Layout section in the ffmpeg-utils(1) manual for
22838 the required syntax.
22839 level_in
22841 Set input volume level. By default, this is 1.
22842 level_out
22844 Set output volume level. By default, this is 1.
22845 lfe Enable LFE channel output if output channel layout has it. By
22847 default, this is enabled.
22848 lfe_low
22850 Set LFE low cut off frequency. By default, this is 128 Hz.
22851 lfe_high
22853 Set LFE high cut off frequency. By default, this is 256 Hz.
22854 lfe_mode
22856 Set LFE mode, can be add or sub. Default is add. In add mode, LFE
22857 channel is created from input audio and added to output. In sub
22858 mode, LFE channel is created from input audio and added to output
22859 but also all non-LFE output channels are subtracted with output LFE
22860 channel.
22861 angle
22863 Set angle of stereo surround transform, Allowed range is from 0 to
22864 360. Default is 90.
22865 fc_in
22867 Set front center input volume. By default, this is 1.
22868 fc_out
22870 Set front center output volume. By default, this is 1.
22871 fl_in
22873 Set front left input volume. By default, this is 1.
22874 fl_out
22876 Set front left output volume. By default, this is 1.
22877 fr_in
22879 Set front right input volume. By default, this is 1.
22880 fr_out
22882 Set front right output volume. By default, this is 1.
22883 sl_in
22885 Set side left input volume. By default, this is 1.
22886 sl_out
22888 Set side left output volume. By default, this is 1.
22889 sr_in
22891 Set side right input volume. By default, this is 1.
22892 sr_out
22894 Set side right output volume. By default, this is 1.
22895 bl_in
22897 Set back left input volume. By default, this is 1.
22898 bl_out
22900 Set back left output volume. By default, this is 1.
22901 br_in
22903 Set back right input volume. By default, this is 1.
22904 br_out
22906 Set back right output volume. By default, this is 1.
22907 bc_in
22909 Set back center input volume. By default, this is 1.
22910 bc_out
22912 Set back center output volume. By default, this is 1.
22913 lfe_in
22915 Set LFE input volume. By default, this is 1.
22916 lfe_out
22918 Set LFE output volume. By default, this is 1.
22919 allx
22921 Set spread usage of stereo image across X axis for all channels.
22922 ally
22924 Set spread usage of stereo image across Y axis for all channels.
22925 fcx, flx, frx, blx, brx, slx, srx, bcx
22927 Set spread usage of stereo image across X axis for each channel.
22928 fcy, fly, fry, bly, bry, sly, sry, bcy
22930 Set spread usage of stereo image across Y axis for each channel.
22931 win_size
22933 Set window size. Allowed range is from 1024 to 65536. Default size
22934 is 4096.
22935 win_func
22937 Set window function.
22938 It accepts the following values:
22940 rect
22942 bartlett
22943 hann, hanning
22944 hamming
22945 blackman
22946 welch
22947 flattop
22948 bharris
22949 bnuttall
22950 bhann
22951 sine
22952 nuttall
22953 lanczos
22954 gauss
22955 tukey
22956 dolph
22957 cauchy
22958 parzen
22959 poisson
22960 bohman
22961 Default is "hann".
22963 overlap
22965 Set window overlap. If set to 1, the recommended overlap for
22966 selected window function will be picked. Default is 0.5.
22967 treble, highshelf
22969 Boost or cut treble (upper) frequencies of the audio using a two-pole
22970 shelving filter with a response similar to that of a standard hi-fi's
22971 tone-controls. This is also known as shelving equalisation (EQ).
22972 The filter accepts the following options:
22974 gain, g
22976 Give the gain at whichever is the lower of ~22 kHz and the Nyquist
22977 frequency. Its useful range is about -20 (for a large cut) to +20
22978 (for a large boost). Beware of clipping when using a positive gain.
22979 frequency, f
22981 Set the filter's central frequency and so can be used to extend or
22982 reduce the frequency range to be boosted or cut. The default value
22983 is 3000 Hz.
22984 width_type, t
22986 Set method to specify band-width of filter.
22987 h Hz
22989 q Q-Factor
22991 o octave
22993 s slope
22995 k kHz
22997 width, w
22999 Determine how steep is the filter's shelf transition.
23000 poles, p
23002 Set number of poles. Default is 2.
23003 mix, m
23005 How much to use filtered signal in output. Default is 1. Range is
23006 between 0 and 1.
23007 channels, c
23009 Specify which channels to filter, by default all available are
23010 filtered.
23011 normalize, n
23013 Normalize biquad coefficients, by default is disabled. Enabling it
23014 will normalize magnitude response at DC to 0dB.
23015 transform, a
23017 Set transform type of IIR filter.
23018 di
23020 dii
23021 tdii
23022 latt
23023 svf
23024 precision, r
23025 Set precison of filtering.
23026 auto
23028 Pick automatic sample format depending on surround filters.
23029 s16 Always use signed 16-bit.
23031 s32 Always use signed 32-bit.
23033 f32 Always use float 32-bit.
23035 f64 Always use float 64-bit.
23037 Commands
23039 This filter supports the following commands:
23041 frequency, f
23043 Change treble frequency. Syntax for the command is : "frequency"
23044 width_type, t
23046 Change treble width_type. Syntax for the command is : "width_type"
23047 width, w
23049 Change treble width. Syntax for the command is : "width"
23050 gain, g
23052 Change treble gain. Syntax for the command is : "gain"
23053 mix, m
23055 Change treble mix. Syntax for the command is : "mix"
23056 tremolo
23058 Sinusoidal amplitude modulation.
23059 The filter accepts the following options:
23061 f Modulation frequency in Hertz. Modulation frequencies in the
23063 subharmonic range (20 Hz or lower) will result in a tremolo effect.
23064 This filter may also be used as a ring modulator by specifying a
23065 modulation frequency higher than 20 Hz. Range is 0.1 - 20000.0.
23066 Default value is 5.0 Hz.
23067 d Depth of modulation as a percentage. Range is 0.0 - 1.0. Default
23069 value is 0.5.
23070 vibrato
23072 Sinusoidal phase modulation.
23073 The filter accepts the following options:
23075 f Modulation frequency in Hertz. Range is 0.1 - 20000.0. Default
23077 value is 5.0 Hz.
23078 d Depth of modulation as a percentage. Range is 0.0 - 1.0. Default
23080 value is 0.5.
23081 volume
23083 Adjust the input audio volume.
23084 It accepts the following parameters:
23086 volume
23088 Set audio volume expression.
23089 Output values are clipped to the maximum value.
23091 The output audio volume is given by the relation:
23093 <output_volume> = <volume> * <input_volume>
23095 The default value for volume is "1.0".
23097 precision
23099 This parameter represents the mathematical precision.
23100 It determines which input sample formats will be allowed, which
23102 affects the precision of the volume scaling.
23103 fixed
23105 8-bit fixed-point; this limits input sample format to U8, S16,
23106 and S32.
23107 float
23109 32-bit floating-point; this limits input sample format to FLT.
23110 (default)
23111 double
23113 64-bit floating-point; this limits input sample format to DBL.
23114 replaygain
23116 Choose the behaviour on encountering ReplayGain side data in input
23117 frames.
23118 drop
23120 Remove ReplayGain side data, ignoring its contents (the
23121 default).
23122 ignore
23124 Ignore ReplayGain side data, but leave it in the frame.
23125 track
23127 Prefer the track gain, if present.
23128 album
23130 Prefer the album gain, if present.
23131 replaygain_preamp
23133 Pre-amplification gain in dB to apply to the selected replaygain
23134 gain.
23135 Default value for replaygain_preamp is 0.0.
23137 replaygain_noclip
23139 Prevent clipping by limiting the gain applied.
23140 Default value for replaygain_noclip is 1.
23142 eval
23144 Set when the volume expression is evaluated.
23145 It accepts the following values:
23147 once
23149 only evaluate expression once during the filter initialization,
23150 or when the volume command is sent
23151 frame
23153 evaluate expression for each incoming frame
23154 Default value is once.
23156 The volume expression can contain the following parameters.
23158 n frame number (starting at zero)
23160 nb_channels
23162 number of channels
23163 nb_consumed_samples
23165 number of samples consumed by the filter
23166 nb_samples
23168 number of samples in the current frame
23169 pos original frame position in the file
23171 pts frame PTS
23173 sample_rate
23175 sample rate
23176 startpts
23178 PTS at start of stream
23179 startt
23181 time at start of stream
23182 t frame time
23184 tb timestamp timebase
23186 volume
23188 last set volume value
23189 Note that when eval is set to once only the sample_rate and tb
23191 variables are available, all other variables will evaluate to NAN.
23192 Commands
23194 This filter supports the following commands:
23196 volume
23198 Modify the volume expression. The command accepts the same syntax
23199 of the corresponding option.
23200 If the specified expression is not valid, it is kept at its current
23202 value.
23203 Examples
23205 • Halve the input audio volume:
23207 volume=volume=0.5
23209 volume=volume=1/2
23210 volume=volume=-6.0206dB
23211 In all the above example the named key for volume can be omitted,
23213 for example like in:
23214 volume=0.5
23216 • Increase input audio power by 6 decibels using fixed-point
23218 precision:
23219 volume=volume=6dB:precision=fixed
23221 • Fade volume after time 10 with an annihilation period of 5 seconds:
23223 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
23225 volumedetect
23227 Detect the volume of the input video.
23228 The filter has no parameters. The input is not modified. Statistics
23230 about the volume will be printed in the log when the input stream end
23231 is reached.
23232 In particular it will show the mean volume (root mean square), maximum
23234 volume (on a per-sample basis), and the beginning of a histogram of the
23235 registered volume values (from the maximum value to a cumulated 1/1000
23236 of the samples).
23237 All volumes are in decibels relative to the maximum PCM value.
23239 Examples
23241 Here is an excerpt of the output:
23243 [Parsed_volumedetect_0 0xa23120] mean_volume: -27 dB
23245 [Parsed_volumedetect_0 0xa23120] max_volume: -4 dB
23246 [Parsed_volumedetect_0 0xa23120] histogram_4db: 6
23247 [Parsed_volumedetect_0 0xa23120] histogram_5db: 62
23248 [Parsed_volumedetect_0 0xa23120] histogram_6db: 286
23249 [Parsed_volumedetect_0 0xa23120] histogram_7db: 1042
23250 [Parsed_volumedetect_0 0xa23120] histogram_8db: 2551
23251 [Parsed_volumedetect_0 0xa23120] histogram_9db: 4609
23252 [Parsed_volumedetect_0 0xa23120] histogram_10db: 8409
23253 It means that:
23255 • The mean square energy is approximately -27 dB, or 10^-2.7.
23257 • The largest sample is at -4 dB, or more precisely between -4 dB and
23259 -5 dB.
23260 • There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
23262 In other words, raising the volume by +4 dB does not cause any
23264 clipping, raising it by +5 dB causes clipping for 6 samples, etc.
23265
23267 Below is a description of the currently available audio sources.
23268 abuffer
23270 Buffer audio frames, and make them available to the filter chain.
23271 This source is mainly intended for a programmatic use, in particular
23273 through the interface defined in libavfilter/buffersrc.h.
23274 It accepts the following parameters:
23276 time_base
23278 The timebase which will be used for timestamps of submitted frames.
23279 It must be either a floating-point number or in
23280 numerator/denominator form.
23281 sample_rate
23283 The sample rate of the incoming audio buffers.
23284 sample_fmt
23286 The sample format of the incoming audio buffers. Either a sample
23287 format name or its corresponding integer representation from the
23288 enum AVSampleFormat in libavutil/samplefmt.h
23289 channel_layout
23291 The channel layout of the incoming audio buffers. Either a channel
23292 layout name from channel_layout_map in libavutil/channel_layout.c
23293 or its corresponding integer representation from the AV_CH_LAYOUT_*
23294 macros in libavutil/channel_layout.h
23295 channels
23297 The number of channels of the incoming audio buffers. If both
23298 channels and channel_layout are specified, then they must be
23299 consistent.
23300 Examples
23302 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
23304 will instruct the source to accept planar 16bit signed stereo at
23306 44100Hz. Since the sample format with name "s16p" corresponds to the
23307 number 6 and the "stereo" channel layout corresponds to the value 0x3,
23308 this is equivalent to:
23309 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
23311 aevalsrc
23313 Generate an audio signal specified by an expression.
23314 This source accepts in input one or more expressions (one for each
23316 channel), which are evaluated and used to generate a corresponding
23317 audio signal.
23318 This source accepts the following options:
23320 exprs
23322 Set the '|'-separated expressions list for each separate channel.
23323 In case the channel_layout option is not specified, the selected
23324 channel layout depends on the number of provided expressions.
23325 Otherwise the last specified expression is applied to the remaining
23326 output channels.
23327 channel_layout, c
23329 Set the channel layout. The number of channels in the specified
23330 layout must be equal to the number of specified expressions.
23331 duration, d
23333 Set the minimum duration of the sourced audio. See the Time
23334 duration section in the ffmpeg-utils(1) manual for the accepted
23335 syntax. Note that the resulting duration may be greater than the
23336 specified duration, as the generated audio is always cut at the end
23337 of a complete frame.
23338 If not specified, or the expressed duration is negative, the audio
23340 is supposed to be generated forever.
23341 nb_samples, n
23343 Set the number of samples per channel per each output frame,
23344 default to 1024.
23345 sample_rate, s
23347 Specify the sample rate, default to 44100.
23348 Each expression in exprs can contain the following constants:
23350 n number of the evaluated sample, starting from 0
23352 t time of the evaluated sample expressed in seconds, starting from 0
23354 s sample rate
23356 Examples
23358 • Generate silence:
23360 aevalsrc=0
23362 • Generate a sin signal with frequency of 440 Hz, set sample rate to
23364 8000 Hz:
23365 aevalsrc="sin(440*2*PI*t):s=8000"
23367 • Generate a two channels signal, specify the channel layout (Front
23369 Center + Back Center) explicitly:
23370 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
23372 • Generate white noise:
23374 aevalsrc="-2+random(0)"
23376 • Generate an amplitude modulated signal:
23378 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
23380 • Generate 2.5 Hz binaural beats on a 360 Hz carrier:
23382 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
23384 afirsrc
23386 Generate a FIR coefficients using frequency sampling method.
23387 The resulting stream can be used with afir filter for filtering the
23389 audio signal.
23390 The filter accepts the following options:
23392 taps, t
23394 Set number of filter coefficents in output audio stream. Default
23395 value is 1025.
23396 frequency, f
23398 Set frequency points from where magnitude and phase are set. This
23399 must be in non decreasing order, and first element must be 0, while
23400 last element must be 1. Elements are separated by white spaces.
23401 magnitude, m
23403 Set magnitude value for every frequency point set by frequency.
23404 Number of values must be same as number of frequency points.
23405 Values are separated by white spaces.
23406 phase, p
23408 Set phase value for every frequency point set by frequency. Number
23409 of values must be same as number of frequency points. Values are
23410 separated by white spaces.
23411 sample_rate, r
23413 Set sample rate, default is 44100.
23414 nb_samples, n
23416 Set number of samples per each frame. Default is 1024.
23417 win_func, w
23419 Set window function. Default is blackman.
23420 anullsrc
23422 The null audio source, return unprocessed audio frames. It is mainly
23423 useful as a template and to be employed in analysis / debugging tools,
23424 or as the source for filters which ignore the input data (for example
23425 the sox synth filter).
23426 This source accepts the following options:
23428 channel_layout, cl
23430 Specifies the channel layout, and can be either an integer or a
23431 string representing a channel layout. The default value of
23432 channel_layout is "stereo".
23433 Check the channel_layout_map definition in
23435 libavutil/channel_layout.c for the mapping between strings and
23436 channel layout values.
23437 sample_rate, r
23439 Specifies the sample rate, and defaults to 44100.
23440 nb_samples, n
23442 Set the number of samples per requested frames.
23443 duration, d
23445 Set the duration of the sourced audio. See the Time duration
23446 section in the ffmpeg-utils(1) manual for the accepted syntax.
23447 If not specified, or the expressed duration is negative, the audio
23449 is supposed to be generated forever.
23450 Examples
23452 • Set the sample rate to 48000 Hz and the channel layout to
23454 AV_CH_LAYOUT_MONO.
23455 anullsrc=r=48000:cl=4
23457 • Do the same operation with a more obvious syntax:
23459 anullsrc=r=48000:cl=mono
23461 All the parameters need to be explicitly defined.
23463 flite
23465 Synthesize a voice utterance using the libflite library.
23466 To enable compilation of this filter you need to configure FFmpeg with
23468 "--enable-libflite".
23469 Note that versions of the flite library prior to 2.0 are not thread-
23471 safe.
23472 The filter accepts the following options:
23474 list_voices
23476 If set to 1, list the names of the available voices and exit
23477 immediately. Default value is 0.
23478 nb_samples, n
23480 Set the maximum number of samples per frame. Default value is 512.
23481 textfile
23483 Set the filename containing the text to speak.
23484 text
23486 Set the text to speak.
23487 voice, v
23489 Set the voice to use for the speech synthesis. Default value is
23490 "kal". See also the list_voices option.
23491 Examples
23493 • Read from file speech.txt, and synthesize the text using the
23495 standard flite voice:
23496 flite=textfile=speech.txt
23498 • Read the specified text selecting the "slt" voice:
23500 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
23502 • Input text to ffmpeg:
23504 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
23506 • Make ffplay speak the specified text, using "flite" and the "lavfi"
23508 device:
23509 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
23511 For more information about libflite, check:
23513 <http://www.festvox.org/flite/>
23514 anoisesrc
23516 Generate a noise audio signal.
23517 The filter accepts the following options:
23519 sample_rate, r
23521 Specify the sample rate. Default value is 48000 Hz.
23522 amplitude, a
23524 Specify the amplitude (0.0 - 1.0) of the generated audio stream.
23525 Default value is 1.0.
23526 duration, d
23528 Specify the duration of the generated audio stream. Not specifying
23529 this option results in noise with an infinite length.
23530 color, colour, c
23532 Specify the color of noise. Available noise colors are white, pink,
23533 brown, blue, violet and velvet. Default color is white.
23534 seed, s
23536 Specify a value used to seed the PRNG.
23537 nb_samples, n
23539 Set the number of samples per each output frame, default is 1024.
23540 Examples
23542 • Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate
23544 and an amplitude of 0.5:
23545 anoisesrc=d=60:c=pink:r=44100:a=0.5
23547 hilbert
23549 Generate odd-tap Hilbert transform FIR coefficients.
23550 The resulting stream can be used with afir filter for phase-shifting
23552 the signal by 90 degrees.
23553 This is used in many matrix coding schemes and for analytic signal
23555 generation. The process is often written as a multiplication by i (or
23556 j), the imaginary unit.
23557 The filter accepts the following options:
23559 sample_rate, s
23561 Set sample rate, default is 44100.
23562 taps, t
23564 Set length of FIR filter, default is 22051.
23565 nb_samples, n
23567 Set number of samples per each frame.
23568 win_func, w
23570 Set window function to be used when generating FIR coefficients.
23571 sinc
23573 Generate a sinc kaiser-windowed low-pass, high-pass, band-pass, or
23574 band-reject FIR coefficients.
23575 The resulting stream can be used with afir filter for filtering the
23577 audio signal.
23578 The filter accepts the following options:
23580 sample_rate, r
23582 Set sample rate, default is 44100.
23583 nb_samples, n
23585 Set number of samples per each frame. Default is 1024.
23586 hp Set high-pass frequency. Default is 0.
23588 lp Set low-pass frequency. Default is 0. If high-pass frequency is
23590 lower than low-pass frequency and low-pass frequency is higher than
23591 0 then filter will create band-pass filter coefficients, otherwise
23592 band-reject filter coefficients.
23593 phase
23595 Set filter phase response. Default is 50. Allowed range is from 0
23596 to 100.
23597 beta
23599 Set Kaiser window beta.
23600 att Set stop-band attenuation. Default is 120dB, allowed range is from
23602 40 to 180 dB.
23603 round
23605 Enable rounding, by default is disabled.
23606 hptaps
23608 Set number of taps for high-pass filter.
23609 lptaps
23611 Set number of taps for low-pass filter.
23612 sine
23614 Generate an audio signal made of a sine wave with amplitude 1/8.
23615 The audio signal is bit-exact.
23617 The filter accepts the following options:
23619 frequency, f
23621 Set the carrier frequency. Default is 440 Hz.
23622 beep_factor, b
23624 Enable a periodic beep every second with frequency beep_factor
23625 times the carrier frequency. Default is 0, meaning the beep is
23626 disabled.
23627 sample_rate, r
23629 Specify the sample rate, default is 44100.
23630 duration, d
23632 Specify the duration of the generated audio stream.
23633 samples_per_frame
23635 Set the number of samples per output frame.
23636 The expression can contain the following constants:
23638 n The (sequential) number of the output audio frame, starting
23640 from 0.
23641 pts The PTS (Presentation TimeStamp) of the output audio frame,
23643 expressed in TB units.
23644 t The PTS of the output audio frame, expressed in seconds.
23646 TB The timebase of the output audio frames.
23648 Default is 1024.
23650 Examples
23652 • Generate a simple 440 Hz sine wave:
23654 sine
23656 • Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5
23658 seconds:
23659 sine=220:4:d=5
23661 sine=f=220:b=4:d=5
23662 sine=frequency=220:beep_factor=4:duration=5
23663 • Generate a 1 kHz sine wave following "1602,1601,1602,1601,1602"
23665 NTSC pattern:
23666 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
23668
23670 Below is a description of the currently available audio sinks.
23671 abuffersink
23673 Buffer audio frames, and make them available to the end of filter
23674 chain.
23675 This sink is mainly intended for programmatic use, in particular
23677 through the interface defined in libavfilter/buffersink.h or the
23678 options system.
23679 It accepts a pointer to an AVABufferSinkContext structure, which
23681 defines the incoming buffers' formats, to be passed as the opaque
23682 parameter to "avfilter_init_filter" for initialization.
23683 anullsink
23685 Null audio sink; do absolutely nothing with the input audio. It is
23686 mainly useful as a template and for use in analysis / debugging tools.
23687
23689 When you configure your FFmpeg build, you can disable any of the
23690 existing filters using "--disable-filters". The configure output will
23691 show the video filters included in your build.
23692 Below is a description of the currently available video filters.
23694 addroi
23696 Mark a region of interest in a video frame.
23697 The frame data is passed through unchanged, but metadata is attached to
23699 the frame indicating regions of interest which can affect the behaviour
23700 of later encoding. Multiple regions can be marked by applying the
23701 filter multiple times.
23702 x Region distance in pixels from the left edge of the frame.
23704 y Region distance in pixels from the top edge of the frame.
23706 w Region width in pixels.
23708 h Region height in pixels.
23710 The parameters x, y, w and h are expressions, and may contain the
23712 following variables:
23713 iw Width of the input frame.
23715 ih Height of the input frame.
23717 qoffset
23719 Quantisation offset to apply within the region.
23720 This must be a real value in the range -1 to +1. A value of zero
23722 indicates no quality change. A negative value asks for better
23723 quality (less quantisation), while a positive value asks for worse
23724 quality (greater quantisation).
23725 The range is calibrated so that the extreme values indicate the
23727 largest possible offset - if the rest of the frame is encoded with
23728 the worst possible quality, an offset of -1 indicates that this
23729 region should be encoded with the best possible quality anyway.
23730 Intermediate values are then interpolated in some codec-dependent
23731 way.
23732 For example, in 10-bit H.264 the quantisation parameter varies
23734 between -12 and 51. A typical qoffset value of -1/10 therefore
23735 indicates that this region should be encoded with a QP around one-
23736 tenth of the full range better than the rest of the frame. So, if
23737 most of the frame were to be encoded with a QP of around 30, this
23738 region would get a QP of around 24 (an offset of approximately
23739 -1/10 * (51 - -12) = -6.3). An extreme value of -1 would indicate
23740 that this region should be encoded with the best possible quality
23741 regardless of the treatment of the rest of the frame - that is,
23742 should be encoded at a QP of -12.
23743 clear
23745 If set to true, remove any existing regions of interest marked on
23746 the frame before adding the new one.
23747 Examples
23749 • Mark the centre quarter of the frame as interesting.
23751 addroi=iw/4:ih/4:iw/2:ih/2:-1/10
23753 • Mark the 100-pixel-wide region on the left edge of the frame as
23755 very uninteresting (to be encoded at much lower quality than the
23756 rest of the frame).
23757 addroi=0:0:100:ih:+1/5
23759 alphaextract
23761 Extract the alpha component from the input as a grayscale video. This
23762 is especially useful with the alphamerge filter.
23763 alphamerge
23765 Add or replace the alpha component of the primary input with the
23766 grayscale value of a second input. This is intended for use with
23767 alphaextract to allow the transmission or storage of frame sequences
23768 that have alpha in a format that doesn't support an alpha channel.
23769 For example, to reconstruct full frames from a normal YUV-encoded video
23771 and a separate video created with alphaextract, you might use:
23772 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
23774 amplify
23776 Amplify differences between current pixel and pixels of adjacent frames
23777 in same pixel location.
23778 This filter accepts the following options:
23780 radius
23782 Set frame radius. Default is 2. Allowed range is from 1 to 63. For
23783 example radius of 3 will instruct filter to calculate average of 7
23784 frames.
23785 factor
23787 Set factor to amplify difference. Default is 2. Allowed range is
23788 from 0 to 65535.
23789 threshold
23791 Set threshold for difference amplification. Any difference greater
23792 or equal to this value will not alter source pixel. Default is 10.
23793 Allowed range is from 0 to 65535.
23794 tolerance
23796 Set tolerance for difference amplification. Any difference lower to
23797 this value will not alter source pixel. Default is 0. Allowed
23798 range is from 0 to 65535.
23799 low Set lower limit for changing source pixel. Default is 65535.
23801 Allowed range is from 0 to 65535. This option controls maximum
23802 possible value that will decrease source pixel value.
23803 high
23805 Set high limit for changing source pixel. Default is 65535. Allowed
23806 range is from 0 to 65535. This option controls maximum possible
23807 value that will increase source pixel value.
23808 planes
23810 Set which planes to filter. Default is all. Allowed range is from 0
23811 to 15.
23812 Commands
23814 This filter supports the following commands that corresponds to option
23816 of same name:
23817 factor
23819 threshold
23820 tolerance
23821 low
23822 high
23823 planes
23824 ass
23826 Same as the subtitles filter, except that it doesn't require libavcodec
23827 and libavformat to work. On the other hand, it is limited to ASS
23828 (Advanced Substation Alpha) subtitles files.
23829 This filter accepts the following option in addition to the common
23831 options from the subtitles filter:
23832 shaping
23834 Set the shaping engine
23835 Available values are:
23837 auto
23839 The default libass shaping engine, which is the best available.
23840 simple
23842 Fast, font-agnostic shaper that can do only substitutions
23843 complex
23845 Slower shaper using OpenType for substitutions and positioning
23846 The default is "auto".
23848 atadenoise
23850 Apply an Adaptive Temporal Averaging Denoiser to the video input.
23851 The filter accepts the following options:
23853 0a Set threshold A for 1st plane. Default is 0.02. Valid range is 0
23855 to 0.3.
23856 0b Set threshold B for 1st plane. Default is 0.04. Valid range is 0
23858 to 5.
23859 1a Set threshold A for 2nd plane. Default is 0.02. Valid range is 0
23861 to 0.3.
23862 1b Set threshold B for 2nd plane. Default is 0.04. Valid range is 0
23864 to 5.
23865 2a Set threshold A for 3rd plane. Default is 0.02. Valid range is 0
23867 to 0.3.
23868 2b Set threshold B for 3rd plane. Default is 0.04. Valid range is 0
23870 to 5.
23871 Threshold A is designed to react on abrupt changes in the input
23873 signal and threshold B is designed to react on continuous changes
23874 in the input signal.
23875 s Set number of frames filter will use for averaging. Default is 9.
23877 Must be odd number in range [5, 129].
23878 p Set what planes of frame filter will use for averaging. Default is
23880 all.
23881 a Set what variant of algorithm filter will use for averaging.
23883 Default is "p" parallel. Alternatively can be set to "s" serial.
23884 Parallel can be faster then serial, while other way around is never
23886 true. Parallel will abort early on first change being greater then
23887 thresholds, while serial will continue processing other side of
23888 frames if they are equal or below thresholds.
23889 0s
23891 1s
23892 2s Set sigma for 1st plane, 2nd plane or 3rd plane. Default is 32767.
23893 Valid range is from 0 to 32767. This options controls weight for
23894 each pixel in radius defined by size. Default value means every
23895 pixel have same weight. Setting this option to 0 effectively
23896 disables filtering.
23897 Commands
23899 This filter supports same commands as options except option "s". The
23901 command accepts the same syntax of the corresponding option.
23902 avgblur
23904 Apply average blur filter.
23905 The filter accepts the following options:
23907 sizeX
23909 Set horizontal radius size.
23910 planes
23912 Set which planes to filter. By default all planes are filtered.
23913 sizeY
23915 Set vertical radius size, if zero it will be same as "sizeX".
23916 Default is 0.
23917 Commands
23919 This filter supports same commands as options. The command accepts the
23921 same syntax of the corresponding option.
23922 If the specified expression is not valid, it is kept at its current
23924 value.
23925 bbox
23927 Compute the bounding box for the non-black pixels in the input frame
23928 luminance plane.
23929 This filter computes the bounding box containing all the pixels with a
23931 luminance value greater than the minimum allowed value. The parameters
23932 describing the bounding box are printed on the filter log.
23933 The filter accepts the following option:
23935 min_val
23937 Set the minimal luminance value. Default is 16.
23938 Commands
23940 This filter supports the all above options as commands.
23942 bilateral
23944 Apply bilateral filter, spatial smoothing while preserving edges.
23945 The filter accepts the following options:
23947 sigmaS
23949 Set sigma of gaussian function to calculate spatial weight.
23950 Allowed range is 0 to 512. Default is 0.1.
23951 sigmaR
23953 Set sigma of gaussian function to calculate range weight. Allowed
23954 range is 0 to 1. Default is 0.1.
23955 planes
23957 Set planes to filter. Default is first only.
23958 Commands
23960 This filter supports the all above options as commands.
23962 bitplanenoise
23964 Show and measure bit plane noise.
23965 The filter accepts the following options:
23967 bitplane
23969 Set which plane to analyze. Default is 1.
23970 filter
23972 Filter out noisy pixels from "bitplane" set above. Default is
23973 disabled.
23974 blackdetect
23976 Detect video intervals that are (almost) completely black. Can be
23977 useful to detect chapter transitions, commercials, or invalid
23978 recordings.
23979 The filter outputs its detection analysis to both the log as well as
23981 frame metadata. If a black segment of at least the specified minimum
23982 duration is found, a line with the start and end timestamps as well as
23983 duration is printed to the log with level "info". In addition, a log
23984 line with level "debug" is printed per frame showing the black amount
23985 detected for that frame.
23986 The filter also attaches metadata to the first frame of a black segment
23988 with key "lavfi.black_start" and to the first frame after the black
23989 segment ends with key "lavfi.black_end". The value is the frame's
23990 timestamp. This metadata is added regardless of the minimum duration
23991 specified.
23992 The filter accepts the following options:
23994 black_min_duration, d
23996 Set the minimum detected black duration expressed in seconds. It
23997 must be a non-negative floating point number.
23998 Default value is 2.0.
24000 picture_black_ratio_th, pic_th
24002 Set the threshold for considering a picture "black". Express the
24003 minimum value for the ratio:
24004 <nb_black_pixels> / <nb_pixels>
24006 for which a picture is considered black. Default value is 0.98.
24008 pixel_black_th, pix_th
24010 Set the threshold for considering a pixel "black".
24011 The threshold expresses the maximum pixel luminance value for which
24013 a pixel is considered "black". The provided value is scaled
24014 according to the following equation:
24015 <absolute_threshold> = <luminance_minimum_value> + <pixel_black_th> * <luminance_range_size>
24017 luminance_range_size and luminance_minimum_value depend on the
24019 input video format, the range is [0-255] for YUV full-range formats
24020 and [16-235] for YUV non full-range formats.
24021 Default value is 0.10.
24023 The following example sets the maximum pixel threshold to the minimum
24025 value, and detects only black intervals of 2 or more seconds:
24026 blackdetect=d=2:pix_th=0.00
24028 blackframe
24030 Detect frames that are (almost) completely black. Can be useful to
24031 detect chapter transitions or commercials. Output lines consist of the
24032 frame number of the detected frame, the percentage of blackness, the
24033 position in the file if known or -1 and the timestamp in seconds.
24034 In order to display the output lines, you need to set the loglevel at
24036 least to the AV_LOG_INFO value.
24037 This filter exports frame metadata "lavfi.blackframe.pblack". The
24039 value represents the percentage of pixels in the picture that are below
24040 the threshold value.
24041 It accepts the following parameters:
24043 amount
24045 The percentage of the pixels that have to be below the threshold;
24046 it defaults to 98.
24047 threshold, thresh
24049 The threshold below which a pixel value is considered black; it
24050 defaults to 32.
24051 blend
24053 Blend two video frames into each other.
24054 The "blend" filter takes two input streams and outputs one stream, the
24056 first input is the "top" layer and second input is "bottom" layer. By
24057 default, the output terminates when the longest input terminates.
24058 The "tblend" (time blend) filter takes two consecutive frames from one
24060 single stream, and outputs the result obtained by blending the new
24061 frame on top of the old frame.
24062 A description of the accepted options follows.
24064 c0_mode
24066 c1_mode
24067 c2_mode
24068 c3_mode
24069 all_mode
24070 Set blend mode for specific pixel component or all pixel components
24071 in case of all_mode. Default value is "normal".
24072 Available values for component modes are:
24074 addition
24076 and
24077 average
24078 bleach
24079 burn
24080 darken
24081 difference
24082 divide
24083 dodge
24084 exclusion
24085 extremity
24086 freeze
24087 geometric
24088 glow
24089 grainextract
24090 grainmerge
24091 hardlight
24092 hardmix
24093 hardoverlay
24094 harmonic
24095 heat
24096 interpolate
24097 lighten
24098 linearlight
24099 multiply
24100 multiply128
24101 negation
24102 normal
24103 or
24104 overlay
24105 phoenix
24106 pinlight
24107 reflect
24108 screen
24109 softdifference
24110 softlight
24111 stain
24112 subtract
24113 vividlight
24114 xor
24115 c0_opacity
24116 c1_opacity
24117 c2_opacity
24118 c3_opacity
24119 all_opacity
24120 Set blend opacity for specific pixel component or all pixel
24121 components in case of all_opacity. Only used in combination with
24122 pixel component blend modes.
24123 c0_expr
24125 c1_expr
24126 c2_expr
24127 c3_expr
24128 all_expr
24129 Set blend expression for specific pixel component or all pixel
24130 components in case of all_expr. Note that related mode options will
24131 be ignored if those are set.
24132 The expressions can use the following variables:
24134 N The sequential number of the filtered frame, starting from 0.
24136 X
24138 Y the coordinates of the current sample
24139 W
24141 H the width and height of currently filtered plane
24142 SW
24144 SH Width and height scale for the plane being filtered. It is the
24145 ratio between the dimensions of the current plane to the luma
24146 plane, e.g. for a "yuv420p" frame, the values are "1,1" for the
24147 luma plane and "0.5,0.5" for the chroma planes.
24148 T Time of the current frame, expressed in seconds.
24150 TOP, A
24152 Value of pixel component at current location for first video
24153 frame (top layer).
24154 BOTTOM, B
24156 Value of pixel component at current location for second video
24157 frame (bottom layer).
24158 The "blend" filter also supports the framesync options.
24160 Examples
24162 • Apply transition from bottom layer to top layer in first 10
24164 seconds:
24165 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
24167 • Apply linear horizontal transition from top layer to bottom layer:
24169 blend=all_expr='A*(X/W)+B*(1-X/W)'
24171 • Apply 1x1 checkerboard effect:
24173 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
24175 • Apply uncover left effect:
24177 blend=all_expr='if(gte(N*SW+X,W),A,B)'
24179 • Apply uncover down effect:
24181 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
24183 • Apply uncover up-left effect:
24185 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
24187 • Split diagonally video and shows top and bottom layer on each side:
24189 blend=all_expr='if(gt(X,Y*(W/H)),A,B)'
24191 • Display differences between the current and the previous frame:
24193 tblend=all_mode=grainextract
24195 Commands
24197 This filter supports same commands as options.
24199 bm3d
24201 Denoise frames using Block-Matching 3D algorithm.
24202 The filter accepts the following options.
24204 sigma
24206 Set denoising strength. Default value is 1. Allowed range is from
24207 0 to 999.9. The denoising algorithm is very sensitive to sigma, so
24208 adjust it according to the source.
24209 block
24211 Set local patch size. This sets dimensions in 2D.
24212 bstep
24214 Set sliding step for processing blocks. Default value is 4.
24215 Allowed range is from 1 to 64. Smaller values allows processing
24216 more reference blocks and is slower.
24217 group
24219 Set maximal number of similar blocks for 3rd dimension. Default
24220 value is 1. When set to 1, no block matching is done. Larger
24221 values allows more blocks in single group. Allowed range is from 1
24222 to 256.
24223 range
24225 Set radius for search block matching. Default is 9. Allowed range
24226 is from 1 to INT32_MAX.
24227 mstep
24229 Set step between two search locations for block matching. Default
24230 is 1. Allowed range is from 1 to 64. Smaller is slower.
24231 thmse
24233 Set threshold of mean square error for block matching. Valid range
24234 is 0 to INT32_MAX.
24235 hdthr
24237 Set thresholding parameter for hard thresholding in 3D transformed
24238 domain. Larger values results in stronger hard-thresholding
24239 filtering in frequency domain.
24240 estim
24242 Set filtering estimation mode. Can be "basic" or "final". Default
24243 is "basic".
24244 ref If enabled, filter will use 2nd stream for block matching. Default
24246 is disabled for "basic" value of estim option, and always enabled
24247 if value of estim is "final".
24248 planes
24250 Set planes to filter. Default is all available except alpha.
24251 Examples
24253 • Basic filtering with bm3d:
24255 bm3d=sigma=3:block=4:bstep=2:group=1:estim=basic
24257 • Same as above, but filtering only luma:
24259 bm3d=sigma=3:block=4:bstep=2:group=1:estim=basic:planes=1
24261 • Same as above, but with both estimation modes:
24263 split[a][b],[a]bm3d=sigma=3:block=4:bstep=2:group=1:estim=basic[a],[b][a]bm3d=sigma=3:block=4:bstep=2:group=16:estim=final:ref=1
24265 • Same as above, but prefilter with nlmeans filter instead:
24267 split[a][b],[a]nlmeans=s=3:r=7:p=3[a],[b][a]bm3d=sigma=3:block=4:bstep=2:group=16:estim=final:ref=1
24269 boxblur
24271 Apply a boxblur algorithm to the input video.
24272 It accepts the following parameters:
24274 luma_radius, lr
24276 luma_power, lp
24277 chroma_radius, cr
24278 chroma_power, cp
24279 alpha_radius, ar
24280 alpha_power, ap
24281 A description of the accepted options follows.
24283 luma_radius, lr
24285 chroma_radius, cr
24286 alpha_radius, ar
24287 Set an expression for the box radius in pixels used for blurring
24288 the corresponding input plane.
24289 The radius value must be a non-negative number, and must not be
24291 greater than the value of the expression "min(w,h)/2" for the luma
24292 and alpha planes, and of "min(cw,ch)/2" for the chroma planes.
24293 Default value for luma_radius is "2". If not specified,
24295 chroma_radius and alpha_radius default to the corresponding value
24296 set for luma_radius.
24297 The expressions can contain the following constants:
24299 w
24301 h The input width and height in pixels.
24302 cw
24304 ch The input chroma image width and height in pixels.
24305 hsub
24307 vsub
24308 The horizontal and vertical chroma subsample values. For
24309 example, for the pixel format "yuv422p", hsub is 2 and vsub is
24310 1.
24311 luma_power, lp
24313 chroma_power, cp
24314 alpha_power, ap
24315 Specify how many times the boxblur filter is applied to the
24316 corresponding plane.
24317 Default value for luma_power is 2. If not specified, chroma_power
24319 and alpha_power default to the corresponding value set for
24320 luma_power.
24321 A value of 0 will disable the effect.
24323 Examples
24325 • Apply a boxblur filter with the luma, chroma, and alpha radii set
24327 to 2:
24328 boxblur=luma_radius=2:luma_power=1
24330 boxblur=2:1
24331 • Set the luma radius to 2, and alpha and chroma radius to 0:
24333 boxblur=2:1:cr=0:ar=0
24335 • Set the luma and chroma radii to a fraction of the video dimension:
24337 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
24339 bwdif
24341 Deinterlace the input video ("bwdif" stands for "Bob Weaver
24342 Deinterlacing Filter").
24343 Motion adaptive deinterlacing based on yadif with the use of w3fdif and
24345 cubic interpolation algorithms. It accepts the following parameters:
24346 mode
24348 The interlacing mode to adopt. It accepts one of the following
24349 values:
24350 0, send_frame
24352 Output one frame for each frame.
24353 1, send_field
24355 Output one frame for each field.
24356 The default value is "send_field".
24358 parity
24360 The picture field parity assumed for the input interlaced video. It
24361 accepts one of the following values:
24362 0, tff
24364 Assume the top field is first.
24365 1, bff
24367 Assume the bottom field is first.
24368 -1, auto
24370 Enable automatic detection of field parity.
24371 The default value is "auto". If the interlacing is unknown or the
24373 decoder does not export this information, top field first will be
24374 assumed.
24375 deint
24377 Specify which frames to deinterlace. Accepts one of the following
24378 values:
24379 0, all
24381 Deinterlace all frames.
24382 1, interlaced
24384 Only deinterlace frames marked as interlaced.
24385 The default value is "all".
24387 cas
24389 Apply Contrast Adaptive Sharpen filter to video stream.
24390 The filter accepts the following options:
24392 strength
24394 Set the sharpening strength. Default value is 0.
24395 planes
24397 Set planes to filter. Default value is to filter all planes except
24398 alpha plane.
24399 Commands
24401 This filter supports same commands as options.
24403 chromahold
24405 Remove all color information for all colors except for certain one.
24406 The filter accepts the following options:
24408 color
24410 The color which will not be replaced with neutral chroma.
24411 similarity
24413 Similarity percentage with the above color. 0.01 matches only the
24414 exact key color, while 1.0 matches everything.
24415 blend
24417 Blend percentage. 0.0 makes pixels either fully gray, or not gray
24418 at all. Higher values result in more preserved color.
24419 yuv Signals that the color passed is already in YUV instead of RGB.
24421 Literal colors like "green" or "red" don't make sense with this
24423 enabled anymore. This can be used to pass exact YUV values as
24424 hexadecimal numbers.
24425 Commands
24427 This filter supports same commands as options. The command accepts the
24429 same syntax of the corresponding option.
24430 If the specified expression is not valid, it is kept at its current
24432 value.
24433 chromakey
24435 YUV colorspace color/chroma keying.
24436 The filter accepts the following options:
24438 color
24440 The color which will be replaced with transparency.
24441 similarity
24443 Similarity percentage with the key color.
24444 0.01 matches only the exact key color, while 1.0 matches
24446 everything.
24447 blend
24449 Blend percentage.
24450 0.0 makes pixels either fully transparent, or not transparent at
24452 all.
24453 Higher values result in semi-transparent pixels, with a higher
24455 transparency the more similar the pixels color is to the key color.
24456 yuv Signals that the color passed is already in YUV instead of RGB.
24458 Literal colors like "green" or "red" don't make sense with this
24460 enabled anymore. This can be used to pass exact YUV values as
24461 hexadecimal numbers.
24462 Commands
24464 This filter supports same commands as options. The command accepts the
24466 same syntax of the corresponding option.
24467 If the specified expression is not valid, it is kept at its current
24469 value.
24470 Examples
24472 • Make every green pixel in the input image transparent:
24474 ffmpeg -i input.png -vf chromakey=green out.png
24476 • Overlay a greenscreen-video on top of a static black background.
24478 ffmpeg -f lavfi -i color=c=black:s=1280x720 -i video.mp4 -shortest -filter_complex "[1:v]chromakey=0x70de77:0.1:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.mkv
24480 chromanr
24482 Reduce chrominance noise.
24483 The filter accepts the following options:
24485 thres
24487 Set threshold for averaging chrominance values. Sum of absolute
24488 difference of Y, U and V pixel components of current pixel and
24489 neighbour pixels lower than this threshold will be used in
24490 averaging. Luma component is left unchanged and is copied to
24491 output. Default value is 30. Allowed range is from 1 to 200.
24492 sizew
24494 Set horizontal radius of rectangle used for averaging. Allowed
24495 range is from 1 to 100. Default value is 5.
24496 sizeh
24498 Set vertical radius of rectangle used for averaging. Allowed range
24499 is from 1 to 100. Default value is 5.
24500 stepw
24502 Set horizontal step when averaging. Default value is 1. Allowed
24503 range is from 1 to 50. Mostly useful to speed-up filtering.
24504 steph
24506 Set vertical step when averaging. Default value is 1. Allowed
24507 range is from 1 to 50. Mostly useful to speed-up filtering.
24508 threy
24510 Set Y threshold for averaging chrominance values. Set finer
24511 control for max allowed difference between Y components of current
24512 pixel and neigbour pixels. Default value is 200. Allowed range is
24513 from 1 to 200.
24514 threu
24516 Set U threshold for averaging chrominance values. Set finer
24517 control for max allowed difference between U components of current
24518 pixel and neigbour pixels. Default value is 200. Allowed range is
24519 from 1 to 200.
24520 threv
24522 Set V threshold for averaging chrominance values. Set finer
24523 control for max allowed difference between V components of current
24524 pixel and neigbour pixels. Default value is 200. Allowed range is
24525 from 1 to 200.
24526 distance
24528 Set distance type used in calculations.
24529 manhattan
24531 Absolute difference.
24532 euclidean
24534 Difference squared.
24535 Default distance type is manhattan.
24537 Commands
24539 This filter supports same commands as options. The command accepts the
24541 same syntax of the corresponding option.
24542 chromashift
24544 Shift chroma pixels horizontally and/or vertically.
24545 The filter accepts the following options:
24547 cbh Set amount to shift chroma-blue horizontally.
24549 cbv Set amount to shift chroma-blue vertically.
24551 crh Set amount to shift chroma-red horizontally.
24553 crv Set amount to shift chroma-red vertically.
24555 edge
24557 Set edge mode, can be smear, default, or warp.
24558 Commands
24560 This filter supports the all above options as commands.
24562 ciescope
24564 Display CIE color diagram with pixels overlaid onto it.
24565 The filter accepts the following options:
24567 system
24569 Set color system.
24570 ntsc, 470m
24572 ebu, 470bg
24573 smpte
24574 240m
24575 apple
24576 widergb
24577 cie1931
24578 rec709, hdtv
24579 uhdtv, rec2020
24580 dcip3
24581 cie Set CIE system.
24582 xyy
24584 ucs
24585 luv
24586 gamuts
24587 Set what gamuts to draw.
24588 See "system" option for available values.
24590 size, s
24592 Set ciescope size, by default set to 512.
24593 intensity, i
24595 Set intensity used to map input pixel values to CIE diagram.
24596 contrast
24598 Set contrast used to draw tongue colors that are out of active
24599 color system gamut.
24600 corrgamma
24602 Correct gamma displayed on scope, by default enabled.
24603 showwhite
24605 Show white point on CIE diagram, by default disabled.
24606 gamma
24608 Set input gamma. Used only with XYZ input color space.
24609 codecview
24611 Visualize information exported by some codecs.
24612 Some codecs can export information through frames using side-data or
24614 other means. For example, some MPEG based codecs export motion vectors
24615 through the export_mvs flag in the codec flags2 option.
24616 The filter accepts the following option:
24618 block
24620 Display block partition structure using the luma plane.
24621 mv Set motion vectors to visualize.
24623 Available flags for mv are:
24625 pf forward predicted MVs of P-frames
24627 bf forward predicted MVs of B-frames
24629 bb backward predicted MVs of B-frames
24631 qp Display quantization parameters using the chroma planes.
24633 mv_type, mvt
24635 Set motion vectors type to visualize. Includes MVs from all frames
24636 unless specified by frame_type option.
24637 Available flags for mv_type are:
24639 fp forward predicted MVs
24641 bp backward predicted MVs
24643 frame_type, ft
24645 Set frame type to visualize motion vectors of.
24646 Available flags for frame_type are:
24648 if intra-coded frames (I-frames)
24650 pf predicted frames (P-frames)
24652 bf bi-directionally predicted frames (B-frames)
24654 Examples
24656 • Visualize forward predicted MVs of all frames using ffplay:
24658 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
24660 • Visualize multi-directionals MVs of P and B-Frames using ffplay:
24662 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
24664 colorbalance
24666 Modify intensity of primary colors (red, green and blue) of input
24667 frames.
24668 The filter allows an input frame to be adjusted in the shadows,
24670 midtones or highlights regions for the red-cyan, green-magenta or blue-
24671 yellow balance.
24672 A positive adjustment value shifts the balance towards the primary
24674 color, a negative value towards the complementary color.
24675 The filter accepts the following options:
24677 rs
24679 gs
24680 bs Adjust red, green and blue shadows (darkest pixels).
24681 rm
24683 gm
24684 bm Adjust red, green and blue midtones (medium pixels).
24685 rh
24687 gh
24688 bh Adjust red, green and blue highlights (brightest pixels).
24689 Allowed ranges for options are "[-1.0, 1.0]". Defaults are 0.
24691 pl Preserve lightness when changing color balance. Default is
24693 disabled.
24694 Examples
24696 • Add red color cast to shadows:
24698 colorbalance=rs=.3
24700 Commands
24702 This filter supports the all above options as commands.
24704 colorcontrast
24706 Adjust color contrast between RGB components.
24707 The filter accepts the following options:
24709 rc Set the red-cyan contrast. Defaults is 0.0. Allowed range is from
24711 -1.0 to 1.0.
24712 gm Set the green-magenta contrast. Defaults is 0.0. Allowed range is
24714 from -1.0 to 1.0.
24715 by Set the blue-yellow contrast. Defaults is 0.0. Allowed range is
24717 from -1.0 to 1.0.
24718 rcw
24720 gmw
24721 byw Set the weight of each "rc", "gm", "by" option value. Default value
24722 is 0.0. Allowed range is from 0.0 to 1.0. If all weights are 0.0
24723 filtering is disabled.
24724 pl Set the amount of preserving lightness. Default value is 0.0.
24726 Allowed range is from 0.0 to 1.0.
24727 Commands
24729 This filter supports the all above options as commands.
24731 colorcorrect
24733 Adjust color white balance selectively for blacks and whites. This
24734 filter operates in YUV colorspace.
24735 The filter accepts the following options:
24737 rl Set the red shadow spot. Allowed range is from -1.0 to 1.0.
24739 Default value is 0.
24740 bl Set the blue shadow spot. Allowed range is from -1.0 to 1.0.
24742 Default value is 0.
24743 rh Set the red highlight spot. Allowed range is from -1.0 to 1.0.
24745 Default value is 0.
24746 bh Set the red highlight spot. Allowed range is from -1.0 to 1.0.
24748 Default value is 0.
24749 saturation
24751 Set the amount of saturation. Allowed range is from -3.0 to 3.0.
24752 Default value is 1.
24753 analyze
24755 If set to anything other than "manual" it will analyze every frame
24756 and use derived parameters for filtering output frame.
24757 Possible values are:
24759 manual
24761 average
24762 minmax
24763 median
24764 Default value is "manual".
24766 Commands
24768 This filter supports the all above options as commands.
24770 colorchannelmixer
24772 Adjust video input frames by re-mixing color channels.
24773 This filter modifies a color channel by adding the values associated to
24775 the other channels of the same pixels. For example if the value to
24776 modify is red, the output value will be:
24777 <red>=<red>*<rr> + <blue>*<rb> + <green>*<rg> + <alpha>*<ra>
24779 The filter accepts the following options:
24781 rr
24783 rg
24784 rb
24785 ra Adjust contribution of input red, green, blue and alpha channels
24786 for output red channel. Default is 1 for rr, and 0 for rg, rb and
24787 ra.
24788 gr
24790 gg
24791 gb
24792 ga Adjust contribution of input red, green, blue and alpha channels
24793 for output green channel. Default is 1 for gg, and 0 for gr, gb
24794 and ga.
24795 br
24797 bg
24798 bb
24799 ba Adjust contribution of input red, green, blue and alpha channels
24800 for output blue channel. Default is 1 for bb, and 0 for br, bg and
24801 ba.
24802 ar
24804 ag
24805 ab
24806 aa Adjust contribution of input red, green, blue and alpha channels
24807 for output alpha channel. Default is 1 for aa, and 0 for ar, ag
24808 and ab.
24809 Allowed ranges for options are "[-2.0, 2.0]".
24811 pc Set preserve color mode. The accepted values are:
24813 none
24815 Disable color preserving, this is default.
24816 lum Preserve luminance.
24818 max Preserve max value of RGB triplet.
24820 avg Preserve average value of RGB triplet.
24822 sum Preserve sum value of RGB triplet.
24824 nrm Preserve normalized value of RGB triplet.
24826 pwr Preserve power value of RGB triplet.
24828 pa Set the preserve color amount when changing colors. Allowed range
24830 is from "[0.0, 1.0]". Default is 0.0, thus disabled.
24831 Examples
24833 • Convert source to grayscale:
24835 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
24837 • Simulate sepia tones:
24839 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
24841 Commands
24843 This filter supports the all above options as commands.
24845 colorize
24847 Overlay a solid color on the video stream.
24848 The filter accepts the following options:
24850 hue Set the color hue. Allowed range is from 0 to 360. Default value
24852 is 0.
24853 saturation
24855 Set the color saturation. Allowed range is from 0 to 1. Default
24856 value is 0.5.
24857 lightness
24859 Set the color lightness. Allowed range is from 0 to 1. Default
24860 value is 0.5.
24861 mix Set the mix of source lightness. By default is set to 1.0. Allowed
24863 range is from 0.0 to 1.0.
24864 Commands
24866 This filter supports the all above options as commands.
24868 colorkey
24870 RGB colorspace color keying. This filter operates on 8-bit RGB format
24871 frames by setting the alpha component of each pixel which falls within
24872 the similarity radius of the key color to 0. The alpha value for pixels
24873 outside the similarity radius depends on the value of the blend option.
24874 The filter accepts the following options:
24876 color
24878 Set the color for which alpha will be set to 0 (full transparency).
24879 See "Color" section in the ffmpeg-utils manual. Default is
24880 "black".
24881 similarity
24883 Set the radius from the key color within which other colors also
24884 have full transparency. The computed distance is related to the
24885 unit fractional distance in 3D space between the RGB values of the
24886 key color and the pixel's color. Range is 0.01 to 1.0. 0.01 matches
24887 within a very small radius around the exact key color, while 1.0
24888 matches everything. Default is 0.01.
24889 blend
24891 Set how the alpha value for pixels that fall outside the similarity
24892 radius is computed. 0.0 makes pixels either fully transparent or
24893 fully opaque. Higher values result in semi-transparent pixels,
24894 with greater transparency the more similar the pixel color is to
24895 the key color. Range is 0.0 to 1.0. Default is 0.0.
24896 Examples
24898 • Make every green pixel in the input image transparent:
24900 ffmpeg -i input.png -vf colorkey=green out.png
24902 • Overlay a greenscreen-video on top of a static background image.
24904 ffmpeg -i background.png -i video.mp4 -filter_complex "[1:v]colorkey=0x3BBD1E:0.3:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.flv
24906 Commands
24908 This filter supports same commands as options. The command accepts the
24910 same syntax of the corresponding option.
24911 If the specified expression is not valid, it is kept at its current
24913 value.
24914 colorhold
24916 Remove all color information for all RGB colors except for certain one.
24917 The filter accepts the following options:
24919 color
24921 The color which will not be replaced with neutral gray.
24922 similarity
24924 Similarity percentage with the above color. 0.01 matches only the
24925 exact key color, while 1.0 matches everything.
24926 blend
24928 Blend percentage. 0.0 makes pixels fully gray. Higher values
24929 result in more preserved color.
24930 Commands
24932 This filter supports same commands as options. The command accepts the
24934 same syntax of the corresponding option.
24935 If the specified expression is not valid, it is kept at its current
24937 value.
24938 colorlevels
24940 Adjust video input frames using levels.
24941 The filter accepts the following options:
24943 rimin
24945 gimin
24946 bimin
24947 aimin
24948 Adjust red, green, blue and alpha input black point. Allowed
24949 ranges for options are "[-1.0, 1.0]". Defaults are 0.
24950 rimax
24952 gimax
24953 bimax
24954 aimax
24955 Adjust red, green, blue and alpha input white point. Allowed
24956 ranges for options are "[-1.0, 1.0]". Defaults are 1.
24957 Input levels are used to lighten highlights (bright tones), darken
24959 shadows (dark tones), change the balance of bright and dark tones.
24960 romin
24962 gomin
24963 bomin
24964 aomin
24965 Adjust red, green, blue and alpha output black point. Allowed
24966 ranges for options are "[0, 1.0]". Defaults are 0.
24967 romax
24969 gomax
24970 bomax
24971 aomax
24972 Adjust red, green, blue and alpha output white point. Allowed
24973 ranges for options are "[0, 1.0]". Defaults are 1.
24974 Output levels allows manual selection of a constrained output level
24976 range.
24977 preserve
24979 Set preserve color mode. The accepted values are:
24980 none
24982 Disable color preserving, this is default.
24983 lum Preserve luminance.
24985 max Preserve max value of RGB triplet.
24987 avg Preserve average value of RGB triplet.
24989 sum Preserve sum value of RGB triplet.
24991 nrm Preserve normalized value of RGB triplet.
24993 pwr Preserve power value of RGB triplet.
24995 Examples
24997 • Make video output darker:
24999 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
25001 • Increase contrast:
25003 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
25005 • Make video output lighter:
25007 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
25009 • Increase brightness:
25011 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
25013 Commands
25015 This filter supports the all above options as commands.
25017 colormatrix
25019 Convert color matrix.
25020 The filter accepts the following options:
25022 src
25024 dst Specify the source and destination color matrix. Both values must
25025 be specified.
25026 The accepted values are:
25028 bt709
25030 BT.709
25031 fcc FCC
25033 bt601
25035 BT.601
25036 bt470
25038 BT.470
25039 bt470bg
25041 BT.470BG
25042 smpte170m
25044 SMPTE-170M
25045 smpte240m
25047 SMPTE-240M
25048 bt2020
25050 BT.2020
25051 For example to convert from BT.601 to SMPTE-240M, use the command:
25053 colormatrix=bt601:smpte240m
25055 colorspace
25057 Convert colorspace, transfer characteristics or color primaries. Input
25058 video needs to have an even size.
25059 The filter accepts the following options:
25061 all Specify all color properties at once.
25063 The accepted values are:
25065 bt470m
25067 BT.470M
25068 bt470bg
25070 BT.470BG
25071 bt601-6-525
25073 BT.601-6 525
25074 bt601-6-625
25076 BT.601-6 625
25077 bt709
25079 BT.709
25080 smpte170m
25082 SMPTE-170M
25083 smpte240m
25085 SMPTE-240M
25086 bt2020
25088 BT.2020
25089 space
25091 Specify output colorspace.
25092 The accepted values are:
25094 bt709
25096 BT.709
25097 fcc FCC
25099 bt470bg
25101 BT.470BG or BT.601-6 625
25102 smpte170m
25104 SMPTE-170M or BT.601-6 525
25105 smpte240m
25107 SMPTE-240M
25108 ycgco
25110 YCgCo
25111 bt2020ncl
25113 BT.2020 with non-constant luminance
25114 trc Specify output transfer characteristics.
25116 The accepted values are:
25118 bt709
25120 BT.709
25121 bt470m
25123 BT.470M
25124 bt470bg
25126 BT.470BG
25127 gamma22
25129 Constant gamma of 2.2
25130 gamma28
25132 Constant gamma of 2.8
25133 smpte170m
25135 SMPTE-170M, BT.601-6 625 or BT.601-6 525
25136 smpte240m
25138 SMPTE-240M
25139 srgb
25141 SRGB
25142 iec61966-2-1
25144 iec61966-2-1
25145 iec61966-2-4
25147 iec61966-2-4
25148 xvycc
25150 xvycc
25151 bt2020-10
25153 BT.2020 for 10-bits content
25154 bt2020-12
25156 BT.2020 for 12-bits content
25157 primaries
25159 Specify output color primaries.
25160 The accepted values are:
25162 bt709
25164 BT.709
25165 bt470m
25167 BT.470M
25168 bt470bg
25170 BT.470BG or BT.601-6 625
25171 smpte170m
25173 SMPTE-170M or BT.601-6 525
25174 smpte240m
25176 SMPTE-240M
25177 film
25179 film
25180 smpte431
25182 SMPTE-431
25183 smpte432
25185 SMPTE-432
25186 bt2020
25188 BT.2020
25189 jedec-p22
25191 JEDEC P22 phosphors
25192 range
25194 Specify output color range.
25195 The accepted values are:
25197 tv TV (restricted) range
25199 mpeg
25201 MPEG (restricted) range
25202 pc PC (full) range
25204 jpeg
25206 JPEG (full) range
25207 format
25209 Specify output color format.
25210 The accepted values are:
25212 yuv420p
25214 YUV 4:2:0 planar 8-bits
25215 yuv420p10
25217 YUV 4:2:0 planar 10-bits
25218 yuv420p12
25220 YUV 4:2:0 planar 12-bits
25221 yuv422p
25223 YUV 4:2:2 planar 8-bits
25224 yuv422p10
25226 YUV 4:2:2 planar 10-bits
25227 yuv422p12
25229 YUV 4:2:2 planar 12-bits
25230 yuv444p
25232 YUV 4:4:4 planar 8-bits
25233 yuv444p10
25235 YUV 4:4:4 planar 10-bits
25236 yuv444p12
25238 YUV 4:4:4 planar 12-bits
25239 fast
25241 Do a fast conversion, which skips gamma/primary correction. This
25242 will take significantly less CPU, but will be mathematically
25243 incorrect. To get output compatible with that produced by the
25244 colormatrix filter, use fast=1.
25245 dither
25247 Specify dithering mode.
25248 The accepted values are:
25250 none
25252 No dithering
25253 fsb Floyd-Steinberg dithering
25255 wpadapt
25257 Whitepoint adaptation mode.
25258 The accepted values are:
25260 bradford
25262 Bradford whitepoint adaptation
25263 vonkries
25265 von Kries whitepoint adaptation
25266 identity
25268 identity whitepoint adaptation (i.e. no whitepoint adaptation)
25269 iall
25271 Override all input properties at once. Same accepted values as all.
25272 ispace
25274 Override input colorspace. Same accepted values as space.
25275 iprimaries
25277 Override input color primaries. Same accepted values as primaries.
25278 itrc
25280 Override input transfer characteristics. Same accepted values as
25281 trc.
25282 irange
25284 Override input color range. Same accepted values as range.
25285 The filter converts the transfer characteristics, color space and color
25287 primaries to the specified user values. The output value, if not
25288 specified, is set to a default value based on the "all" property. If
25289 that property is also not specified, the filter will log an error. The
25290 output color range and format default to the same value as the input
25291 color range and format. The input transfer characteristics, color
25292 space, color primaries and color range should be set on the input data.
25293 If any of these are missing, the filter will log an error and no
25294 conversion will take place.
25295 For example to convert the input to SMPTE-240M, use the command:
25297 colorspace=smpte240m
25299 colortemperature
25301 Adjust color temperature in video to simulate variations in ambient
25302 color temperature.
25303 The filter accepts the following options:
25305 temperature
25307 Set the temperature in Kelvin. Allowed range is from 1000 to 40000.
25308 Default value is 6500 K.
25309 mix Set mixing with filtered output. Allowed range is from 0 to 1.
25311 Default value is 1.
25312 pl Set the amount of preserving lightness. Allowed range is from 0 to
25314 1. Default value is 0.
25315 Commands
25317 This filter supports same commands as options.
25319 convolution
25321 Apply convolution of 3x3, 5x5, 7x7 or horizontal/vertical up to 49
25322 elements.
25323 The filter accepts the following options:
25325 0m
25327 1m
25328 2m
25329 3m Set matrix for each plane. Matrix is sequence of 9, 25 or 49
25330 signed integers in square mode, and from 1 to 49 odd number of
25331 signed integers in row mode.
25332 0rdiv
25334 1rdiv
25335 2rdiv
25336 3rdiv
25337 Set multiplier for calculated value for each plane. If unset or 0,
25338 it will be sum of all matrix elements.
25339 0bias
25341 1bias
25342 2bias
25343 3bias
25344 Set bias for each plane. This value is added to the result of the
25345 multiplication. Useful for making the overall image brighter or
25346 darker. Default is 0.0.
25347 0mode
25349 1mode
25350 2mode
25351 3mode
25352 Set matrix mode for each plane. Can be square, row or column.
25353 Default is square.
25354 Commands
25356 This filter supports the all above options as commands.
25358 Examples
25360 • Apply sharpen:
25362 convolution="0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0"
25364 • Apply blur:
25366 convolution="1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1/9:1/9:1/9:1/9"
25368 • Apply edge enhance:
25370 convolution="0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:5:1:1:1:0:128:128:128"
25372 • Apply edge detect:
25374 convolution="0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:5:5:5:1:0:128:128:128"
25376 • Apply laplacian edge detector which includes diagonals:
25378 convolution="1 1 1 1 -8 1 1 1 1:1 1 1 1 -8 1 1 1 1:1 1 1 1 -8 1 1 1 1:1 1 1 1 -8 1 1 1 1:5:5:5:1:0:128:128:0"
25380 • Apply emboss:
25382 convolution="-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2"
25384 convolve
25386 Apply 2D convolution of video stream in frequency domain using second
25387 stream as impulse.
25388 The filter accepts the following options:
25390 planes
25392 Set which planes to process.
25393 impulse
25395 Set which impulse video frames will be processed, can be first or
25396 all. Default is all.
25397 The "convolve" filter also supports the framesync options.
25399 copy
25401 Copy the input video source unchanged to the output. This is mainly
25402 useful for testing purposes.
25403 coreimage
25405 Video filtering on GPU using Apple's CoreImage API on OSX.
25406 Hardware acceleration is based on an OpenGL context. Usually, this
25408 means it is processed by video hardware. However, software-based OpenGL
25409 implementations exist which means there is no guarantee for hardware
25410 processing. It depends on the respective OSX.
25411 There are many filters and image generators provided by Apple that come
25413 with a large variety of options. The filter has to be referenced by its
25414 name along with its options.
25415 The coreimage filter accepts the following options:
25417 list_filters
25419 List all available filters and generators along with all their
25420 respective options as well as possible minimum and maximum values
25421 along with the default values.
25422 list_filters=true
25424 filter
25426 Specify all filters by their respective name and options. Use
25427 list_filters to determine all valid filter names and options.
25428 Numerical options are specified by a float value and are
25429 automatically clamped to their respective value range. Vector and
25430 color options have to be specified by a list of space separated
25431 float values. Character escaping has to be done. A special option
25432 name "default" is available to use default options for a filter.
25433 It is required to specify either "default" or at least one of the
25435 filter options. All omitted options are used with their default
25436 values. The syntax of the filter string is as follows:
25437 filter=<NAME>@<OPTION>=<VALUE>[@<OPTION>=<VALUE>][@...][#<NAME>@<OPTION>=<VALUE>[@<OPTION>=<VALUE>][@...]][#...]
25439 output_rect
25441 Specify a rectangle where the output of the filter chain is copied
25442 into the input image. It is given by a list of space separated
25443 float values:
25444 output_rect=x\ y\ width\ height
25446 If not given, the output rectangle equals the dimensions of the
25448 input image. The output rectangle is automatically cropped at the
25449 borders of the input image. Negative values are valid for each
25450 component.
25451 output_rect=25\ 25\ 100\ 100
25453 Several filters can be chained for successive processing without GPU-
25455 HOST transfers allowing for fast processing of complex filter chains.
25456 Currently, only filters with zero (generators) or exactly one (filters)
25457 input image and one output image are supported. Also, transition
25458 filters are not yet usable as intended.
25459 Some filters generate output images with additional padding depending
25461 on the respective filter kernel. The padding is automatically removed
25462 to ensure the filter output has the same size as the input image.
25463 For image generators, the size of the output image is determined by the
25465 previous output image of the filter chain or the input image of the
25466 whole filterchain, respectively. The generators do not use the pixel
25467 information of this image to generate their output. However, the
25468 generated output is blended onto this image, resulting in partial or
25469 complete coverage of the output image.
25470 The coreimagesrc video source can be used for generating input images
25472 which are directly fed into the filter chain. By using it, providing
25473 input images by another video source or an input video is not required.
25474 Examples
25476 • List all filters available:
25478 coreimage=list_filters=true
25480 • Use the CIBoxBlur filter with default options to blur an image:
25482 coreimage=filter=CIBoxBlur@default
25484 • Use a filter chain with CISepiaTone at default values and
25486 CIVignetteEffect with its center at 100x100 and a radius of 50
25487 pixels:
25488 coreimage=filter=CIBoxBlur@default#CIVignetteEffect@inputCenter=100\ 100@inputRadius=50
25490 • Use nullsrc and CIQRCodeGenerator to create a QR code for the
25492 FFmpeg homepage, given as complete and escaped command-line for
25493 Apple's standard bash shell:
25494 ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@inputMessage=https\\\\\://FFmpeg.org/@inputCorrectionLevel=H -frames:v 1 QRCode.png
25496 cover_rect
25498 Cover a rectangular object
25499 It accepts the following options:
25501 cover
25503 Filepath of the optional cover image, needs to be in yuv420.
25504 mode
25506 Set covering mode.
25507 It accepts the following values:
25509 cover
25511 cover it by the supplied image
25512 blur
25514 cover it by interpolating the surrounding pixels
25515 Default value is blur.
25517 Examples
25519 • Cover a rectangular object by the supplied image of a given video
25521 using ffmpeg:
25522 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
25524 crop
25526 Crop the input video to given dimensions.
25527 It accepts the following parameters:
25529 w, out_w
25531 The width of the output video. It defaults to "iw". This
25532 expression is evaluated only once during the filter configuration,
25533 or when the w or out_w command is sent.
25534 h, out_h
25536 The height of the output video. It defaults to "ih". This
25537 expression is evaluated only once during the filter configuration,
25538 or when the h or out_h command is sent.
25539 x The horizontal position, in the input video, of the left edge of
25541 the output video. It defaults to "(in_w-out_w)/2". This expression
25542 is evaluated per-frame.
25543 y The vertical position, in the input video, of the top edge of the
25545 output video. It defaults to "(in_h-out_h)/2". This expression is
25546 evaluated per-frame.
25547 keep_aspect
25549 If set to 1 will force the output display aspect ratio to be the
25550 same of the input, by changing the output sample aspect ratio. It
25551 defaults to 0.
25552 exact
25554 Enable exact cropping. If enabled, subsampled videos will be
25555 cropped at exact width/height/x/y as specified and will not be
25556 rounded to nearest smaller value. It defaults to 0.
25557 The out_w, out_h, x, y parameters are expressions containing the
25559 following constants:
25560 x
25562 y The computed values for x and y. They are evaluated for each new
25563 frame.
25564 in_w
25566 in_h
25567 The input width and height.
25568 iw
25570 ih These are the same as in_w and in_h.
25571 out_w
25573 out_h
25574 The output (cropped) width and height.
25575 ow
25577 oh These are the same as out_w and out_h.
25578 a same as iw / ih
25580 sar input sample aspect ratio
25582 dar input display aspect ratio, it is the same as (iw / ih) * sar
25584 hsub
25586 vsub
25587 horizontal and vertical chroma subsample values. For example for
25588 the pixel format "yuv422p" hsub is 2 and vsub is 1.
25589 n The number of the input frame, starting from 0.
25591 pos the position in the file of the input frame, NAN if unknown
25593 t The timestamp expressed in seconds. It's NAN if the input timestamp
25595 is unknown.
25596 The expression for out_w may depend on the value of out_h, and the
25598 expression for out_h may depend on out_w, but they cannot depend on x
25599 and y, as x and y are evaluated after out_w and out_h.
25600 The x and y parameters specify the expressions for the position of the
25602 top-left corner of the output (non-cropped) area. They are evaluated
25603 for each frame. If the evaluated value is not valid, it is approximated
25604 to the nearest valid value.
25605 The expression for x may depend on y, and the expression for y may
25607 depend on x.
25608 Examples
25610 • Crop area with size 100x100 at position (12,34).
25612 crop=100:100:12:34
25614 Using named options, the example above becomes:
25616 crop=w=100:h=100:x=12:y=34
25618 • Crop the central input area with size 100x100:
25620 crop=100:100
25622 • Crop the central input area with size 2/3 of the input video:
25624 crop=2/3*in_w:2/3*in_h
25626 • Crop the input video central square:
25628 crop=out_w=in_h
25630 crop=in_h
25631 • Delimit the rectangle with the top-left corner placed at position
25633 100:100 and the right-bottom corner corresponding to the right-
25634 bottom corner of the input image.
25635 crop=in_w-100:in_h-100:100:100
25637 • Crop 10 pixels from the left and right borders, and 20 pixels from
25639 the top and bottom borders
25640 crop=in_w-2*10:in_h-2*20
25642 • Keep only the bottom right quarter of the input image:
25644 crop=in_w/2:in_h/2:in_w/2:in_h/2
25646 • Crop height for getting Greek harmony:
25648 crop=in_w:1/PHI*in_w
25650 • Apply trembling effect:
25652 crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(n/10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(n/7)
25654 • Apply erratic camera effect depending on timestamp:
25656 crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(t*10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(t*13)"
25658 • Set x depending on the value of y:
25660 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
25662 Commands
25664 This filter supports the following commands:
25666 w, out_w
25668 h, out_h
25669 x
25670 y Set width/height of the output video and the horizontal/vertical
25671 position in the input video. The command accepts the same syntax
25672 of the corresponding option.
25673 If the specified expression is not valid, it is kept at its current
25675 value.
25676 cropdetect
25678 Auto-detect the crop size.
25679 It calculates the necessary cropping parameters and prints the
25681 recommended parameters via the logging system. The detected dimensions
25682 correspond to the non-black area of the input video.
25683 It accepts the following parameters:
25685 limit
25687 Set higher black value threshold, which can be optionally specified
25688 from nothing (0) to everything (255 for 8-bit based formats). An
25689 intensity value greater to the set value is considered non-black.
25690 It defaults to 24. You can also specify a value between 0.0 and
25691 1.0 which will be scaled depending on the bitdepth of the pixel
25692 format.
25693 round
25695 The value which the width/height should be divisible by. It
25696 defaults to 16. The offset is automatically adjusted to center the
25697 video. Use 2 to get only even dimensions (needed for 4:2:2 video).
25698 16 is best when encoding to most video codecs.
25699 skip
25701 Set the number of initial frames for which evaluation is skipped.
25702 Default is 2. Range is 0 to INT_MAX.
25703 reset_count, reset
25705 Set the counter that determines after how many frames cropdetect
25706 will reset the previously detected largest video area and start
25707 over to detect the current optimal crop area. Default value is 0.
25708 This can be useful when channel logos distort the video area. 0
25710 indicates 'never reset', and returns the largest area encountered
25711 during playback.
25712 cue
25714 Delay video filtering until a given wallclock timestamp. The filter
25715 first passes on preroll amount of frames, then it buffers at most
25716 buffer amount of frames and waits for the cue. After reaching the cue
25717 it forwards the buffered frames and also any subsequent frames coming
25718 in its input.
25719 The filter can be used synchronize the output of multiple ffmpeg
25721 processes for realtime output devices like decklink. By putting the
25722 delay in the filtering chain and pre-buffering frames the process can
25723 pass on data to output almost immediately after the target wallclock
25724 timestamp is reached.
25725 Perfect frame accuracy cannot be guaranteed, but the result is good
25727 enough for some use cases.
25728 cue The cue timestamp expressed in a UNIX timestamp in microseconds.
25730 Default is 0.
25731 preroll
25733 The duration of content to pass on as preroll expressed in seconds.
25734 Default is 0.
25735 buffer
25737 The maximum duration of content to buffer before waiting for the
25738 cue expressed in seconds. Default is 0.
25739 curves
25741 Apply color adjustments using curves.
25742 This filter is similar to the Adobe Photoshop and GIMP curves tools.
25744 Each component (red, green and blue) has its values defined by N key
25745 points tied from each other using a smooth curve. The x-axis represents
25746 the pixel values from the input frame, and the y-axis the new pixel
25747 values to be set for the output frame.
25748 By default, a component curve is defined by the two points (0;0) and
25750 (1;1). This creates a straight line where each original pixel value is
25751 "adjusted" to its own value, which means no change to the image.
25752 The filter allows you to redefine these two points and add some more. A
25754 new curve (using a natural cubic spline interpolation) will be define
25755 to pass smoothly through all these new coordinates. The new defined
25756 points needs to be strictly increasing over the x-axis, and their x and
25757 y values must be in the [0;1] interval. If the computed curves
25758 happened to go outside the vector spaces, the values will be clipped
25759 accordingly.
25760 The filter accepts the following options:
25762 preset
25764 Select one of the available color presets. This option can be used
25765 in addition to the r, g, b parameters; in this case, the later
25766 options takes priority on the preset values. Available presets
25767 are:
25768 none
25770 color_negative
25771 cross_process
25772 darker
25773 increase_contrast
25774 lighter
25775 linear_contrast
25776 medium_contrast
25777 negative
25778 strong_contrast
25779 vintage
25780 Default is "none".
25782 master, m
25784 Set the master key points. These points will define a second pass
25785 mapping. It is sometimes called a "luminance" or "value" mapping.
25786 It can be used with r, g, b or all since it acts like a post-
25787 processing LUT.
25788 red, r
25790 Set the key points for the red component.
25791 green, g
25793 Set the key points for the green component.
25794 blue, b
25796 Set the key points for the blue component.
25797 all Set the key points for all components (not including master). Can
25799 be used in addition to the other key points component options. In
25800 this case, the unset component(s) will fallback on this all
25801 setting.
25802 psfile
25804 Specify a Photoshop curves file (".acv") to import the settings
25805 from.
25806 plot
25808 Save Gnuplot script of the curves in specified file.
25809 To avoid some filtergraph syntax conflicts, each key points list need
25811 to be defined using the following syntax: "x0/y0 x1/y1 x2/y2 ...".
25812 Commands
25814 This filter supports same commands as options.
25816 Examples
25818 • Increase slightly the middle level of blue:
25820 curves=blue='0/0 0.5/0.58 1/1'
25822 • Vintage effect:
25824 curves=r='0/0.11 .42/.51 1/0.95':g='0/0 0.50/0.48 1/1':b='0/0.22 .49/.44 1/0.8'
25826 Here we obtain the following coordinates for each components:
25828 red "(0;0.11) (0.42;0.51) (1;0.95)"
25830 green
25832 "(0;0) (0.50;0.48) (1;1)"
25833 blue
25835 "(0;0.22) (0.49;0.44) (1;0.80)"
25836 • The previous example can also be achieved with the associated
25838 built-in preset:
25839 curves=preset=vintage
25841 • Or simply:
25843 curves=vintage
25845 • Use a Photoshop preset and redefine the points of the green
25847 component:
25848 curves=psfile='MyCurvesPresets/purple.acv':green='0/0 0.45/0.53 1/1'
25850 • Check out the curves of the "cross_process" profile using ffmpeg
25852 and gnuplot:
25853 ffmpeg -f lavfi -i color -vf curves=cross_process:plot=/tmp/curves.plt -frames:v 1 -f null -
25855 gnuplot -p /tmp/curves.plt
25856 datascope
25858 Video data analysis filter.
25859 This filter shows hexadecimal pixel values of part of video.
25861 The filter accepts the following options:
25863 size, s
25865 Set output video size.
25866 x Set x offset from where to pick pixels.
25868 y Set y offset from where to pick pixels.
25870 mode
25872 Set scope mode, can be one of the following:
25873 mono
25875 Draw hexadecimal pixel values with white color on black
25876 background.
25877 color
25879 Draw hexadecimal pixel values with input video pixel color on
25880 black background.
25881 color2
25883 Draw hexadecimal pixel values on color background picked from
25884 input video, the text color is picked in such way so its always
25885 visible.
25886 axis
25888 Draw rows and columns numbers on left and top of video.
25889 opacity
25891 Set background opacity.
25892 format
25894 Set display number format. Can be "hex", or "dec". Default is
25895 "hex".
25896 components
25898 Set pixel components to display. By default all pixel components
25899 are displayed.
25900 Commands
25902 This filter supports same commands as options excluding "size" option.
25904 dblur
25906 Apply Directional blur filter.
25907 The filter accepts the following options:
25909 angle
25911 Set angle of directional blur. Default is 45.
25912 radius
25914 Set radius of directional blur. Default is 5.
25915 planes
25917 Set which planes to filter. By default all planes are filtered.
25918 Commands
25920 This filter supports same commands as options. The command accepts the
25922 same syntax of the corresponding option.
25923 If the specified expression is not valid, it is kept at its current
25925 value.
25926 dctdnoiz
25928 Denoise frames using 2D DCT (frequency domain filtering).
25929 This filter is not designed for real time.
25931 The filter accepts the following options:
25933 sigma, s
25935 Set the noise sigma constant.
25936 This sigma defines a hard threshold of "3 * sigma"; every DCT
25938 coefficient (absolute value) below this threshold with be dropped.
25939 If you need a more advanced filtering, see expr.
25941 Default is 0.
25943 overlap
25945 Set number overlapping pixels for each block. Since the filter can
25946 be slow, you may want to reduce this value, at the cost of a less
25947 effective filter and the risk of various artefacts.
25948 If the overlapping value doesn't permit processing the whole input
25950 width or height, a warning will be displayed and according borders
25951 won't be denoised.
25952 Default value is blocksize-1, which is the best possible setting.
25954 expr, e
25956 Set the coefficient factor expression.
25957 For each coefficient of a DCT block, this expression will be
25959 evaluated as a multiplier value for the coefficient.
25960 If this is option is set, the sigma option will be ignored.
25962 The absolute value of the coefficient can be accessed through the c
25964 variable.
25965 n Set the blocksize using the number of bits. "1<<n" defines the
25967 blocksize, which is the width and height of the processed blocks.
25968 The default value is 3 (8x8) and can be raised to 4 for a blocksize
25970 of 16x16. Note that changing this setting has huge consequences on
25971 the speed processing. Also, a larger block size does not
25972 necessarily means a better de-noising.
25973 Examples
25975 Apply a denoise with a sigma of 4.5:
25977 dctdnoiz=4.5
25979 The same operation can be achieved using the expression system:
25981 dctdnoiz=e='gte(c, 4.5*3)'
25983 Violent denoise using a block size of "16x16":
25985 dctdnoiz=15:n=4
25987 deband
25989 Remove banding artifacts from input video. It works by replacing
25990 banded pixels with average value of referenced pixels.
25991 The filter accepts the following options:
25993 1thr
25995 2thr
25996 3thr
25997 4thr
25998 Set banding detection threshold for each plane. Default is 0.02.
25999 Valid range is 0.00003 to 0.5. If difference between current pixel
26000 and reference pixel is less than threshold, it will be considered
26001 as banded.
26002 range, r
26004 Banding detection range in pixels. Default is 16. If positive,
26005 random number in range 0 to set value will be used. If negative,
26006 exact absolute value will be used. The range defines square of
26007 four pixels around current pixel.
26008 direction, d
26010 Set direction in radians from which four pixel will be compared. If
26011 positive, random direction from 0 to set direction will be picked.
26012 If negative, exact of absolute value will be picked. For example
26013 direction 0, -PI or -2*PI radians will pick only pixels on same row
26014 and -PI/2 will pick only pixels on same column.
26015 blur, b
26017 If enabled, current pixel is compared with average value of all
26018 four surrounding pixels. The default is enabled. If disabled
26019 current pixel is compared with all four surrounding pixels. The
26020 pixel is considered banded if only all four differences with
26021 surrounding pixels are less than threshold.
26022 coupling, c
26024 If enabled, current pixel is changed if and only if all pixel
26025 components are banded, e.g. banding detection threshold is
26026 triggered for all color components. The default is disabled.
26027 Commands
26029 This filter supports the all above options as commands.
26031 deblock
26033 Remove blocking artifacts from input video.
26034 The filter accepts the following options:
26036 filter
26038 Set filter type, can be weak or strong. Default is strong. This
26039 controls what kind of deblocking is applied.
26040 block
26042 Set size of block, allowed range is from 4 to 512. Default is 8.
26043 alpha
26045 beta
26046 gamma
26047 delta
26048 Set blocking detection thresholds. Allowed range is 0 to 1.
26049 Defaults are: 0.098 for alpha and 0.05 for the rest. Using higher
26050 threshold gives more deblocking strength. Setting alpha controls
26051 threshold detection at exact edge of block. Remaining options
26052 controls threshold detection near the edge. Each one for
26053 below/above or left/right. Setting any of those to 0 disables
26054 deblocking.
26055 planes
26057 Set planes to filter. Default is to filter all available planes.
26058 Examples
26060 • Deblock using weak filter and block size of 4 pixels.
26062 deblock=filter=weak:block=4
26064 • Deblock using strong filter, block size of 4 pixels and custom
26066 thresholds for deblocking more edges.
26067 deblock=filter=strong:block=4:alpha=0.12:beta=0.07:gamma=0.06:delta=0.05
26069 • Similar as above, but filter only first plane.
26071 deblock=filter=strong:block=4:alpha=0.12:beta=0.07:gamma=0.06:delta=0.05:planes=1
26073 • Similar as above, but filter only second and third plane.
26075 deblock=filter=strong:block=4:alpha=0.12:beta=0.07:gamma=0.06:delta=0.05:planes=6
26077 Commands
26079 This filter supports the all above options as commands.
26081 decimate
26083 Drop duplicated frames at regular intervals.
26084 The filter accepts the following options:
26086 cycle
26088 Set the number of frames from which one will be dropped. Setting
26089 this to N means one frame in every batch of N frames will be
26090 dropped. Default is 5.
26091 dupthresh
26093 Set the threshold for duplicate detection. If the difference metric
26094 for a frame is less than or equal to this value, then it is
26095 declared as duplicate. Default is 1.1
26096 scthresh
26098 Set scene change threshold. Default is 15.
26099 blockx
26101 blocky
26102 Set the size of the x and y-axis blocks used during metric
26103 calculations. Larger blocks give better noise suppression, but
26104 also give worse detection of small movements. Must be a power of
26105 two. Default is 32.
26106 ppsrc
26108 Mark main input as a pre-processed input and activate clean source
26109 input stream. This allows the input to be pre-processed with
26110 various filters to help the metrics calculation while keeping the
26111 frame selection lossless. When set to 1, the first stream is for
26112 the pre-processed input, and the second stream is the clean source
26113 from where the kept frames are chosen. Default is 0.
26114 chroma
26116 Set whether or not chroma is considered in the metric calculations.
26117 Default is 1.
26118 deconvolve
26120 Apply 2D deconvolution of video stream in frequency domain using second
26121 stream as impulse.
26122 The filter accepts the following options:
26124 planes
26126 Set which planes to process.
26127 impulse
26129 Set which impulse video frames will be processed, can be first or
26130 all. Default is all.
26131 noise
26133 Set noise when doing divisions. Default is 0.0000001. Useful when
26134 width and height are not same and not power of 2 or if stream prior
26135 to convolving had noise.
26136 The "deconvolve" filter also supports the framesync options.
26138 dedot
26140 Reduce cross-luminance (dot-crawl) and cross-color (rainbows) from
26141 video.
26142 It accepts the following options:
26144 m Set mode of operation. Can be combination of dotcrawl for cross-
26146 luminance reduction and/or rainbows for cross-color reduction.
26147 lt Set spatial luma threshold. Lower values increases reduction of
26149 cross-luminance.
26150 tl Set tolerance for temporal luma. Higher values increases reduction
26152 of cross-luminance.
26153 tc Set tolerance for chroma temporal variation. Higher values
26155 increases reduction of cross-color.
26156 ct Set temporal chroma threshold. Lower values increases reduction of
26158 cross-color.
26159 deflate
26161 Apply deflate effect to the video.
26162 This filter replaces the pixel by the local(3x3) average by taking into
26164 account only values lower than the pixel.
26165 It accepts the following options:
26167 threshold0
26169 threshold1
26170 threshold2
26171 threshold3
26172 Limit the maximum change for each plane, default is 65535. If 0,
26173 plane will remain unchanged.
26174 Commands
26176 This filter supports the all above options as commands.
26178 deflicker
26180 Remove temporal frame luminance variations.
26181 It accepts the following options:
26183 size, s
26185 Set moving-average filter size in frames. Default is 5. Allowed
26186 range is 2 - 129.
26187 mode, m
26189 Set averaging mode to smooth temporal luminance variations.
26190 Available values are:
26192 am Arithmetic mean
26194 gm Geometric mean
26196 hm Harmonic mean
26198 qm Quadratic mean
26200 cm Cubic mean
26202 pm Power mean
26204 median
26206 Median
26207 bypass
26209 Do not actually modify frame. Useful when one only wants metadata.
26210 dejudder
26212 Remove judder produced by partially interlaced telecined content.
26213 Judder can be introduced, for instance, by pullup filter. If the
26215 original source was partially telecined content then the output of
26216 "pullup,dejudder" will have a variable frame rate. May change the
26217 recorded frame rate of the container. Aside from that change, this
26218 filter will not affect constant frame rate video.
26219 The option available in this filter is:
26221 cycle
26223 Specify the length of the window over which the judder repeats.
26224 Accepts any integer greater than 1. Useful values are:
26226 4 If the original was telecined from 24 to 30 fps (Film to NTSC).
26228 5 If the original was telecined from 25 to 30 fps (PAL to NTSC).
26230 20 If a mixture of the two.
26232 The default is 4.
26234 delogo
26236 Suppress a TV station logo by a simple interpolation of the surrounding
26237 pixels. Just set a rectangle covering the logo and watch it disappear
26238 (and sometimes something even uglier appear - your mileage may vary).
26239 It accepts the following parameters:
26241 x
26243 y Specify the top left corner coordinates of the logo. They must be
26244 specified.
26245 w
26247 h Specify the width and height of the logo to clear. They must be
26248 specified.
26249 show
26251 When set to 1, a green rectangle is drawn on the screen to simplify
26252 finding the right x, y, w, and h parameters. The default value is
26253 0.
26254 The rectangle is drawn on the outermost pixels which will be
26256 (partly) replaced with interpolated values. The values of the next
26257 pixels immediately outside this rectangle in each direction will be
26258 used to compute the interpolated pixel values inside the rectangle.
26259 Examples
26261 • Set a rectangle covering the area with top left corner coordinates
26263 0,0 and size 100x77:
26264 delogo=x=0:y=0:w=100:h=77
26266 derain
26268 Remove the rain in the input image/video by applying the derain methods
26269 based on convolutional neural networks. Supported models:
26270 • Recurrent Squeeze-and-Excitation Context Aggregation Net (RESCAN).
26272 See
26273 <http://openaccess.thecvf.com/content_ECCV_2018/papers/Xia_Li_Recurrent_Squeeze-and-Excitation_Context_ECCV_2018_paper.pdf>.
26274 Training as well as model generation scripts are provided in the
26276 repository at <https://github.com/XueweiMeng/derain_filter.git>.
26277 Native model files (.model) can be generated from TensorFlow model
26279 files (.pb) by using tools/python/convert.py
26280 The filter accepts the following options:
26282 filter_type
26284 Specify which filter to use. This option accepts the following
26285 values:
26286 derain
26288 Derain filter. To conduct derain filter, you need to use a
26289 derain model.
26290 dehaze
26292 Dehaze filter. To conduct dehaze filter, you need to use a
26293 dehaze model.
26294 Default value is derain.
26296 dnn_backend
26298 Specify which DNN backend to use for model loading and execution.
26299 This option accepts the following values:
26300 native
26302 Native implementation of DNN loading and execution.
26303 tensorflow
26305 TensorFlow backend. To enable this backend you need to install
26306 the TensorFlow for C library (see
26307 <https://www.tensorflow.org/install/lang_c>) and configure
26308 FFmpeg with "--enable-libtensorflow"
26309 Default value is native.
26311 model
26313 Set path to model file specifying network architecture and its
26314 parameters. Note that different backends use different file
26315 formats. TensorFlow and native backend can load files for only its
26316 format.
26317 To get full functionality (such as async execution), please use the
26319 dnn_processing filter.
26320 deshake
26322 Attempt to fix small changes in horizontal and/or vertical shift. This
26323 filter helps remove camera shake from hand-holding a camera, bumping a
26324 tripod, moving on a vehicle, etc.
26325 The filter accepts the following options:
26327 x
26329 y
26330 w
26331 h Specify a rectangular area where to limit the search for motion
26332 vectors. If desired the search for motion vectors can be limited
26333 to a rectangular area of the frame defined by its top left corner,
26334 width and height. These parameters have the same meaning as the
26335 drawbox filter which can be used to visualise the position of the
26336 bounding box.
26337 This is useful when simultaneous movement of subjects within the
26339 frame might be confused for camera motion by the motion vector
26340 search.
26341 If any or all of x, y, w and h are set to -1 then the full frame is
26343 used. This allows later options to be set without specifying the
26344 bounding box for the motion vector search.
26345 Default - search the whole frame.
26347 rx
26349 ry Specify the maximum extent of movement in x and y directions in the
26350 range 0-64 pixels. Default 16.
26351 edge
26353 Specify how to generate pixels to fill blanks at the edge of the
26354 frame. Available values are:
26355 blank, 0
26357 Fill zeroes at blank locations
26358 original, 1
26360 Original image at blank locations
26361 clamp, 2
26363 Extruded edge value at blank locations
26364 mirror, 3
26366 Mirrored edge at blank locations
26367 Default value is mirror.
26369 blocksize
26371 Specify the blocksize to use for motion search. Range 4-128 pixels,
26372 default 8.
26373 contrast
26375 Specify the contrast threshold for blocks. Only blocks with more
26376 than the specified contrast (difference between darkest and
26377 lightest pixels) will be considered. Range 1-255, default 125.
26378 search
26380 Specify the search strategy. Available values are:
26381 exhaustive, 0
26383 Set exhaustive search
26384 less, 1
26386 Set less exhaustive search.
26387 Default value is exhaustive.
26389 filename
26391 If set then a detailed log of the motion search is written to the
26392 specified file.
26393 despill
26395 Remove unwanted contamination of foreground colors, caused by reflected
26396 color of greenscreen or bluescreen.
26397 This filter accepts the following options:
26399 type
26401 Set what type of despill to use.
26402 mix Set how spillmap will be generated.
26404 expand
26406 Set how much to get rid of still remaining spill.
26407 red Controls amount of red in spill area.
26409 green
26411 Controls amount of green in spill area. Should be -1 for
26412 greenscreen.
26413 blue
26415 Controls amount of blue in spill area. Should be -1 for
26416 bluescreen.
26417 brightness
26419 Controls brightness of spill area, preserving colors.
26420 alpha
26422 Modify alpha from generated spillmap.
26423 Commands
26425 This filter supports the all above options as commands.
26427 detelecine
26429 Apply an exact inverse of the telecine operation. It requires a
26430 predefined pattern specified using the pattern option which must be the
26431 same as that passed to the telecine filter.
26432 This filter accepts the following options:
26434 first_field
26436 top, t
26437 top field first
26438 bottom, b
26440 bottom field first The default value is "top".
26441 pattern
26443 A string of numbers representing the pulldown pattern you wish to
26444 apply. The default value is 23.
26445 start_frame
26447 A number representing position of the first frame with respect to
26448 the telecine pattern. This is to be used if the stream is cut. The
26449 default value is 0.
26450 dilation
26452 Apply dilation effect to the video.
26453 This filter replaces the pixel by the local(3x3) maximum.
26455 It accepts the following options:
26457 threshold0
26459 threshold1
26460 threshold2
26461 threshold3
26462 Limit the maximum change for each plane, default is 65535. If 0,
26463 plane will remain unchanged.
26464 coordinates
26466 Flag which specifies the pixel to refer to. Default is 255 i.e. all
26467 eight pixels are used.
26468 Flags to local 3x3 coordinates maps like this:
26470 1 2 3
26472 4 5
26473 6 7 8
26474 Commands
26476 This filter supports the all above options as commands.
26478 displace
26480 Displace pixels as indicated by second and third input stream.
26481 It takes three input streams and outputs one stream, the first input is
26483 the source, and second and third input are displacement maps.
26484 The second input specifies how much to displace pixels along the
26486 x-axis, while the third input specifies how much to displace pixels
26487 along the y-axis. If one of displacement map streams terminates, last
26488 frame from that displacement map will be used.
26489 Note that once generated, displacements maps can be reused over and
26491 over again.
26492 A description of the accepted options follows.
26494 edge
26496 Set displace behavior for pixels that are out of range.
26497 Available values are:
26499 blank
26501 Missing pixels are replaced by black pixels.
26502 smear
26504 Adjacent pixels will spread out to replace missing pixels.
26505 wrap
26507 Out of range pixels are wrapped so they point to pixels of
26508 other side.
26509 mirror
26511 Out of range pixels will be replaced with mirrored pixels.
26512 Default is smear.
26514 Examples
26516 • Add ripple effect to rgb input of video size hd720:
26518 ffmpeg -i INPUT -f lavfi -i nullsrc=s=hd720,lutrgb=128:128:128 -f lavfi -i nullsrc=s=hd720,geq='r=128+30*sin(2*PI*X/400+T):g=128+30*sin(2*PI*X/400+T):b=128+30*sin(2*PI*X/400+T)' -lavfi '[0][1][2]displace' OUTPUT
26520 • Add wave effect to rgb input of video size hd720:
26522 ffmpeg -i INPUT -f lavfi -i nullsrc=hd720,geq='r=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T)):g=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T)):b=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T))' -lavfi '[1]split[x][y],[0][x][y]displace' OUTPUT
26524 dnn_classify
26526 Do classification with deep neural networks based on bounding boxes.
26527 The filter accepts the following options:
26529 dnn_backend
26531 Specify which DNN backend to use for model loading and execution.
26532 This option accepts only openvino now, tensorflow backends will be
26533 added.
26534 model
26536 Set path to model file specifying network architecture and its
26537 parameters. Note that different backends use different file
26538 formats.
26539 input
26541 Set the input name of the dnn network.
26542 output
26544 Set the output name of the dnn network.
26545 confidence
26547 Set the confidence threshold (default: 0.5).
26548 labels
26550 Set path to label file specifying the mapping between label id and
26551 name. Each label name is written in one line, tailing spaces and
26552 empty lines are skipped. The first line is the name of label id 0,
26553 and the second line is the name of label id 1, etc. The label id
26554 is considered as name if the label file is not provided.
26555 backend_configs
26557 Set the configs to be passed into backend
26558 For tensorflow backend, you can set its configs with sess_config
26560 options, please use tools/python/tf_sess_config.py to get the
26561 configs for your system.
26562 dnn_detect
26564 Do object detection with deep neural networks.
26565 The filter accepts the following options:
26567 dnn_backend
26569 Specify which DNN backend to use for model loading and execution.
26570 This option accepts only openvino now, tensorflow backends will be
26571 added.
26572 model
26574 Set path to model file specifying network architecture and its
26575 parameters. Note that different backends use different file
26576 formats.
26577 input
26579 Set the input name of the dnn network.
26580 output
26582 Set the output name of the dnn network.
26583 confidence
26585 Set the confidence threshold (default: 0.5).
26586 labels
26588 Set path to label file specifying the mapping between label id and
26589 name. Each label name is written in one line, tailing spaces and
26590 empty lines are skipped. The first line is the name of label id 0
26591 (usually it is 'background'), and the second line is the name of
26592 label id 1, etc. The label id is considered as name if the label
26593 file is not provided.
26594 backend_configs
26596 Set the configs to be passed into backend. To use async execution,
26597 set async (default: set). Roll back to sync execution if the
26598 backend does not support async.
26599 dnn_processing
26601 Do image processing with deep neural networks. It works together with
26602 another filter which converts the pixel format of the Frame to what the
26603 dnn network requires.
26604 The filter accepts the following options:
26606 dnn_backend
26608 Specify which DNN backend to use for model loading and execution.
26609 This option accepts the following values:
26610 native
26612 Native implementation of DNN loading and execution.
26613 tensorflow
26615 TensorFlow backend. To enable this backend you need to install
26616 the TensorFlow for C library (see
26617 <https://www.tensorflow.org/install/lang_c>) and configure
26618 FFmpeg with "--enable-libtensorflow"
26619 openvino
26621 OpenVINO backend. To enable this backend you need to build and
26622 install the OpenVINO for C library (see
26623 <https://github.com/openvinotoolkit/openvino/blob/master/build-instruction.md>)
26624 and configure FFmpeg with "--enable-libopenvino"
26625 (--extra-cflags=-I... --extra-ldflags=-L... might be needed if
26626 the header files and libraries are not installed into system
26627 path)
26628 Default value is native.
26630 model
26632 Set path to model file specifying network architecture and its
26633 parameters. Note that different backends use different file
26634 formats. TensorFlow, OpenVINO and native backend can load files for
26635 only its format.
26636 Native model file (.model) can be generated from TensorFlow model
26638 file (.pb) by using tools/python/convert.py
26639 input
26641 Set the input name of the dnn network.
26642 output
26644 Set the output name of the dnn network.
26645 backend_configs
26647 Set the configs to be passed into backend. To use async execution,
26648 set async (default: set). Roll back to sync execution if the
26649 backend does not support async.
26650 For tensorflow backend, you can set its configs with sess_config
26652 options, please use tools/python/tf_sess_config.py to get the
26653 configs of TensorFlow backend for your system.
26654 Examples
26656 • Remove rain in rgb24 frame with can.pb (see derain filter):
26658 ./ffmpeg -i rain.jpg -vf format=rgb24,dnn_processing=dnn_backend=tensorflow:model=can.pb:input=x:output=y derain.jpg
26660 • Halve the pixel value of the frame with format gray32f:
26662 ffmpeg -i input.jpg -vf format=grayf32,dnn_processing=model=halve_gray_float.model:input=dnn_in:output=dnn_out:dnn_backend=native -y out.native.png
26664 • Handle the Y channel with srcnn.pb (see sr filter) for frame with
26666 yuv420p (planar YUV formats supported):
26667 ./ffmpeg -i 480p.jpg -vf format=yuv420p,scale=w=iw*2:h=ih*2,dnn_processing=dnn_backend=tensorflow:model=srcnn.pb:input=x:output=y -y srcnn.jpg
26669 • Handle the Y channel with espcn.pb (see sr filter), which changes
26671 frame size, for format yuv420p (planar YUV formats supported),
26672 please use tools/python/tf_sess_config.py to get the configs of
26673 TensorFlow backend for your system.
26674 ./ffmpeg -i 480p.jpg -vf format=yuv420p,dnn_processing=dnn_backend=tensorflow:model=espcn.pb:input=x:output=y:backend_configs=sess_config=0x10022805320e09cdccccccccccec3f20012a01303801 -y tmp.espcn.jpg
26676 drawbox
26678 Draw a colored box on the input image.
26679 It accepts the following parameters:
26681 x
26683 y The expressions which specify the top left corner coordinates of
26684 the box. It defaults to 0.
26685 width, w
26687 height, h
26688 The expressions which specify the width and height of the box; if 0
26689 they are interpreted as the input width and height. It defaults to
26690 0.
26691 color, c
26693 Specify the color of the box to write. For the general syntax of
26694 this option, check the "Color" section in the ffmpeg-utils manual.
26695 If the special value "invert" is used, the box edge color is the
26696 same as the video with inverted luma.
26697 thickness, t
26699 The expression which sets the thickness of the box edge. A value
26700 of "fill" will create a filled box. Default value is 3.
26701 See below for the list of accepted constants.
26703 replace
26705 Applicable if the input has alpha. With value 1, the pixels of the
26706 painted box will overwrite the video's color and alpha pixels.
26707 Default is 0, which composites the box onto the input, leaving the
26708 video's alpha intact.
26709 The parameters for x, y, w and h and t are expressions containing the
26711 following constants:
26712 dar The input display aspect ratio, it is the same as (w / h) * sar.
26714 hsub
26716 vsub
26717 horizontal and vertical chroma subsample values. For example for
26718 the pixel format "yuv422p" hsub is 2 and vsub is 1.
26719 in_h, ih
26721 in_w, iw
26722 The input width and height.
26723 sar The input sample aspect ratio.
26725 x
26727 y The x and y offset coordinates where the box is drawn.
26728 w
26730 h The width and height of the drawn box.
26731 box_source
26733 Box source can be set as side_data_detection_bboxes if you want to
26734 use box data in detection bboxes of side data.
26735 If box_source is set, the x, y, width and height will be ignored
26737 and still use box data in detection bboxes of side data. So please
26738 do not use this parameter if you were not sure about the box
26739 source.
26740 t The thickness of the drawn box.
26742 These constants allow the x, y, w, h and t expressions to refer to
26744 each other, so you may for example specify "y=x/dar" or "h=w/dar".
26745 Examples
26747 • Draw a black box around the edge of the input image:
26749 drawbox
26751 • Draw a box with color red and an opacity of 50%:
26753 drawbox=10:20:200:60:red@0.5
26755 The previous example can be specified as:
26757 drawbox=x=10:y=20:w=200:h=60:color=red@0.5
26759 • Fill the box with pink color:
26761 drawbox=x=10:y=10:w=100:h=100:color=pink@0.5:t=fill
26763 • Draw a 2-pixel red 2.40:1 mask:
26765 drawbox=x=-t:y=0.5*(ih-iw/2.4)-t:w=iw+t*2:h=iw/2.4+t*2:t=2:c=red
26767 Commands
26769 This filter supports same commands as options. The command accepts the
26771 same syntax of the corresponding option.
26772 If the specified expression is not valid, it is kept at its current
26774 value.
26775 drawgraph
26777 Draw a graph using input video metadata.
26778 It accepts the following parameters:
26780 m1 Set 1st frame metadata key from which metadata values will be used
26782 to draw a graph.
26783 fg1 Set 1st foreground color expression.
26785 m2 Set 2nd frame metadata key from which metadata values will be used
26787 to draw a graph.
26788 fg2 Set 2nd foreground color expression.
26790 m3 Set 3rd frame metadata key from which metadata values will be used
26792 to draw a graph.
26793 fg3 Set 3rd foreground color expression.
26795 m4 Set 4th frame metadata key from which metadata values will be used
26797 to draw a graph.
26798 fg4 Set 4th foreground color expression.
26800 min Set minimal value of metadata value.
26802 max Set maximal value of metadata value.
26804 bg Set graph background color. Default is white.
26806 mode
26808 Set graph mode.
26809 Available values for mode is:
26811 bar
26813 dot
26814 line
26815 Default is "line".
26817 slide
26819 Set slide mode.
26820 Available values for slide is:
26822 frame
26824 Draw new frame when right border is reached.
26825 replace
26827 Replace old columns with new ones.
26828 scroll
26830 Scroll from right to left.
26831 rscroll
26833 Scroll from left to right.
26834 picture
26836 Draw single picture.
26837 Default is "frame".
26839 size
26841 Set size of graph video. For the syntax of this option, check the
26842 "Video size" section in the ffmpeg-utils manual. The default value
26843 is "900x256".
26844 rate, r
26846 Set the output frame rate. Default value is 25.
26847 The foreground color expressions can use the following variables:
26849 MIN Minimal value of metadata value.
26851 MAX Maximal value of metadata value.
26853 VAL Current metadata key value.
26855 The color is defined as 0xAABBGGRR.
26857 Example using metadata from signalstats filter:
26859 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
26861 Example using metadata from ebur128 filter:
26863 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
26865 drawgrid
26867 Draw a grid on the input image.
26868 It accepts the following parameters:
26870 x
26872 y The expressions which specify the coordinates of some point of grid
26873 intersection (meant to configure offset). Both default to 0.
26874 width, w
26876 height, h
26877 The expressions which specify the width and height of the grid
26878 cell, if 0 they are interpreted as the input width and height,
26879 respectively, minus "thickness", so image gets framed. Default to
26880 0.
26881 color, c
26883 Specify the color of the grid. For the general syntax of this
26884 option, check the "Color" section in the ffmpeg-utils manual. If
26885 the special value "invert" is used, the grid color is the same as
26886 the video with inverted luma.
26887 thickness, t
26889 The expression which sets the thickness of the grid line. Default
26890 value is 1.
26891 See below for the list of accepted constants.
26893 replace
26895 Applicable if the input has alpha. With 1 the pixels of the painted
26896 grid will overwrite the video's color and alpha pixels. Default is
26897 0, which composites the grid onto the input, leaving the video's
26898 alpha intact.
26899 The parameters for x, y, w and h and t are expressions containing the
26901 following constants:
26902 dar The input display aspect ratio, it is the same as (w / h) * sar.
26904 hsub
26906 vsub
26907 horizontal and vertical chroma subsample values. For example for
26908 the pixel format "yuv422p" hsub is 2 and vsub is 1.
26909 in_h, ih
26911 in_w, iw
26912 The input grid cell width and height.
26913 sar The input sample aspect ratio.
26915 x
26917 y The x and y coordinates of some point of grid intersection (meant
26918 to configure offset).
26919 w
26921 h The width and height of the drawn cell.
26922 t The thickness of the drawn cell.
26924 These constants allow the x, y, w, h and t expressions to refer to
26926 each other, so you may for example specify "y=x/dar" or "h=w/dar".
26927 Examples
26929 • Draw a grid with cell 100x100 pixels, thickness 2 pixels, with
26931 color red and an opacity of 50%:
26932 drawgrid=width=100:height=100:thickness=2:color=red@0.5
26934 • Draw a white 3x3 grid with an opacity of 50%:
26936 drawgrid=w=iw/3:h=ih/3:t=2:c=white@0.5
26938 Commands
26940 This filter supports same commands as options. The command accepts the
26942 same syntax of the corresponding option.
26943 If the specified expression is not valid, it is kept at its current
26945 value.
26946 drawtext
26948 Draw a text string or text from a specified file on top of a video,
26949 using the libfreetype library.
26950 To enable compilation of this filter, you need to configure FFmpeg with
26952 "--enable-libfreetype". To enable default font fallback and the font
26953 option you need to configure FFmpeg with "--enable-libfontconfig". To
26954 enable the text_shaping option, you need to configure FFmpeg with
26955 "--enable-libfribidi".
26956 Syntax
26958 It accepts the following parameters:
26960 box Used to draw a box around text using the background color. The
26962 value must be either 1 (enable) or 0 (disable). The default value
26963 of box is 0.
26964 boxborderw
26966 Set the width of the border to be drawn around the box using
26967 boxcolor. The default value of boxborderw is 0.
26968 boxcolor
26970 The color to be used for drawing box around text. For the syntax of
26971 this option, check the "Color" section in the ffmpeg-utils manual.
26972 The default value of boxcolor is "white".
26974 line_spacing
26976 Set the line spacing in pixels of the border to be drawn around the
26977 box using box. The default value of line_spacing is 0.
26978 borderw
26980 Set the width of the border to be drawn around the text using
26981 bordercolor. The default value of borderw is 0.
26982 bordercolor
26984 Set the color to be used for drawing border around text. For the
26985 syntax of this option, check the "Color" section in the ffmpeg-
26986 utils manual.
26987 The default value of bordercolor is "black".
26989 expansion
26991 Select how the text is expanded. Can be either "none", "strftime"
26992 (deprecated) or "normal" (default). See the drawtext_expansion,
26993 Text expansion section below for details.
26994 basetime
26996 Set a start time for the count. Value is in microseconds. Only
26997 applied in the deprecated strftime expansion mode. To emulate in
26998 normal expansion mode use the "pts" function, supplying the start
26999 time (in seconds) as the second argument.
27000 fix_bounds
27002 If true, check and fix text coords to avoid clipping.
27003 fontcolor
27005 The color to be used for drawing fonts. For the syntax of this
27006 option, check the "Color" section in the ffmpeg-utils manual.
27007 The default value of fontcolor is "black".
27009 fontcolor_expr
27011 String which is expanded the same way as text to obtain dynamic
27012 fontcolor value. By default this option has empty value and is not
27013 processed. When this option is set, it overrides fontcolor option.
27014 font
27016 The font family to be used for drawing text. By default Sans.
27017 fontfile
27019 The font file to be used for drawing text. The path must be
27020 included. This parameter is mandatory if the fontconfig support is
27021 disabled.
27022 alpha
27024 Draw the text applying alpha blending. The value can be a number
27025 between 0.0 and 1.0. The expression accepts the same variables x,
27026 y as well. The default value is 1. Please see fontcolor_expr.
27027 fontsize
27029 The font size to be used for drawing text. The default value of
27030 fontsize is 16.
27031 text_shaping
27033 If set to 1, attempt to shape the text (for example, reverse the
27034 order of right-to-left text and join Arabic characters) before
27035 drawing it. Otherwise, just draw the text exactly as given. By
27036 default 1 (if supported).
27037 ft_load_flags
27039 The flags to be used for loading the fonts.
27040 The flags map the corresponding flags supported by libfreetype, and
27042 are a combination of the following values:
27043 default
27045 no_scale
27046 no_hinting
27047 render
27048 no_bitmap
27049 vertical_layout
27050 force_autohint
27051 crop_bitmap
27052 pedantic
27053 ignore_global_advance_width
27054 no_recurse
27055 ignore_transform
27056 monochrome
27057 linear_design
27058 no_autohint
27059 Default value is "default".
27061 For more information consult the documentation for the FT_LOAD_*
27063 libfreetype flags.
27064 shadowcolor
27066 The color to be used for drawing a shadow behind the drawn text.
27067 For the syntax of this option, check the "Color" section in the
27068 ffmpeg-utils manual.
27069 The default value of shadowcolor is "black".
27071 shadowx
27073 shadowy
27074 The x and y offsets for the text shadow position with respect to
27075 the position of the text. They can be either positive or negative
27076 values. The default value for both is "0".
27077 start_number
27079 The starting frame number for the n/frame_num variable. The default
27080 value is "0".
27081 tabsize
27083 The size in number of spaces to use for rendering the tab. Default
27084 value is 4.
27085 timecode
27087 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
27088 format. It can be used with or without text parameter.
27089 timecode_rate option must be specified.
27090 timecode_rate, rate, r
27092 Set the timecode frame rate (timecode only). Value will be rounded
27093 to nearest integer. Minimum value is "1". Drop-frame timecode is
27094 supported for frame rates 30 & 60.
27095 tc24hmax
27097 If set to 1, the output of the timecode option will wrap around at
27098 24 hours. Default is 0 (disabled).
27099 text
27101 The text string to be drawn. The text must be a sequence of UTF-8
27102 encoded characters. This parameter is mandatory if no file is
27103 specified with the parameter textfile.
27104 textfile
27106 A text file containing text to be drawn. The text must be a
27107 sequence of UTF-8 encoded characters.
27108 This parameter is mandatory if no text string is specified with the
27110 parameter text.
27111 If both text and textfile are specified, an error is thrown.
27113 text_source
27115 Text source should be set as side_data_detection_bboxes if you want
27116 to use text data in detection bboxes of side data.
27117 If text source is set, text and textfile will be ignored and still
27119 use text data in detection bboxes of side data. So please do not
27120 use this parameter if you are not sure about the text source.
27121 reload
27123 If set to 1, the textfile will be reloaded before each frame. Be
27124 sure to update it atomically, or it may be read partially, or even
27125 fail.
27126 x
27128 y The expressions which specify the offsets where text will be drawn
27129 within the video frame. They are relative to the top/left border of
27130 the output image.
27131 The default value of x and y is "0".
27133 See below for the list of accepted constants and functions.
27135 The parameters for x and y are expressions containing the following
27137 constants and functions:
27138 dar input display aspect ratio, it is the same as (w / h) * sar
27140 hsub
27142 vsub
27143 horizontal and vertical chroma subsample values. For example for
27144 the pixel format "yuv422p" hsub is 2 and vsub is 1.
27145 line_h, lh
27147 the height of each text line
27148 main_h, h, H
27150 the input height
27151 main_w, w, W
27153 the input width
27154 max_glyph_a, ascent
27156 the maximum distance from the baseline to the highest/upper grid
27157 coordinate used to place a glyph outline point, for all the
27158 rendered glyphs. It is a positive value, due to the grid's
27159 orientation with the Y axis upwards.
27160 max_glyph_d, descent
27162 the maximum distance from the baseline to the lowest grid
27163 coordinate used to place a glyph outline point, for all the
27164 rendered glyphs. This is a negative value, due to the grid's
27165 orientation, with the Y axis upwards.
27166 max_glyph_h
27168 maximum glyph height, that is the maximum height for all the glyphs
27169 contained in the rendered text, it is equivalent to ascent -
27170 descent.
27171 max_glyph_w
27173 maximum glyph width, that is the maximum width for all the glyphs
27174 contained in the rendered text
27175 n the number of input frame, starting from 0
27177 rand(min, max)
27179 return a random number included between min and max
27180 sar The input sample aspect ratio.
27182 t timestamp expressed in seconds, NAN if the input timestamp is
27184 unknown
27185 text_h, th
27187 the height of the rendered text
27188 text_w, tw
27190 the width of the rendered text
27191 x
27193 y the x and y offset coordinates where the text is drawn.
27194 These parameters allow the x and y expressions to refer to each
27196 other, so you can for example specify "y=x/dar".
27197 pict_type
27199 A one character description of the current frame's picture type.
27200 pkt_pos
27202 The current packet's position in the input file or stream (in
27203 bytes, from the start of the input). A value of -1 indicates this
27204 info is not available.
27205 pkt_duration
27207 The current packet's duration, in seconds.
27208 pkt_size
27210 The current packet's size (in bytes).
27211 Text expansion
27213 If expansion is set to "strftime", the filter recognizes strftime()
27215 sequences in the provided text and expands them accordingly. Check the
27216 documentation of strftime(). This feature is deprecated.
27217 If expansion is set to "none", the text is printed verbatim.
27219 If expansion is set to "normal" (which is the default), the following
27221 expansion mechanism is used.
27222 The backslash character \, followed by any character, always expands to
27224 the second character.
27225 Sequences of the form "%{...}" are expanded. The text between the
27227 braces is a function name, possibly followed by arguments separated by
27228 ':'. If the arguments contain special characters or delimiters (':' or
27229 '}'), they should be escaped.
27230 Note that they probably must also be escaped as the value for the text
27232 option in the filter argument string and as the filter argument in the
27233 filtergraph description, and possibly also for the shell, that makes up
27234 to four levels of escaping; using a text file avoids these problems.
27235 The following functions are available:
27237 expr, e
27239 The expression evaluation result.
27240 It must take one argument specifying the expression to be
27242 evaluated, which accepts the same constants and functions as the x
27243 and y values. Note that not all constants should be used, for
27244 example the text size is not known when evaluating the expression,
27245 so the constants text_w and text_h will have an undefined value.
27246 expr_int_format, eif
27248 Evaluate the expression's value and output as formatted integer.
27249 The first argument is the expression to be evaluated, just as for
27251 the expr function. The second argument specifies the output
27252 format. Allowed values are x, X, d and u. They are treated exactly
27253 as in the "printf" function. The third parameter is optional and
27254 sets the number of positions taken by the output. It can be used
27255 to add padding with zeros from the left.
27256 gmtime
27258 The time at which the filter is running, expressed in UTC. It can
27259 accept an argument: a strftime() format string.
27260 localtime
27262 The time at which the filter is running, expressed in the local
27263 time zone. It can accept an argument: a strftime() format string.
27264 metadata
27266 Frame metadata. Takes one or two arguments.
27267 The first argument is mandatory and specifies the metadata key.
27269 The second argument is optional and specifies a default value, used
27271 when the metadata key is not found or empty.
27272 Available metadata can be identified by inspecting entries starting
27274 with TAG included within each frame section printed by running
27275 "ffprobe -show_frames".
27276 String metadata generated in filters leading to the drawtext filter
27278 are also available.
27279 n, frame_num
27281 The frame number, starting from 0.
27282 pict_type
27284 A one character description of the current picture type.
27285 pts The timestamp of the current frame. It can take up to three
27287 arguments.
27288 The first argument is the format of the timestamp; it defaults to
27290 "flt" for seconds as a decimal number with microsecond accuracy;
27291 "hms" stands for a formatted [-]HH:MM:SS.mmm timestamp with
27292 millisecond accuracy. "gmtime" stands for the timestamp of the
27293 frame formatted as UTC time; "localtime" stands for the timestamp
27294 of the frame formatted as local time zone time.
27295 The second argument is an offset added to the timestamp.
27297 If the format is set to "hms", a third argument "24HH" may be
27299 supplied to present the hour part of the formatted timestamp in 24h
27300 format (00-23).
27301 If the format is set to "localtime" or "gmtime", a third argument
27303 may be supplied: a strftime() format string. By default, YYYY-MM-
27304 DD HH:MM:SS format will be used.
27305 Commands
27307 This filter supports altering parameters via commands:
27309 reinit
27311 Alter existing filter parameters.
27312 Syntax for the argument is the same as for filter invocation, e.g.
27314 fontsize=56:fontcolor=green:text='Hello World'
27316 Full filter invocation with sendcmd would look like this:
27318 sendcmd=c='56.0 drawtext reinit fontsize=56\:fontcolor=green\:text=Hello\\ World'
27320 If the entire argument can't be parsed or applied as valid values then
27322 the filter will continue with its existing parameters.
27323 Examples
27325 • Draw "Test Text" with font FreeSerif, using the default values for
27327 the optional parameters.
27328 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
27330 • Draw 'Test Text' with font FreeSerif of size 24 at position x=100
27332 and y=50 (counting from the top-left corner of the screen), text is
27333 yellow with a red box around it. Both the text and the box have an
27334 opacity of 20%.
27335 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
27337 x=100: y=50: fontsize=24: fontcolor=yellow@0.2: box=1: boxcolor=red@0.2"
27338 Note that the double quotes are not necessary if spaces are not
27340 used within the parameter list.
27341 • Show the text at the center of the video frame:
27343 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
27345 • Show the text at a random position, switching to a new position
27347 every 30 seconds:
27348 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=if(eq(mod(t\,30)\,0)\,rand(0\,(w-text_w))\,x):y=if(eq(mod(t\,30)\,0)\,rand(0\,(h-text_h))\,y)"
27350 • Show a text line sliding from right to left in the last row of the
27352 video frame. The file LONG_LINE is assumed to contain a single line
27353 with no newlines.
27354 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
27356 • Show the content of file CREDITS off the bottom of the frame and
27358 scroll up.
27359 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
27361 • Draw a single green letter "g", at the center of the input video.
27363 The glyph baseline is placed at half screen height.
27364 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
27366 • Show text for 1 second every 3 seconds:
27368 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
27370 • Use fontconfig to set the font. Note that the colons need to be
27372 escaped.
27373 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
27375 • Draw "Test Text" with font size dependent on height of the video.
27377 drawtext="text='Test Text': fontsize=h/30: x=(w-text_w)/2: y=(h-text_h*2)"
27379 • Print the date of a real-time encoding (see strftime(3)):
27381 drawtext='fontfile=FreeSans.ttf:text=%{localtime\:%a %b %d %Y}'
27383 • Show text fading in and out (appearing/disappearing):
27385 #!/bin/sh
27387 DS=1.0 # display start
27388 DE=10.0 # display end
27389 FID=1.5 # fade in duration
27390 FOD=5 # fade out duration
27391 ffplay -f lavfi "color,drawtext=text=TEST:fontsize=50:fontfile=FreeSerif.ttf:fontcolor_expr=ff0000%{eif\\\\: clip(255*(1*between(t\\, $DS + $FID\\, $DE - $FOD) + ((t - $DS)/$FID)*between(t\\, $DS\\, $DS + $FID) + (-(t - $DE)/$FOD)*between(t\\, $DE - $FOD\\, $DE) )\\, 0\\, 255) \\\\: x\\\\: 2 }"
27392 • Horizontally align multiple separate texts. Note that max_glyph_a
27394 and the fontsize value are included in the y offset.
27395 drawtext=fontfile=FreeSans.ttf:text=DOG:fontsize=24:x=10:y=20+24-max_glyph_a,
27397 drawtext=fontfile=FreeSans.ttf:text=cow:fontsize=24:x=80:y=20+24-max_glyph_a
27398 • Plot special lavf.image2dec.source_basename metadata onto each
27400 frame if such metadata exists. Otherwise, plot the string "NA".
27401 Note that image2 demuxer must have option -export_path_metadata 1
27402 for the special metadata fields to be available for filters.
27403 drawtext="fontsize=20:fontcolor=white:fontfile=FreeSans.ttf:text='%{metadata\:lavf.image2dec.source_basename\:NA}':x=10:y=10"
27405 For more information about libfreetype, check:
27407 <http://www.freetype.org/>.
27408 For more information about fontconfig, check:
27410 <http://freedesktop.org/software/fontconfig/fontconfig-user.html>.
27411 For more information about libfribidi, check: <http://fribidi.org/>.
27413 edgedetect
27415 Detect and draw edges. The filter uses the Canny Edge Detection
27416 algorithm.
27417 The filter accepts the following options:
27419 low
27421 high
27422 Set low and high threshold values used by the Canny thresholding
27423 algorithm.
27424 The high threshold selects the "strong" edge pixels, which are then
27426 connected through 8-connectivity with the "weak" edge pixels
27427 selected by the low threshold.
27428 low and high threshold values must be chosen in the range [0,1],
27430 and low should be lesser or equal to high.
27431 Default value for low is "20/255", and default value for high is
27433 "50/255".
27434 mode
27436 Define the drawing mode.
27437 wires
27439 Draw white/gray wires on black background.
27440 colormix
27442 Mix the colors to create a paint/cartoon effect.
27443 canny
27445 Apply Canny edge detector on all selected planes.
27446 Default value is wires.
27448 planes
27450 Select planes for filtering. By default all available planes are
27451 filtered.
27452 Examples
27454 • Standard edge detection with custom values for the hysteresis
27456 thresholding:
27457 edgedetect=low=0.1:high=0.4
27459 • Painting effect without thresholding:
27461 edgedetect=mode=colormix:high=0
27463 elbg
27465 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
27466 For each input image, the filter will compute the optimal mapping from
27468 the input to the output given the codebook length, that is the number
27469 of distinct output colors.
27470 This filter accepts the following options.
27472 codebook_length, l
27474 Set codebook length. The value must be a positive integer, and
27475 represents the number of distinct output colors. Default value is
27476 256.
27477 nb_steps, n
27479 Set the maximum number of iterations to apply for computing the
27480 optimal mapping. The higher the value the better the result and the
27481 higher the computation time. Default value is 1.
27482 seed, s
27484 Set a random seed, must be an integer included between 0 and
27485 UINT32_MAX. If not specified, or if explicitly set to -1, the
27486 filter will try to use a good random seed on a best effort basis.
27487 pal8
27489 Set pal8 output pixel format. This option does not work with
27490 codebook length greater than 256. Default is disabled.
27491 use_alpha
27493 Include alpha values in the quantization calculation. Allows
27494 creating palettized output images (e.g. PNG8) with multiple alpha
27495 smooth blending.
27496 entropy
27498 Measure graylevel entropy in histogram of color channels of video
27499 frames.
27500 It accepts the following parameters:
27502 mode
27504 Can be either normal or diff. Default is normal.
27505 diff mode measures entropy of histogram delta values, absolute
27507 differences between neighbour histogram values.
27508 epx
27510 Apply the EPX magnification filter which is designed for pixel art.
27511 It accepts the following option:
27513 n Set the scaling dimension: 2 for "2xEPX", 3 for "3xEPX". Default
27515 is 3.
27516 eq
27518 Set brightness, contrast, saturation and approximate gamma adjustment.
27519 The filter accepts the following options:
27521 contrast
27523 Set the contrast expression. The value must be a float value in
27524 range "-1000.0" to 1000.0. The default value is "1".
27525 brightness
27527 Set the brightness expression. The value must be a float value in
27528 range "-1.0" to 1.0. The default value is "0".
27529 saturation
27531 Set the saturation expression. The value must be a float in range
27532 0.0 to 3.0. The default value is "1".
27533 gamma
27535 Set the gamma expression. The value must be a float in range 0.1 to
27536 10.0. The default value is "1".
27537 gamma_r
27539 Set the gamma expression for red. The value must be a float in
27540 range 0.1 to 10.0. The default value is "1".
27541 gamma_g
27543 Set the gamma expression for green. The value must be a float in
27544 range 0.1 to 10.0. The default value is "1".
27545 gamma_b
27547 Set the gamma expression for blue. The value must be a float in
27548 range 0.1 to 10.0. The default value is "1".
27549 gamma_weight
27551 Set the gamma weight expression. It can be used to reduce the
27552 effect of a high gamma value on bright image areas, e.g. keep them
27553 from getting overamplified and just plain white. The value must be
27554 a float in range 0.0 to 1.0. A value of 0.0 turns the gamma
27555 correction all the way down while 1.0 leaves it at its full
27556 strength. Default is "1".
27557 eval
27559 Set when the expressions for brightness, contrast, saturation and
27560 gamma expressions are evaluated.
27561 It accepts the following values:
27563 init
27565 only evaluate expressions once during the filter initialization
27566 or when a command is processed
27567 frame
27569 evaluate expressions for each incoming frame
27570 Default value is init.
27572 The expressions accept the following parameters:
27574 n frame count of the input frame starting from 0
27576 pos byte position of the corresponding packet in the input file, NAN if
27578 unspecified
27579 r frame rate of the input video, NAN if the input frame rate is
27581 unknown
27582 t timestamp expressed in seconds, NAN if the input timestamp is
27584 unknown
27585 Commands
27587 The filter supports the following commands:
27589 contrast
27591 Set the contrast expression.
27592 brightness
27594 Set the brightness expression.
27595 saturation
27597 Set the saturation expression.
27598 gamma
27600 Set the gamma expression.
27601 gamma_r
27603 Set the gamma_r expression.
27604 gamma_g
27606 Set gamma_g expression.
27607 gamma_b
27609 Set gamma_b expression.
27610 gamma_weight
27612 Set gamma_weight expression.
27613 The command accepts the same syntax of the corresponding option.
27615 If the specified expression is not valid, it is kept at its current
27617 value.
27618 erosion
27620 Apply erosion effect to the video.
27621 This filter replaces the pixel by the local(3x3) minimum.
27623 It accepts the following options:
27625 threshold0
27627 threshold1
27628 threshold2
27629 threshold3
27630 Limit the maximum change for each plane, default is 65535. If 0,
27631 plane will remain unchanged.
27632 coordinates
27634 Flag which specifies the pixel to refer to. Default is 255 i.e. all
27635 eight pixels are used.
27636 Flags to local 3x3 coordinates maps like this:
27638 1 2 3
27640 4 5
27641 6 7 8
27642 Commands
27644 This filter supports the all above options as commands.
27646 estdif
27648 Deinterlace the input video ("estdif" stands for "Edge Slope Tracing
27649 Deinterlacing Filter").
27650 Spatial only filter that uses edge slope tracing algorithm to
27652 interpolate missing lines. It accepts the following parameters:
27653 mode
27655 The interlacing mode to adopt. It accepts one of the following
27656 values:
27657 frame
27659 Output one frame for each frame.
27660 field
27662 Output one frame for each field.
27663 The default value is "field".
27665 parity
27667 The picture field parity assumed for the input interlaced video. It
27668 accepts one of the following values:
27669 tff Assume the top field is first.
27671 bff Assume the bottom field is first.
27673 auto
27675 Enable automatic detection of field parity.
27676 The default value is "auto". If the interlacing is unknown or the
27678 decoder does not export this information, top field first will be
27679 assumed.
27680 deint
27682 Specify which frames to deinterlace. Accepts one of the following
27683 values:
27684 all Deinterlace all frames.
27686 interlaced
27688 Only deinterlace frames marked as interlaced.
27689 The default value is "all".
27691 rslope
27693 Specify the search radius for edge slope tracing. Default value is
27694 1. Allowed range is from 1 to 15.
27695 redge
27697 Specify the search radius for best edge matching. Default value is
27698 2. Allowed range is from 0 to 15.
27699 ecost
27701 Specify the edge cost for edge matching. Default value is 0.03125.
27702 Allowed range is from 0 to 1.
27703 mcost
27705 Specify the middle cost for edge matching. Default value is 0.5.
27706 Allowed range is from 0 to 1.
27707 dcost
27709 Specify the distance cost for edge matching. Default value is 0.5.
27710 Allowed range is from 0 to 1.
27711 interp
27713 Specify the interpolation used. Default is 4-point interpolation.
27714 It accepts one of the following values:
27715 2p Two-point interpolation.
27717 4p Four-point interpolation.
27719 6p Six-point interpolation.
27721 Commands
27723 This filter supports same commands as options.
27725 exposure
27727 Adjust exposure of the video stream.
27728 The filter accepts the following options:
27730 exposure
27732 Set the exposure correction in EV. Allowed range is from -3.0 to
27733 3.0 EV Default value is 0 EV.
27734 black
27736 Set the black level correction. Allowed range is from -1.0 to 1.0.
27737 Default value is 0.
27738 Commands
27740 This filter supports same commands as options.
27742 extractplanes
27744 Extract color channel components from input video stream into separate
27745 grayscale video streams.
27746 The filter accepts the following option:
27748 planes
27750 Set plane(s) to extract.
27751 Available values for planes are:
27753 y
27755 u
27756 v
27757 a
27758 r
27759 g
27760 b
27761 Choosing planes not available in the input will result in an error.
27763 That means you cannot select "r", "g", "b" planes with "y", "u",
27764 "v" planes at same time.
27765 Examples
27767 • Extract luma, u and v color channel component from input video
27769 frame into 3 grayscale outputs:
27770 ffmpeg -i video.avi -filter_complex 'extractplanes=y+u+v[y][u][v]' -map '[y]' y.avi -map '[u]' u.avi -map '[v]' v.avi
27772 fade
27774 Apply a fade-in/out effect to the input video.
27775 It accepts the following parameters:
27777 type, t
27779 The effect type can be either "in" for a fade-in, or "out" for a
27780 fade-out effect. Default is "in".
27781 start_frame, s
27783 Specify the number of the frame to start applying the fade effect
27784 at. Default is 0.
27785 nb_frames, n
27787 The number of frames that the fade effect lasts. At the end of the
27788 fade-in effect, the output video will have the same intensity as
27789 the input video. At the end of the fade-out transition, the output
27790 video will be filled with the selected color. Default is 25.
27791 alpha
27793 If set to 1, fade only alpha channel, if one exists on the input.
27794 Default value is 0.
27795 start_time, st
27797 Specify the timestamp (in seconds) of the frame to start to apply
27798 the fade effect. If both start_frame and start_time are specified,
27799 the fade will start at whichever comes last. Default is 0.
27800 duration, d
27802 The number of seconds for which the fade effect has to last. At the
27803 end of the fade-in effect the output video will have the same
27804 intensity as the input video, at the end of the fade-out transition
27805 the output video will be filled with the selected color. If both
27806 duration and nb_frames are specified, duration is used. Default is
27807 0 (nb_frames is used by default).
27808 color, c
27810 Specify the color of the fade. Default is "black".
27811 Examples
27813 • Fade in the first 30 frames of video:
27815 fade=in:0:30
27817 The command above is equivalent to:
27819 fade=t=in:s=0:n=30
27821 • Fade out the last 45 frames of a 200-frame video:
27823 fade=out:155:45
27825 fade=type=out:start_frame=155:nb_frames=45
27826 • Fade in the first 25 frames and fade out the last 25 frames of a
27828 1000-frame video:
27829 fade=in:0:25, fade=out:975:25
27831 • Make the first 5 frames yellow, then fade in from frame 5-24:
27833 fade=in:5:20:color=yellow
27835 • Fade in alpha over first 25 frames of video:
27837 fade=in:0:25:alpha=1
27839 • Make the first 5.5 seconds black, then fade in for 0.5 seconds:
27841 fade=t=in:st=5.5:d=0.5
27843 fftdnoiz
27845 Denoise frames using 3D FFT (frequency domain filtering).
27846 The filter accepts the following options:
27848 sigma
27850 Set the noise sigma constant. This sets denoising strength.
27851 Default value is 1. Allowed range is from 0 to 30. Using very high
27852 sigma with low overlap may give blocking artifacts.
27853 amount
27855 Set amount of denoising. By default all detected noise is reduced.
27856 Default value is 1. Allowed range is from 0 to 1.
27857 block
27859 Set size of block, Default is 4, can be 3, 4, 5 or 6. Actual size
27860 of block in pixels is 2 to power of block, so by default block size
27861 in pixels is 2^4 which is 16.
27862 overlap
27864 Set block overlap. Default is 0.5. Allowed range is from 0.2 to
27865 0.8.
27866 prev
27868 Set number of previous frames to use for denoising. By default is
27869 set to 0.
27870 next
27872 Set number of next frames to to use for denoising. By default is
27873 set to 0.
27874 planes
27876 Set planes which will be filtered, by default are all available
27877 filtered except alpha.
27878 fftfilt
27880 Apply arbitrary expressions to samples in frequency domain
27881 dc_Y
27883 Adjust the dc value (gain) of the luma plane of the image. The
27884 filter accepts an integer value in range 0 to 1000. The default
27885 value is set to 0.
27886 dc_U
27888 Adjust the dc value (gain) of the 1st chroma plane of the image.
27889 The filter accepts an integer value in range 0 to 1000. The default
27890 value is set to 0.
27891 dc_V
27893 Adjust the dc value (gain) of the 2nd chroma plane of the image.
27894 The filter accepts an integer value in range 0 to 1000. The default
27895 value is set to 0.
27896 weight_Y
27898 Set the frequency domain weight expression for the luma plane.
27899 weight_U
27901 Set the frequency domain weight expression for the 1st chroma
27902 plane.
27903 weight_V
27905 Set the frequency domain weight expression for the 2nd chroma
27906 plane.
27907 eval
27909 Set when the expressions are evaluated.
27910 It accepts the following values:
27912 init
27914 Only evaluate expressions once during the filter
27915 initialization.
27916 frame
27918 Evaluate expressions for each incoming frame.
27919 Default value is init.
27921 The filter accepts the following variables:
27923 X
27925 Y The coordinates of the current sample.
27926 W
27928 H The width and height of the image.
27929 N The number of input frame, starting from 0.
27931 WS
27933 HS The size of FFT array for horizontal and vertical processing.
27934 Examples
27936 • High-pass:
27938 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
27940 • Low-pass:
27942 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
27944 • Sharpen:
27946 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
27948 • Blur:
27950 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
27952 field
27954 Extract a single field from an interlaced image using stride arithmetic
27955 to avoid wasting CPU time. The output frames are marked as non-
27956 interlaced.
27957 The filter accepts the following options:
27959 type
27961 Specify whether to extract the top (if the value is 0 or "top") or
27962 the bottom field (if the value is 1 or "bottom").
27963 fieldhint
27965 Create new frames by copying the top and bottom fields from surrounding
27966 frames supplied as numbers by the hint file.
27967 hint
27969 Set file containing hints: absolute/relative frame numbers.
27970 There must be one line for each frame in a clip. Each line must
27972 contain two numbers separated by the comma, optionally followed by
27973 "-" or "+". Numbers supplied on each line of file can not be out
27974 of [N-1,N+1] where N is current frame number for "absolute" mode or
27975 out of [-1, 1] range for "relative" mode. First number tells from
27976 which frame to pick up top field and second number tells from which
27977 frame to pick up bottom field.
27978 If optionally followed by "+" output frame will be marked as
27980 interlaced, else if followed by "-" output frame will be marked as
27981 progressive, else it will be marked same as input frame. If
27982 optionally followed by "t" output frame will use only top field, or
27983 in case of "b" it will use only bottom field. If line starts with
27984 "#" or ";" that line is skipped.
27985 mode
27987 Can be item "absolute" or "relative". Default is "absolute".
27988 Example of first several lines of "hint" file for "relative" mode:
27990 0,0 - # first frame
27992 1,0 - # second frame, use third's frame top field and second's frame bottom field
27993 1,0 - # third frame, use fourth's frame top field and third's frame bottom field
27994 1,0 -
27995 0,0 -
27996 0,0 -
27997 1,0 -
27998 1,0 -
27999 1,0 -
28000 0,0 -
28001 0,0 -
28002 1,0 -
28003 1,0 -
28004 1,0 -
28005 0,0 -
28006 fieldmatch
28008 Field matching filter for inverse telecine. It is meant to reconstruct
28009 the progressive frames from a telecined stream. The filter does not
28010 drop duplicated frames, so to achieve a complete inverse telecine
28011 "fieldmatch" needs to be followed by a decimation filter such as
28012 decimate in the filtergraph.
28013 The separation of the field matching and the decimation is notably
28015 motivated by the possibility of inserting a de-interlacing filter
28016 fallback between the two. If the source has mixed telecined and real
28017 interlaced content, "fieldmatch" will not be able to match fields for
28018 the interlaced parts. But these remaining combed frames will be marked
28019 as interlaced, and thus can be de-interlaced by a later filter such as
28020 yadif before decimation.
28021 In addition to the various configuration options, "fieldmatch" can take
28023 an optional second stream, activated through the ppsrc option. If
28024 enabled, the frames reconstruction will be based on the fields and
28025 frames from this second stream. This allows the first input to be pre-
28026 processed in order to help the various algorithms of the filter, while
28027 keeping the output lossless (assuming the fields are matched properly).
28028 Typically, a field-aware denoiser, or brightness/contrast adjustments
28029 can help.
28030 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth
28032 project) and VIVTC/VFM (VapourSynth project). The later is a light
28033 clone of TFM from which "fieldmatch" is based on. While the semantic
28034 and usage are very close, some behaviour and options names can differ.
28035 The decimate filter currently only works for constant frame rate input.
28037 If your input has mixed telecined (30fps) and progressive content with
28038 a lower framerate like 24fps use the following filterchain to produce
28039 the necessary cfr stream:
28040 "dejudder,fps=30000/1001,fieldmatch,decimate".
28041 The filter accepts the following options:
28043 order
28045 Specify the assumed field order of the input stream. Available
28046 values are:
28047 auto
28049 Auto detect parity (use FFmpeg's internal parity value).
28050 bff Assume bottom field first.
28052 tff Assume top field first.
28054 Note that it is sometimes recommended not to trust the parity
28056 announced by the stream.
28057 Default value is auto.
28059 mode
28061 Set the matching mode or strategy to use. pc mode is the safest in
28062 the sense that it won't risk creating jerkiness due to duplicate
28063 frames when possible, but if there are bad edits or blended fields
28064 it will end up outputting combed frames when a good match might
28065 actually exist. On the other hand, pcn_ub mode is the most risky in
28066 terms of creating jerkiness, but will almost always find a good
28067 frame if there is one. The other values are all somewhere in
28068 between pc and pcn_ub in terms of risking jerkiness and creating
28069 duplicate frames versus finding good matches in sections with bad
28070 edits, orphaned fields, blended fields, etc.
28071 More details about p/c/n/u/b are available in p/c/n/u/b meaning
28073 section.
28074 Available values are:
28076 pc 2-way matching (p/c)
28078 pc_n
28080 2-way matching, and trying 3rd match if still combed (p/c + n)
28081 pc_u
28083 2-way matching, and trying 3rd match (same order) if still
28084 combed (p/c + u)
28085 pc_n_ub
28087 2-way matching, trying 3rd match if still combed, and trying
28088 4th/5th matches if still combed (p/c + n + u/b)
28089 pcn 3-way matching (p/c/n)
28091 pcn_ub
28093 3-way matching, and trying 4th/5th matches if all 3 of the
28094 original matches are detected as combed (p/c/n + u/b)
28095 The parenthesis at the end indicate the matches that would be used
28097 for that mode assuming order=tff (and field on auto or top).
28098 In terms of speed pc mode is by far the fastest and pcn_ub is the
28100 slowest.
28101 Default value is pc_n.
28103 ppsrc
28105 Mark the main input stream as a pre-processed input, and enable the
28106 secondary input stream as the clean source to pick the fields from.
28107 See the filter introduction for more details. It is similar to the
28108 clip2 feature from VFM/TFM.
28109 Default value is 0 (disabled).
28111 field
28113 Set the field to match from. It is recommended to set this to the
28114 same value as order unless you experience matching failures with
28115 that setting. In certain circumstances changing the field that is
28116 used to match from can have a large impact on matching performance.
28117 Available values are:
28118 auto
28120 Automatic (same value as order).
28121 bottom
28123 Match from the bottom field.
28124 top Match from the top field.
28126 Default value is auto.
28128 mchroma
28130 Set whether or not chroma is included during the match comparisons.
28131 In most cases it is recommended to leave this enabled. You should
28132 set this to 0 only if your clip has bad chroma problems such as
28133 heavy rainbowing or other artifacts. Setting this to 0 could also
28134 be used to speed things up at the cost of some accuracy.
28135 Default value is 1.
28137 y0
28139 y1 These define an exclusion band which excludes the lines between y0
28140 and y1 from being included in the field matching decision. An
28141 exclusion band can be used to ignore subtitles, a logo, or other
28142 things that may interfere with the matching. y0 sets the starting
28143 scan line and y1 sets the ending line; all lines in between y0 and
28144 y1 (including y0 and y1) will be ignored. Setting y0 and y1 to the
28145 same value will disable the feature. y0 and y1 defaults to 0.
28146 scthresh
28148 Set the scene change detection threshold as a percentage of maximum
28149 change on the luma plane. Good values are in the "[8.0, 14.0]"
28150 range. Scene change detection is only relevant in case
28151 combmatch=sc. The range for scthresh is "[0.0, 100.0]".
28152 Default value is 12.0.
28154 combmatch
28156 When combatch is not none, "fieldmatch" will take into account the
28157 combed scores of matches when deciding what match to use as the
28158 final match. Available values are:
28159 none
28161 No final matching based on combed scores.
28162 sc Combed scores are only used when a scene change is detected.
28164 full
28166 Use combed scores all the time.
28167 Default is sc.
28169 combdbg
28171 Force "fieldmatch" to calculate the combed metrics for certain
28172 matches and print them. This setting is known as micout in TFM/VFM
28173 vocabulary. Available values are:
28174 none
28176 No forced calculation.
28177 pcn Force p/c/n calculations.
28179 pcnub
28181 Force p/c/n/u/b calculations.
28182 Default value is none.
28184 cthresh
28186 This is the area combing threshold used for combed frame detection.
28187 This essentially controls how "strong" or "visible" combing must be
28188 to be detected. Larger values mean combing must be more visible
28189 and smaller values mean combing can be less visible or strong and
28190 still be detected. Valid settings are from "-1" (every pixel will
28191 be detected as combed) to 255 (no pixel will be detected as
28192 combed). This is basically a pixel difference value. A good range
28193 is "[8, 12]".
28194 Default value is 9.
28196 chroma
28198 Sets whether or not chroma is considered in the combed frame
28199 decision. Only disable this if your source has chroma problems
28200 (rainbowing, etc.) that are causing problems for the combed frame
28201 detection with chroma enabled. Actually, using chroma=0 is usually
28202 more reliable, except for the case where there is chroma only
28203 combing in the source.
28204 Default value is 0.
28206 blockx
28208 blocky
28209 Respectively set the x-axis and y-axis size of the window used
28210 during combed frame detection. This has to do with the size of the
28211 area in which combpel pixels are required to be detected as combed
28212 for a frame to be declared combed. See the combpel parameter
28213 description for more info. Possible values are any number that is
28214 a power of 2 starting at 4 and going up to 512.
28215 Default value is 16.
28217 combpel
28219 The number of combed pixels inside any of the blocky by blockx size
28220 blocks on the frame for the frame to be detected as combed. While
28221 cthresh controls how "visible" the combing must be, this setting
28222 controls "how much" combing there must be in any localized area (a
28223 window defined by the blockx and blocky settings) on the frame.
28224 Minimum value is 0 and maximum is "blocky x blockx" (at which point
28225 no frames will ever be detected as combed). This setting is known
28226 as MI in TFM/VFM vocabulary.
28227 Default value is 80.
28229 p/c/n/u/b meaning
28231 p/c/n
28233 We assume the following telecined stream:
28235 Top fields: 1 2 2 3 4
28237 Bottom fields: 1 2 3 4 4
28238 The numbers correspond to the progressive frame the fields relate to.
28240 Here, the first two frames are progressive, the 3rd and 4th are combed,
28241 and so on.
28242 When "fieldmatch" is configured to run a matching from bottom
28244 (field=bottom) this is how this input stream get transformed:
28245 Input stream:
28247 T 1 2 2 3 4
28248 B 1 2 3 4 4 <-- matching reference
28249 Matches: c c n n c
28251 Output stream:
28253 T 1 2 3 4 4
28254 B 1 2 3 4 4
28255 As a result of the field matching, we can see that some frames get
28257 duplicated. To perform a complete inverse telecine, you need to rely
28258 on a decimation filter after this operation. See for instance the
28259 decimate filter.
28260 The same operation now matching from top fields (field=top) looks like
28262 this:
28263 Input stream:
28265 T 1 2 2 3 4 <-- matching reference
28266 B 1 2 3 4 4
28267 Matches: c c p p c
28269 Output stream:
28271 T 1 2 2 3 4
28272 B 1 2 2 3 4
28273 In these examples, we can see what p, c and n mean; basically, they
28275 refer to the frame and field of the opposite parity:
28276 *<p matches the field of the opposite parity in the previous frame>
28278 *<c matches the field of the opposite parity in the current frame>
28279 *<n matches the field of the opposite parity in the next frame>
28280 u/b
28282 The u and b matching are a bit special in the sense that they match
28284 from the opposite parity flag. In the following examples, we assume
28285 that we are currently matching the 2nd frame (Top:2, bottom:2).
28286 According to the match, a 'x' is placed above and below each matched
28287 fields.
28288 With bottom matching (field=bottom):
28290 Match: c p n b u
28292 x x x x x
28294 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
28295 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
28296 x x x x x
28297 Output frames:
28299 2 1 2 2 2
28300 2 2 2 1 3
28301 With top matching (field=top):
28303 Match: c p n b u
28305 x x x x x
28307 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
28308 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
28309 x x x x x
28310 Output frames:
28312 2 2 2 1 2
28313 2 1 3 2 2
28314 Examples
28316 Simple IVTC of a top field first telecined stream:
28318 fieldmatch=order=tff:combmatch=none, decimate
28320 Advanced IVTC, with fallback on yadif for still combed frames:
28322 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
28324 fieldorder
28326 Transform the field order of the input video.
28327 It accepts the following parameters:
28329 order
28331 The output field order. Valid values are tff for top field first or
28332 bff for bottom field first.
28333 The default value is tff.
28335 The transformation is done by shifting the picture content up or down
28337 by one line, and filling the remaining line with appropriate picture
28338 content. This method is consistent with most broadcast field order
28339 converters.
28340 If the input video is not flagged as being interlaced, or it is already
28342 flagged as being of the required output field order, then this filter
28343 does not alter the incoming video.
28344 It is very useful when converting to or from PAL DV material, which is
28346 bottom field first.
28347 For example:
28349 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
28351 fifo, afifo
28353 Buffer input images and send them when they are requested.
28354 It is mainly useful when auto-inserted by the libavfilter framework.
28356 It does not take parameters.
28358 fillborders
28360 Fill borders of the input video, without changing video stream
28361 dimensions. Sometimes video can have garbage at the four edges and you
28362 may not want to crop video input to keep size multiple of some number.
28363 This filter accepts the following options:
28365 left
28367 Number of pixels to fill from left border.
28368 right
28370 Number of pixels to fill from right border.
28371 top Number of pixels to fill from top border.
28373 bottom
28375 Number of pixels to fill from bottom border.
28376 mode
28378 Set fill mode.
28379 It accepts the following values:
28381 smear
28383 fill pixels using outermost pixels
28384 mirror
28386 fill pixels using mirroring (half sample symmetric)
28387 fixed
28389 fill pixels with constant value
28390 reflect
28392 fill pixels using reflecting (whole sample symmetric)
28393 wrap
28395 fill pixels using wrapping
28396 fade
28398 fade pixels to constant value
28399 margins
28401 fill pixels at top and bottom with weighted averages pixels
28402 near borders
28403 Default is smear.
28405 color
28407 Set color for pixels in fixed or fade mode. Default is black.
28408 Commands
28410 This filter supports same commands as options. The command accepts the
28412 same syntax of the corresponding option.
28413 If the specified expression is not valid, it is kept at its current
28415 value.
28416 find_rect
28418 Find a rectangular object
28419 It accepts the following options:
28421 object
28423 Filepath of the object image, needs to be in gray8.
28424 threshold
28426 Detection threshold, default is 0.5.
28427 mipmaps
28429 Number of mipmaps, default is 3.
28430 xmin, ymin, xmax, ymax
28432 Specifies the rectangle in which to search.
28433 discard
28435 Discard frames where object is not detected. Default is disabled.
28436 Examples
28438 • Cover a rectangular object by the supplied image of a given video
28440 using ffmpeg:
28441 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
28443 floodfill
28445 Flood area with values of same pixel components with another values.
28446 It accepts the following options:
28448 x Set pixel x coordinate.
28450 y Set pixel y coordinate.
28452 s0 Set source #0 component value.
28454 s1 Set source #1 component value.
28456 s2 Set source #2 component value.
28458 s3 Set source #3 component value.
28460 d0 Set destination #0 component value.
28462 d1 Set destination #1 component value.
28464 d2 Set destination #2 component value.
28466 d3 Set destination #3 component value.
28468 format
28470 Convert the input video to one of the specified pixel formats.
28471 Libavfilter will try to pick one that is suitable as input to the next
28472 filter.
28473 It accepts the following parameters:
28475 pix_fmts
28477 A '|'-separated list of pixel format names, such as
28478 "pix_fmts=yuv420p|monow|rgb24".
28479 Examples
28481 • Convert the input video to the yuv420p format
28483 format=pix_fmts=yuv420p
28485 Convert the input video to any of the formats in the list
28487 format=pix_fmts=yuv420p|yuv444p|yuv410p
28489 fps
28491 Convert the video to specified constant frame rate by duplicating or
28492 dropping frames as necessary.
28493 It accepts the following parameters:
28495 fps The desired output frame rate. It accepts expressions containing
28497 the following constants:
28498 source_fps
28500 The input's frame rate
28501 ntsc
28503 NTSC frame rate of "30000/1001"
28504 pal PAL frame rate of 25.0
28506 film
28508 Film frame rate of 24.0
28509 ntsc_film
28511 NTSC-film frame rate of "24000/1001"
28512 The default is 25.
28514 start_time
28516 Assume the first PTS should be the given value, in seconds. This
28517 allows for padding/trimming at the start of stream. By default, no
28518 assumption is made about the first frame's expected PTS, so no
28519 padding or trimming is done. For example, this could be set to 0
28520 to pad the beginning with duplicates of the first frame if a video
28521 stream starts after the audio stream or to trim any frames with a
28522 negative PTS.
28523 round
28525 Timestamp (PTS) rounding method.
28526 Possible values are:
28528 zero
28530 round towards 0
28531 inf round away from 0
28533 down
28535 round towards -infinity
28536 up round towards +infinity
28538 near
28540 round to nearest
28541 The default is "near".
28543 eof_action
28545 Action performed when reading the last frame.
28546 Possible values are:
28548 round
28550 Use same timestamp rounding method as used for other frames.
28551 pass
28553 Pass through last frame if input duration has not been reached
28554 yet.
28555 The default is "round".
28557 Alternatively, the options can be specified as a flat string:
28559 fps[:start_time[:round]].
28560 See also the setpts filter.
28562 Examples
28564 • A typical usage in order to set the fps to 25:
28566 fps=fps=25
28568 • Sets the fps to 24, using abbreviation and rounding method to round
28570 to nearest:
28571 fps=fps=film:round=near
28573 framepack
28575 Pack two different video streams into a stereoscopic video, setting
28576 proper metadata on supported codecs. The two views should have the same
28577 size and framerate and processing will stop when the shorter video
28578 ends. Please note that you may conveniently adjust view properties with
28579 the scale and fps filters.
28580 It accepts the following parameters:
28582 format
28584 The desired packing format. Supported values are:
28585 sbs The views are next to each other (default).
28587 tab The views are on top of each other.
28589 lines
28591 The views are packed by line.
28592 columns
28594 The views are packed by column.
28595 frameseq
28597 The views are temporally interleaved.
28598 Some examples:
28600 # Convert left and right views into a frame-sequential video
28602 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
28603 # Convert views into a side-by-side video with the same output resolution as the input
28605 ffmpeg -i LEFT -i RIGHT -filter_complex [0:v]scale=w=iw/2[left],[1:v]scale=w=iw/2[right],[left][right]framepack=sbs OUTPUT
28606 framerate
28608 Change the frame rate by interpolating new video output frames from the
28609 source frames.
28610 This filter is not designed to function correctly with interlaced
28612 media. If you wish to change the frame rate of interlaced media then
28613 you are required to deinterlace before this filter and re-interlace
28614 after this filter.
28615 A description of the accepted options follows.
28617 fps Specify the output frames per second. This option can also be
28619 specified as a value alone. The default is 50.
28620 interp_start
28622 Specify the start of a range where the output frame will be created
28623 as a linear interpolation of two frames. The range is [0-255], the
28624 default is 15.
28625 interp_end
28627 Specify the end of a range where the output frame will be created
28628 as a linear interpolation of two frames. The range is [0-255], the
28629 default is 240.
28630 scene
28632 Specify the level at which a scene change is detected as a value
28633 between 0 and 100 to indicate a new scene; a low value reflects a
28634 low probability for the current frame to introduce a new scene,
28635 while a higher value means the current frame is more likely to be
28636 one. The default is 8.2.
28637 flags
28639 Specify flags influencing the filter process.
28640 Available value for flags is:
28642 scene_change_detect, scd
28644 Enable scene change detection using the value of the option
28645 scene. This flag is enabled by default.
28646 framestep
28648 Select one frame every N-th frame.
28649 This filter accepts the following option:
28651 step
28653 Select frame after every "step" frames. Allowed values are
28654 positive integers higher than 0. Default value is 1.
28655 freezedetect
28657 Detect frozen video.
28658 This filter logs a message and sets frame metadata when it detects that
28660 the input video has no significant change in content during a specified
28661 duration. Video freeze detection calculates the mean average absolute
28662 difference of all the components of video frames and compares it to a
28663 noise floor.
28664 The printed times and duration are expressed in seconds. The
28666 "lavfi.freezedetect.freeze_start" metadata key is set on the first
28667 frame whose timestamp equals or exceeds the detection duration and it
28668 contains the timestamp of the first frame of the freeze. The
28669 "lavfi.freezedetect.freeze_duration" and
28670 "lavfi.freezedetect.freeze_end" metadata keys are set on the first
28671 frame after the freeze.
28672 The filter accepts the following options:
28674 noise, n
28676 Set noise tolerance. Can be specified in dB (in case "dB" is
28677 appended to the specified value) or as a difference ratio between 0
28678 and 1. Default is -60dB, or 0.001.
28679 duration, d
28681 Set freeze duration until notification (default is 2 seconds).
28682 freezeframes
28684 Freeze video frames.
28685 This filter freezes video frames using frame from 2nd input.
28687 The filter accepts the following options:
28689 first
28691 Set number of first frame from which to start freeze.
28692 last
28694 Set number of last frame from which to end freeze.
28695 replace
28697 Set number of frame from 2nd input which will be used instead of
28698 replaced frames.
28699 frei0r
28701 Apply a frei0r effect to the input video.
28702 To enable the compilation of this filter, you need to install the
28704 frei0r header and configure FFmpeg with "--enable-frei0r".
28705 It accepts the following parameters:
28707 filter_name
28709 The name of the frei0r effect to load. If the environment variable
28710 FREI0R_PATH is defined, the frei0r effect is searched for in each
28711 of the directories specified by the colon-separated list in
28712 FREI0R_PATH. Otherwise, the standard frei0r paths are searched, in
28713 this order: HOME/.frei0r-1/lib/, /usr/local/lib/frei0r-1/,
28714 /usr/lib/frei0r-1/.
28715 filter_params
28717 A '|'-separated list of parameters to pass to the frei0r effect.
28718 A frei0r effect parameter can be a boolean (its value is either "y" or
28720 "n"), a double, a color (specified as R/G/B, where R, G, and B are
28721 floating point numbers between 0.0 and 1.0, inclusive) or a color
28722 description as specified in the "Color" section in the ffmpeg-utils
28723 manual, a position (specified as X/Y, where X and Y are floating point
28724 numbers) and/or a string.
28725 The number and types of parameters depend on the loaded effect. If an
28727 effect parameter is not specified, the default value is set.
28728 Examples
28730 • Apply the distort0r effect, setting the first two double
28732 parameters:
28733 frei0r=filter_name=distort0r:filter_params=0.5|0.01
28735 • Apply the colordistance effect, taking a color as the first
28737 parameter:
28738 frei0r=colordistance:0.2/0.3/0.4
28740 frei0r=colordistance:violet
28741 frei0r=colordistance:0x112233
28742 • Apply the perspective effect, specifying the top left and top right
28744 image positions:
28745 frei0r=perspective:0.2/0.2|0.8/0.2
28747 For more information, see <http://frei0r.dyne.org>
28749 Commands
28751 This filter supports the filter_params option as commands.
28753 fspp
28755 Apply fast and simple postprocessing. It is a faster version of spp.
28756 It splits (I)DCT into horizontal/vertical passes. Unlike the simple
28758 post- processing filter, one of them is performed once per block, not
28759 per pixel. This allows for much higher speed.
28760 The filter accepts the following options:
28762 quality
28764 Set quality. This option defines the number of levels for
28765 averaging. It accepts an integer in the range 4-5. Default value is
28766 4.
28767 qp Force a constant quantization parameter. It accepts an integer in
28769 range 0-63. If not set, the filter will use the QP from the video
28770 stream (if available).
28771 strength
28773 Set filter strength. It accepts an integer in range -15 to 32.
28774 Lower values mean more details but also more artifacts, while
28775 higher values make the image smoother but also blurrier. Default
28776 value is 0 X PSNR optimal.
28777 use_bframe_qp
28779 Enable the use of the QP from the B-Frames if set to 1. Using this
28780 option may cause flicker since the B-Frames have often larger QP.
28781 Default is 0 (not enabled).
28782 gblur
28784 Apply Gaussian blur filter.
28785 The filter accepts the following options:
28787 sigma
28789 Set horizontal sigma, standard deviation of Gaussian blur. Default
28790 is 0.5.
28791 steps
28793 Set number of steps for Gaussian approximation. Default is 1.
28794 planes
28796 Set which planes to filter. By default all planes are filtered.
28797 sigmaV
28799 Set vertical sigma, if negative it will be same as "sigma".
28800 Default is "-1".
28801 Commands
28803 This filter supports same commands as options. The command accepts the
28805 same syntax of the corresponding option.
28806 If the specified expression is not valid, it is kept at its current
28808 value.
28809 geq
28811 Apply generic equation to each pixel.
28812 The filter accepts the following options:
28814 lum_expr, lum
28816 Set the luminance expression.
28817 cb_expr, cb
28819 Set the chrominance blue expression.
28820 cr_expr, cr
28822 Set the chrominance red expression.
28823 alpha_expr, a
28825 Set the alpha expression.
28826 red_expr, r
28828 Set the red expression.
28829 green_expr, g
28831 Set the green expression.
28832 blue_expr, b
28834 Set the blue expression.
28835 The colorspace is selected according to the specified options. If one
28837 of the lum_expr, cb_expr, or cr_expr options is specified, the filter
28838 will automatically select a YCbCr colorspace. If one of the red_expr,
28839 green_expr, or blue_expr options is specified, it will select an RGB
28840 colorspace.
28841 If one of the chrominance expression is not defined, it falls back on
28843 the other one. If no alpha expression is specified it will evaluate to
28844 opaque value. If none of chrominance expressions are specified, they
28845 will evaluate to the luminance expression.
28846 The expressions can use the following variables and functions:
28848 N The sequential number of the filtered frame, starting from 0.
28850 X
28852 Y The coordinates of the current sample.
28853 W
28855 H The width and height of the image.
28856 SW
28858 SH Width and height scale depending on the currently filtered plane.
28859 It is the ratio between the corresponding luma plane number of
28860 pixels and the current plane ones. E.g. for YUV4:2:0 the values are
28861 "1,1" for the luma plane, and "0.5,0.5" for chroma planes.
28862 T Time of the current frame, expressed in seconds.
28864 p(x, y)
28866 Return the value of the pixel at location (x,y) of the current
28867 plane.
28868 lum(x, y)
28870 Return the value of the pixel at location (x,y) of the luminance
28871 plane.
28872 cb(x, y)
28874 Return the value of the pixel at location (x,y) of the blue-
28875 difference chroma plane. Return 0 if there is no such plane.
28876 cr(x, y)
28878 Return the value of the pixel at location (x,y) of the red-
28879 difference chroma plane. Return 0 if there is no such plane.
28880 r(x, y)
28882 g(x, y)
28883 b(x, y)
28884 Return the value of the pixel at location (x,y) of the
28885 red/green/blue component. Return 0 if there is no such component.
28886 alpha(x, y)
28888 Return the value of the pixel at location (x,y) of the alpha plane.
28889 Return 0 if there is no such plane.
28890 psum(x,y), lumsum(x, y), cbsum(x,y), crsum(x,y), rsum(x,y), gsum(x,y),
28892 bsum(x,y), alphasum(x,y)
28893 Sum of sample values in the rectangle from (0,0) to (x,y), this
28894 allows obtaining sums of samples within a rectangle. See the
28895 functions without the sum postfix.
28896 interpolation
28898 Set one of interpolation methods:
28899 nearest, n
28901 bilinear, b
28902 Default is bilinear.
28904 For functions, if x and y are outside the area, the value will be
28906 automatically clipped to the closer edge.
28907 Please note that this filter can use multiple threads in which case
28909 each slice will have its own expression state. If you want to use only
28910 a single expression state because your expressions depend on previous
28911 state then you should limit the number of filter threads to 1.
28912 Examples
28914 • Flip the image horizontally:
28916 geq=p(W-X\,Y)
28918 • Generate a bidimensional sine wave, with angle "PI/3" and a
28920 wavelength of 100 pixels:
28921 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
28923 • Generate a fancy enigmatic moving light:
28925 nullsrc=s=256x256,geq=random(1)/hypot(X-cos(N*0.07)*W/2-W/2\,Y-sin(N*0.09)*H/2-H/2)^2*1000000*sin(N*0.02):128:128
28927 • Generate a quick emboss effect:
28929 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
28931 • Modify RGB components depending on pixel position:
28933 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
28935 • Create a radial gradient that is the same size as the input (also
28937 see the vignette filter):
28938 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
28940 gradfun
28942 Fix the banding artifacts that are sometimes introduced into nearly
28943 flat regions by truncation to 8-bit color depth. Interpolate the
28944 gradients that should go where the bands are, and dither them.
28945 It is designed for playback only. Do not use it prior to lossy
28947 compression, because compression tends to lose the dither and bring
28948 back the bands.
28949 It accepts the following parameters:
28951 strength
28953 The maximum amount by which the filter will change any one pixel.
28954 This is also the threshold for detecting nearly flat regions.
28955 Acceptable values range from .51 to 64; the default value is 1.2.
28956 Out-of-range values will be clipped to the valid range.
28957 radius
28959 The neighborhood to fit the gradient to. A larger radius makes for
28960 smoother gradients, but also prevents the filter from modifying the
28961 pixels near detailed regions. Acceptable values are 8-32; the
28962 default value is 16. Out-of-range values will be clipped to the
28963 valid range.
28964 Alternatively, the options can be specified as a flat string:
28966 strength[:radius]
28967 Examples
28969 • Apply the filter with a 3.5 strength and radius of 8:
28971 gradfun=3.5:8
28973 • Specify radius, omitting the strength (which will fall-back to the
28975 default value):
28976 gradfun=radius=8
28978 graphmonitor
28980 Show various filtergraph stats.
28981 With this filter one can debug complete filtergraph. Especially issues
28983 with links filling with queued frames.
28984 The filter accepts the following options:
28986 size, s
28988 Set video output size. Default is hd720.
28989 opacity, o
28991 Set video opacity. Default is 0.9. Allowed range is from 0 to 1.
28992 mode, m
28994 Set output mode, can be fulll or compact. In compact mode only
28995 filters with some queued frames have displayed stats.
28996 flags, f
28998 Set flags which enable which stats are shown in video.
28999 Available values for flags are:
29001 queue
29003 Display number of queued frames in each link.
29004 frame_count_in
29006 Display number of frames taken from filter.
29007 frame_count_out
29009 Display number of frames given out from filter.
29010 pts Display current filtered frame pts.
29012 time
29014 Display current filtered frame time.
29015 timebase
29017 Display time base for filter link.
29018 format
29020 Display used format for filter link.
29021 size
29023 Display video size or number of audio channels in case of audio
29024 used by filter link.
29025 rate
29027 Display video frame rate or sample rate in case of audio used
29028 by filter link.
29029 eof Display link output status.
29031 sample_count_in
29033 Display number of samples taken from filter.
29034 sample_count_out
29036 Display number of samples given out from filter.
29037 rate, r
29039 Set upper limit for video rate of output stream, Default value is
29040 25. This guarantee that output video frame rate will not be higher
29041 than this value.
29042 grayworld
29044 A color constancy filter that applies color correction based on the
29045 grayworld assumption
29046 See:
29048 <https://www.researchgate.net/publication/275213614_A_New_Color_Correction_Method_for_Underwater_Imaging>
29049 The algorithm uses linear light, so input data should be linearized
29051 beforehand (and possibly correctly tagged).
29052 ffmpeg -i INPUT -vf zscale=transfer=linear,grayworld,zscale=transfer=bt709,format=yuv420p OUTPUT
29054 greyedge
29056 A color constancy variation filter which estimates scene illumination
29057 via grey edge algorithm and corrects the scene colors accordingly.
29058 See: <https://staff.science.uva.nl/th.gevers/pub/GeversTIP07.pdf>
29060 The filter accepts the following options:
29062 difford
29064 The order of differentiation to be applied on the scene. Must be
29065 chosen in the range [0,2] and default value is 1.
29066 minknorm
29068 The Minkowski parameter to be used for calculating the Minkowski
29069 distance. Must be chosen in the range [0,20] and default value is
29070 1. Set to 0 for getting max value instead of calculating Minkowski
29071 distance.
29072 sigma
29074 The standard deviation of Gaussian blur to be applied on the scene.
29075 Must be chosen in the range [0,1024.0] and default value = 1.
29076 floor( sigma * break_off_sigma(3) ) can't be equal to 0 if difford
29077 is greater than 0.
29078 Examples
29080 • Grey Edge:
29082 greyedge=difford=1:minknorm=5:sigma=2
29084 • Max Edge:
29086 greyedge=difford=1:minknorm=0:sigma=2
29088 guided
29090 Apply guided filter for edge-preserving smoothing, dehazing and so on.
29091 The filter accepts the following options:
29093 radius
29095 Set the box radius in pixels. Allowed range is 1 to 20. Default is
29096 3.
29097 eps Set regularization parameter (with square). Allowed range is 0 to
29099 1. Default is 0.01.
29100 mode
29102 Set filter mode. Can be "basic" or "fast". Default is "basic".
29103 sub Set subsampling ratio for "fast" mode. Range is 2 to 64. Default
29105 is 4. No subsampling occurs in "basic" mode.
29106 guidance
29108 Set guidance mode. Can be "off" or "on". Default is "off". If
29109 "off", single input is required. If "on", two inputs of the same
29110 resolution and pixel format are required. The second input serves
29111 as the guidance.
29112 planes
29114 Set planes to filter. Default is first only.
29115 Commands
29117 This filter supports the all above options as commands.
29119 Examples
29121 • Edge-preserving smoothing with guided filter:
29123 ffmpeg -i in.png -vf guided out.png
29125 • Dehazing, structure-transferring filtering, detail enhancement with
29127 guided filter. For the generation of guidance image, refer to
29128 paper "Guided Image Filtering". See:
29129 <http://kaiminghe.com/publications/pami12guidedfilter.pdf>.
29130 ffmpeg -i in.png -i guidance.png -filter_complex guided=guidance=on out.png
29132 haldclut
29134 Apply a Hald CLUT to a video stream.
29135 First input is the video stream to process, and second one is the Hald
29137 CLUT. The Hald CLUT input can be a simple picture or a complete video
29138 stream.
29139 The filter accepts the following options:
29141 shortest
29143 Force termination when the shortest input terminates. Default is 0.
29144 repeatlast
29146 Continue applying the last CLUT after the end of the stream. A
29147 value of 0 disable the filter after the last frame of the CLUT is
29148 reached. Default is 1.
29149 "haldclut" also has the same interpolation options as lut3d (both
29151 filters share the same internals).
29152 This filter also supports the framesync options.
29154 More information about the Hald CLUT can be found on Eskil Steenberg's
29156 website (Hald CLUT author) at
29157 <http://www.quelsolaar.com/technology/clut.html>.
29158 Commands
29160 This filter supports the "interp" option as commands.
29162 Workflow examples
29164 Hald CLUT video stream
29166 Generate an identity Hald CLUT stream altered with various effects:
29168 ffmpeg -f lavfi -i B<haldclutsrc>=8 -vf "hue=H=2*PI*t:s=sin(2*PI*t)+1, curves=cross_process" -t 10 -c:v ffv1 clut.nut
29170 Note: make sure you use a lossless codec.
29172 Then use it with "haldclut" to apply it on some random stream:
29174 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
29176 The Hald CLUT will be applied to the 10 first seconds (duration of
29178 clut.nut), then the latest picture of that CLUT stream will be applied
29179 to the remaining frames of the "mandelbrot" stream.
29180 Hald CLUT with preview
29182 A Hald CLUT is supposed to be a squared image of "Level*Level*Level" by
29184 "Level*Level*Level" pixels. For a given Hald CLUT, FFmpeg will select
29185 the biggest possible square starting at the top left of the picture.
29186 The remaining padding pixels (bottom or right) will be ignored. This
29187 area can be used to add a preview of the Hald CLUT.
29188 Typically, the following generated Hald CLUT will be supported by the
29190 "haldclut" filter:
29191 ffmpeg -f lavfi -i B<haldclutsrc>=8 -vf "
29193 pad=iw+320 [padded_clut];
29194 smptebars=s=320x256, split [a][b];
29195 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
29196 [main][b] overlay=W-320" -frames:v 1 clut.png
29197 It contains the original and a preview of the effect of the CLUT: SMPTE
29199 color bars are displayed on the right-top, and below the same color
29200 bars processed by the color changes.
29201 Then, the effect of this Hald CLUT can be visualized with:
29203 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
29205 hflip
29207 Flip the input video horizontally.
29208 For example, to horizontally flip the input video with ffmpeg:
29210 ffmpeg -i in.avi -vf "hflip" out.avi
29212 histeq
29214 This filter applies a global color histogram equalization on a per-
29215 frame basis.
29216 It can be used to correct video that has a compressed range of pixel
29218 intensities. The filter redistributes the pixel intensities to
29219 equalize their distribution across the intensity range. It may be
29220 viewed as an "automatically adjusting contrast filter". This filter is
29221 useful only for correcting degraded or poorly captured source video.
29222 The filter accepts the following options:
29224 strength
29226 Determine the amount of equalization to be applied. As the
29227 strength is reduced, the distribution of pixel intensities more-
29228 and-more approaches that of the input frame. The value must be a
29229 float number in the range [0,1] and defaults to 0.200.
29230 intensity
29232 Set the maximum intensity that can generated and scale the output
29233 values appropriately. The strength should be set as desired and
29234 then the intensity can be limited if needed to avoid washing-out.
29235 The value must be a float number in the range [0,1] and defaults to
29236 0.210.
29237 antibanding
29239 Set the antibanding level. If enabled the filter will randomly vary
29240 the luminance of output pixels by a small amount to avoid banding
29241 of the histogram. Possible values are "none", "weak" or "strong".
29242 It defaults to "none".
29243 histogram
29245 Compute and draw a color distribution histogram for the input video.
29246 The computed histogram is a representation of the color component
29248 distribution in an image.
29249 Standard histogram displays the color components distribution in an
29251 image. Displays color graph for each color component. Shows
29252 distribution of the Y, U, V, A or R, G, B components, depending on
29253 input format, in the current frame. Below each graph a color component
29254 scale meter is shown.
29255 The filter accepts the following options:
29257 level_height
29259 Set height of level. Default value is 200. Allowed range is [50,
29260 2048].
29261 scale_height
29263 Set height of color scale. Default value is 12. Allowed range is
29264 [0, 40].
29265 display_mode
29267 Set display mode. It accepts the following values:
29268 stack
29270 Per color component graphs are placed below each other.
29271 parade
29273 Per color component graphs are placed side by side.
29274 overlay
29276 Presents information identical to that in the "parade", except
29277 that the graphs representing color components are superimposed
29278 directly over one another.
29279 Default is "stack".
29281 levels_mode
29283 Set mode. Can be either "linear", or "logarithmic". Default is
29284 "linear".
29285 components
29287 Set what color components to display. Default is 7.
29288 fgopacity
29290 Set foreground opacity. Default is 0.7.
29291 bgopacity
29293 Set background opacity. Default is 0.5.
29294 colors_mode
29296 Set colors mode. It accepts the following values:
29297 whiteonblack
29299 blackonwhite
29300 whiteongray
29301 blackongray
29302 coloronblack
29303 coloronwhite
29304 colorongray
29305 blackoncolor
29306 whiteoncolor
29307 grayoncolor
29308 Default is "whiteonblack".
29310 Examples
29312 • Calculate and draw histogram:
29314 ffplay -i input -vf histogram
29316 hqdn3d
29318 This is a high precision/quality 3d denoise filter. It aims to reduce
29319 image noise, producing smooth images and making still images really
29320 still. It should enhance compressibility.
29321 It accepts the following optional parameters:
29323 luma_spatial
29325 A non-negative floating point number which specifies spatial luma
29326 strength. It defaults to 4.0.
29327 chroma_spatial
29329 A non-negative floating point number which specifies spatial chroma
29330 strength. It defaults to 3.0*luma_spatial/4.0.
29331 luma_tmp
29333 A floating point number which specifies luma temporal strength. It
29334 defaults to 6.0*luma_spatial/4.0.
29335 chroma_tmp
29337 A floating point number which specifies chroma temporal strength.
29338 It defaults to luma_tmp*chroma_spatial/luma_spatial.
29339 Commands
29341 This filter supports same commands as options. The command accepts the
29343 same syntax of the corresponding option.
29344 If the specified expression is not valid, it is kept at its current
29346 value.
29347 hwdownload
29349 Download hardware frames to system memory.
29350 The input must be in hardware frames, and the output a non-hardware
29352 format. Not all formats will be supported on the output - it may be
29353 necessary to insert an additional format filter immediately following
29354 in the graph to get the output in a supported format.
29355 hwmap
29357 Map hardware frames to system memory or to another device.
29358 This filter has several different modes of operation; which one is used
29360 depends on the input and output formats:
29361 • Hardware frame input, normal frame output
29363 Map the input frames to system memory and pass them to the output.
29365 If the original hardware frame is later required (for example,
29366 after overlaying something else on part of it), the hwmap filter
29367 can be used again in the next mode to retrieve it.
29368 • Normal frame input, hardware frame output
29370 If the input is actually a software-mapped hardware frame, then
29372 unmap it - that is, return the original hardware frame.
29373 Otherwise, a device must be provided. Create new hardware surfaces
29375 on that device for the output, then map them back to the software
29376 format at the input and give those frames to the preceding filter.
29377 This will then act like the hwupload filter, but may be able to
29378 avoid an additional copy when the input is already in a compatible
29379 format.
29380 • Hardware frame input and output
29382 A device must be supplied for the output, either directly or with
29384 the derive_device option. The input and output devices must be of
29385 different types and compatible - the exact meaning of this is
29386 system-dependent, but typically it means that they must refer to
29387 the same underlying hardware context (for example, refer to the
29388 same graphics card).
29389 If the input frames were originally created on the output device,
29391 then unmap to retrieve the original frames.
29392 Otherwise, map the frames to the output device - create new
29394 hardware frames on the output corresponding to the frames on the
29395 input.
29396 The following additional parameters are accepted:
29398 mode
29400 Set the frame mapping mode. Some combination of:
29401 read
29403 The mapped frame should be readable.
29404 write
29406 The mapped frame should be writeable.
29407 overwrite
29409 The mapping will always overwrite the entire frame.
29410 This may improve performance in some cases, as the original
29412 contents of the frame need not be loaded.
29413 direct
29415 The mapping must not involve any copying.
29416 Indirect mappings to copies of frames are created in some cases
29418 where either direct mapping is not possible or it would have
29419 unexpected properties. Setting this flag ensures that the
29420 mapping is direct and will fail if that is not possible.
29421 Defaults to read+write if not specified.
29423 derive_device type
29425 Rather than using the device supplied at initialisation, instead
29426 derive a new device of type type from the device the input frames
29427 exist on.
29428 reverse
29430 In a hardware to hardware mapping, map in reverse - create frames
29431 in the sink and map them back to the source. This may be necessary
29432 in some cases where a mapping in one direction is required but only
29433 the opposite direction is supported by the devices being used.
29434 This option is dangerous - it may break the preceding filter in
29436 undefined ways if there are any additional constraints on that
29437 filter's output. Do not use it without fully understanding the
29438 implications of its use.
29439 hwupload
29441 Upload system memory frames to hardware surfaces.
29442 The device to upload to must be supplied when the filter is
29444 initialised. If using ffmpeg, select the appropriate device with the
29445 -filter_hw_device option or with the derive_device option. The input
29446 and output devices must be of different types and compatible - the
29447 exact meaning of this is system-dependent, but typically it means that
29448 they must refer to the same underlying hardware context (for example,
29449 refer to the same graphics card).
29450 The following additional parameters are accepted:
29452 derive_device type
29454 Rather than using the device supplied at initialisation, instead
29455 derive a new device of type type from the device the input frames
29456 exist on.
29457 hwupload_cuda
29459 Upload system memory frames to a CUDA device.
29460 It accepts the following optional parameters:
29462 device
29464 The number of the CUDA device to use
29465 hqx
29467 Apply a high-quality magnification filter designed for pixel art. This
29468 filter was originally created by Maxim Stepin.
29469 It accepts the following option:
29471 n Set the scaling dimension: 2 for "hq2x", 3 for "hq3x" and 4 for
29473 "hq4x". Default is 3.
29474 hstack
29476 Stack input videos horizontally.
29477 All streams must be of same pixel format and of same height.
29479 Note that this filter is faster than using overlay and pad filter to
29481 create same output.
29482 The filter accepts the following option:
29484 inputs
29486 Set number of input streams. Default is 2.
29487 shortest
29489 If set to 1, force the output to terminate when the shortest input
29490 terminates. Default value is 0.
29491 hsvhold
29493 Turns a certain HSV range into gray values.
29494 This filter measures color difference between set HSV color in options
29496 and ones measured in video stream. Depending on options, output colors
29497 can be changed to be gray or not.
29498 The filter accepts the following options:
29500 hue Set the hue value which will be used in color difference
29502 calculation. Allowed range is from -360 to 360. Default value is
29503 0.
29504 sat Set the saturation value which will be used in color difference
29506 calculation. Allowed range is from -1 to 1. Default value is 0.
29507 val Set the value which will be used in color difference calculation.
29509 Allowed range is from -1 to 1. Default value is 0.
29510 similarity
29512 Set similarity percentage with the key color. Allowed range is
29513 from 0 to 1. Default value is 0.01.
29514 0.00001 matches only the exact key color, while 1.0 matches
29516 everything.
29517 blend
29519 Blend percentage. Allowed range is from 0 to 1. Default value is
29520 0.
29521 0.0 makes pixels either fully gray, or not gray at all.
29523 Higher values result in more gray pixels, with a higher gray pixel
29525 the more similar the pixels color is to the key color.
29526 hsvkey
29528 Turns a certain HSV range into transparency.
29529 This filter measures color difference between set HSV color in options
29531 and ones measured in video stream. Depending on options, output colors
29532 can be changed to transparent by adding alpha channel.
29533 The filter accepts the following options:
29535 hue Set the hue value which will be used in color difference
29537 calculation. Allowed range is from -360 to 360. Default value is
29538 0.
29539 sat Set the saturation value which will be used in color difference
29541 calculation. Allowed range is from -1 to 1. Default value is 0.
29542 val Set the value which will be used in color difference calculation.
29544 Allowed range is from -1 to 1. Default value is 0.
29545 similarity
29547 Set similarity percentage with the key color. Allowed range is
29548 from 0 to 1. Default value is 0.01.
29549 0.00001 matches only the exact key color, while 1.0 matches
29551 everything.
29552 blend
29554 Blend percentage. Allowed range is from 0 to 1. Default value is
29555 0.
29556 0.0 makes pixels either fully transparent, or not transparent at
29558 all.
29559 Higher values result in semi-transparent pixels, with a higher
29561 transparency the more similar the pixels color is to the key color.
29562 hue
29564 Modify the hue and/or the saturation of the input.
29565 It accepts the following parameters:
29567 h Specify the hue angle as a number of degrees. It accepts an
29569 expression, and defaults to "0".
29570 s Specify the saturation in the [-10,10] range. It accepts an
29572 expression and defaults to "1".
29573 H Specify the hue angle as a number of radians. It accepts an
29575 expression, and defaults to "0".
29576 b Specify the brightness in the [-10,10] range. It accepts an
29578 expression and defaults to "0".
29579 h and H are mutually exclusive, and can't be specified at the same
29581 time.
29582 The b, h, H and s option values are expressions containing the
29584 following constants:
29585 n frame count of the input frame starting from 0
29587 pts presentation timestamp of the input frame expressed in time base
29589 units
29590 r frame rate of the input video, NAN if the input frame rate is
29592 unknown
29593 t timestamp expressed in seconds, NAN if the input timestamp is
29595 unknown
29596 tb time base of the input video
29598 Examples
29600 • Set the hue to 90 degrees and the saturation to 1.0:
29602 hue=h=90:s=1
29604 • Same command but expressing the hue in radians:
29606 hue=H=PI/2:s=1
29608 • Rotate hue and make the saturation swing between 0 and 2 over a
29610 period of 1 second:
29611 hue="H=2*PI*t: s=sin(2*PI*t)+1"
29613 • Apply a 3 seconds saturation fade-in effect starting at 0:
29615 hue="s=min(t/3\,1)"
29617 The general fade-in expression can be written as:
29619 hue="s=min(0\, max((t-START)/DURATION\, 1))"
29621 • Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
29623 hue="s=max(0\, min(1\, (8-t)/3))"
29625 The general fade-out expression can be written as:
29627 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
29629 Commands
29631 This filter supports the following commands:
29633 b
29635 s
29636 h
29637 H Modify the hue and/or the saturation and/or brightness of the input
29638 video. The command accepts the same syntax of the corresponding
29639 option.
29640 If the specified expression is not valid, it is kept at its current
29642 value.
29643 huesaturation
29645 Apply hue-saturation-intensity adjustments to input video stream.
29646 This filter operates in RGB colorspace.
29648 This filter accepts the following options:
29650 hue Set the hue shift in degrees to apply. Default is 0. Allowed range
29652 is from -180 to 180.
29653 saturation
29655 Set the saturation shift. Default is 0. Allowed range is from -1
29656 to 1.
29657 intensity
29659 Set the intensity shift. Default is 0. Allowed range is from -1 to
29660 1.
29661 colors
29663 Set which primary and complementary colors are going to be
29664 adjusted. This options is set by providing one or multiple values.
29665 This can select multiple colors at once. By default all colors are
29666 selected.
29667 r Adjust reds.
29669 y Adjust yellows.
29671 g Adjust greens.
29673 c Adjust cyans.
29675 b Adjust blues.
29677 m Adjust magentas.
29679 a Adjust all colors.
29681 strength
29683 Set strength of filtering. Allowed range is from 0 to 100. Default
29684 value is 1.
29685 rw, gw, bw
29687 Set weight for each RGB component. Allowed range is from 0 to 1.
29688 By default is set to 0.333, 0.334, 0.333. Those options are used
29689 in saturation and lightess processing.
29690 lightness
29692 Set preserving lightness, by default is disabled. Adjusting hues
29693 can change lightness from original RGB triplet, with this option
29694 enabled lightness is kept at same value.
29695 hysteresis
29697 Grow first stream into second stream by connecting components. This
29698 makes it possible to build more robust edge masks.
29699 This filter accepts the following options:
29701 planes
29703 Set which planes will be processed as bitmap, unprocessed planes
29704 will be copied from first stream. By default value 0xf, all planes
29705 will be processed.
29706 threshold
29708 Set threshold which is used in filtering. If pixel component value
29709 is higher than this value filter algorithm for connecting
29710 components is activated. By default value is 0.
29711 The "hysteresis" filter also supports the framesync options.
29713 identity
29715 Obtain the identity score between two input videos.
29716 This filter takes two input videos.
29718 Both input videos must have the same resolution and pixel format for
29720 this filter to work correctly. Also it assumes that both inputs have
29721 the same number of frames, which are compared one by one.
29722 The obtained per component, average, min and max identity score is
29724 printed through the logging system.
29725 The filter stores the calculated identity scores of each frame in frame
29727 metadata.
29728 In the below example the input file main.mpg being processed is
29730 compared with the reference file ref.mpg.
29731 ffmpeg -i main.mpg -i ref.mpg -lavfi identity -f null -
29733 idet
29735 Detect video interlacing type.
29736 This filter tries to detect if the input frames are interlaced,
29738 progressive, top or bottom field first. It will also try to detect
29739 fields that are repeated between adjacent frames (a sign of telecine).
29740 Single frame detection considers only immediately adjacent frames when
29742 classifying each frame. Multiple frame detection incorporates the
29743 classification history of previous frames.
29744 The filter will log these metadata values:
29746 single.current_frame
29748 Detected type of current frame using single-frame detection. One
29749 of: ``tff'' (top field first), ``bff'' (bottom field first),
29750 ``progressive'', or ``undetermined''
29751 single.tff
29753 Cumulative number of frames detected as top field first using
29754 single-frame detection.
29755 multiple.tff
29757 Cumulative number of frames detected as top field first using
29758 multiple-frame detection.
29759 single.bff
29761 Cumulative number of frames detected as bottom field first using
29762 single-frame detection.
29763 multiple.current_frame
29765 Detected type of current frame using multiple-frame detection. One
29766 of: ``tff'' (top field first), ``bff'' (bottom field first),
29767 ``progressive'', or ``undetermined''
29768 multiple.bff
29770 Cumulative number of frames detected as bottom field first using
29771 multiple-frame detection.
29772 single.progressive
29774 Cumulative number of frames detected as progressive using single-
29775 frame detection.
29776 multiple.progressive
29778 Cumulative number of frames detected as progressive using multiple-
29779 frame detection.
29780 single.undetermined
29782 Cumulative number of frames that could not be classified using
29783 single-frame detection.
29784 multiple.undetermined
29786 Cumulative number of frames that could not be classified using
29787 multiple-frame detection.
29788 repeated.current_frame
29790 Which field in the current frame is repeated from the last. One of
29791 ``neither'', ``top'', or ``bottom''.
29792 repeated.neither
29794 Cumulative number of frames with no repeated field.
29795 repeated.top
29797 Cumulative number of frames with the top field repeated from the
29798 previous frame's top field.
29799 repeated.bottom
29801 Cumulative number of frames with the bottom field repeated from the
29802 previous frame's bottom field.
29803 The filter accepts the following options:
29805 intl_thres
29807 Set interlacing threshold.
29808 prog_thres
29810 Set progressive threshold.
29811 rep_thres
29813 Threshold for repeated field detection.
29814 half_life
29816 Number of frames after which a given frame's contribution to the
29817 statistics is halved (i.e., it contributes only 0.5 to its
29818 classification). The default of 0 means that all frames seen are
29819 given full weight of 1.0 forever.
29820 analyze_interlaced_flag
29822 When this is not 0 then idet will use the specified number of
29823 frames to determine if the interlaced flag is accurate, it will not
29824 count undetermined frames. If the flag is found to be accurate it
29825 will be used without any further computations, if it is found to be
29826 inaccurate it will be cleared without any further computations.
29827 This allows inserting the idet filter as a low computational method
29828 to clean up the interlaced flag
29829 il
29831 Deinterleave or interleave fields.
29832 This filter allows one to process interlaced images fields without
29834 deinterlacing them. Deinterleaving splits the input frame into 2 fields
29835 (so called half pictures). Odd lines are moved to the top half of the
29836 output image, even lines to the bottom half. You can process (filter)
29837 them independently and then re-interleave them.
29838 The filter accepts the following options:
29840 luma_mode, l
29842 chroma_mode, c
29843 alpha_mode, a
29844 Available values for luma_mode, chroma_mode and alpha_mode are:
29845 none
29847 Do nothing.
29848 deinterleave, d
29850 Deinterleave fields, placing one above the other.
29851 interleave, i
29853 Interleave fields. Reverse the effect of deinterleaving.
29854 Default value is "none".
29856 luma_swap, ls
29858 chroma_swap, cs
29859 alpha_swap, as
29860 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default
29861 value is 0.
29862 Commands
29864 This filter supports the all above options as commands.
29866 inflate
29868 Apply inflate effect to the video.
29869 This filter replaces the pixel by the local(3x3) average by taking into
29871 account only values higher than the pixel.
29872 It accepts the following options:
29874 threshold0
29876 threshold1
29877 threshold2
29878 threshold3
29879 Limit the maximum change for each plane, default is 65535. If 0,
29880 plane will remain unchanged.
29881 Commands
29883 This filter supports the all above options as commands.
29885 interlace
29887 Simple interlacing filter from progressive contents. This interleaves
29888 upper (or lower) lines from odd frames with lower (or upper) lines from
29889 even frames, halving the frame rate and preserving image height.
29890 Original Original New Frame
29892 Frame 'j' Frame 'j+1' (tff)
29893 ========== =========== ==================
29894 Line 0 --------------------> Frame 'j' Line 0
29895 Line 1 Line 1 ----> Frame 'j+1' Line 1
29896 Line 2 ---------------------> Frame 'j' Line 2
29897 Line 3 Line 3 ----> Frame 'j+1' Line 3
29898 ... ... ...
29899 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
29900 It accepts the following optional parameters:
29902 scan
29904 This determines whether the interlaced frame is taken from the even
29905 (tff - default) or odd (bff) lines of the progressive frame.
29906 lowpass
29908 Vertical lowpass filter to avoid twitter interlacing and reduce
29909 moire patterns.
29910 0, off
29912 Disable vertical lowpass filter
29913 1, linear
29915 Enable linear filter (default)
29916 2, complex
29918 Enable complex filter. This will slightly less reduce twitter
29919 and moire but better retain detail and subjective sharpness
29920 impression.
29921 kerndeint
29923 Deinterlace input video by applying Donald Graft's adaptive kernel
29924 deinterling. Work on interlaced parts of a video to produce progressive
29925 frames.
29926 The description of the accepted parameters follows.
29928 thresh
29930 Set the threshold which affects the filter's tolerance when
29931 determining if a pixel line must be processed. It must be an
29932 integer in the range [0,255] and defaults to 10. A value of 0 will
29933 result in applying the process on every pixels.
29934 map Paint pixels exceeding the threshold value to white if set to 1.
29936 Default is 0.
29937 order
29939 Set the fields order. Swap fields if set to 1, leave fields alone
29940 if 0. Default is 0.
29941 sharp
29943 Enable additional sharpening if set to 1. Default is 0.
29944 twoway
29946 Enable twoway sharpening if set to 1. Default is 0.
29947 Examples
29949 • Apply default values:
29951 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
29953 • Enable additional sharpening:
29955 kerndeint=sharp=1
29957 • Paint processed pixels in white:
29959 kerndeint=map=1
29961 kirsch
29963 Apply kirsch operator to input video stream.
29964 The filter accepts the following option:
29966 planes
29968 Set which planes will be processed, unprocessed planes will be
29969 copied. By default value 0xf, all planes will be processed.
29970 scale
29972 Set value which will be multiplied with filtered result.
29973 delta
29975 Set value which will be added to filtered result.
29976 Commands
29978 This filter supports the all above options as commands.
29980 lagfun
29982 Slowly update darker pixels.
29983 This filter makes short flashes of light appear longer. This filter
29985 accepts the following options:
29986 decay
29988 Set factor for decaying. Default is .95. Allowed range is from 0 to
29989 1.
29990 planes
29992 Set which planes to filter. Default is all. Allowed range is from 0
29993 to 15.
29994 Commands
29996 This filter supports the all above options as commands.
29998 lenscorrection
30000 Correct radial lens distortion
30001 This filter can be used to correct for radial distortion as can result
30003 from the use of wide angle lenses, and thereby re-rectify the image. To
30004 find the right parameters one can use tools available for example as
30005 part of opencv or simply trial-and-error. To use opencv use the
30006 calibration sample (under samples/cpp) from the opencv sources and
30007 extract the k1 and k2 coefficients from the resulting matrix.
30008 Note that effectively the same filter is available in the open-source
30010 tools Krita and Digikam from the KDE project.
30011 In contrast to the vignette filter, which can also be used to
30013 compensate lens errors, this filter corrects the distortion of the
30014 image, whereas vignette corrects the brightness distribution, so you
30015 may want to use both filters together in certain cases, though you will
30016 have to take care of ordering, i.e. whether vignetting should be
30017 applied before or after lens correction.
30018 Options
30020 The filter accepts the following options:
30022 cx Relative x-coordinate of the focal point of the image, and thereby
30024 the center of the distortion. This value has a range [0,1] and is
30025 expressed as fractions of the image width. Default is 0.5.
30026 cy Relative y-coordinate of the focal point of the image, and thereby
30028 the center of the distortion. This value has a range [0,1] and is
30029 expressed as fractions of the image height. Default is 0.5.
30030 k1 Coefficient of the quadratic correction term. This value has a
30032 range [-1,1]. 0 means no correction. Default is 0.
30033 k2 Coefficient of the double quadratic correction term. This value has
30035 a range [-1,1]. 0 means no correction. Default is 0.
30036 i Set interpolation type. Can be "nearest" or "bilinear". Default is
30038 "nearest".
30039 fc Specify the color of the unmapped pixels. For the syntax of this
30041 option, check the "Color" section in the ffmpeg-utils manual.
30042 Default color is "black@0".
30043 The formula that generates the correction is:
30045 r_src = r_tgt * (1 + k1 * (r_tgt / r_0)^2 + k2 * (r_tgt / r_0)^4)
30047 where r_0 is halve of the image diagonal and r_src and r_tgt are the
30049 distances from the focal point in the source and target images,
30050 respectively.
30051 Commands
30053 This filter supports the all above options as commands.
30055 lensfun
30057 Apply lens correction via the lensfun library
30058 (<http://lensfun.sourceforge.net/>).
30059 The "lensfun" filter requires the camera make, camera model, and lens
30061 model to apply the lens correction. The filter will load the lensfun
30062 database and query it to find the corresponding camera and lens entries
30063 in the database. As long as these entries can be found with the given
30064 options, the filter can perform corrections on frames. Note that
30065 incomplete strings will result in the filter choosing the best match
30066 with the given options, and the filter will output the chosen camera
30067 and lens models (logged with level "info"). You must provide the make,
30068 camera model, and lens model as they are required.
30069 The filter accepts the following options:
30071 make
30073 The make of the camera (for example, "Canon"). This option is
30074 required.
30075 model
30077 The model of the camera (for example, "Canon EOS 100D"). This
30078 option is required.
30079 lens_model
30081 The model of the lens (for example, "Canon EF-S 18-55mm f/3.5-5.6
30082 IS STM"). This option is required.
30083 mode
30085 The type of correction to apply. The following values are valid
30086 options:
30087 vignetting
30089 Enables fixing lens vignetting.
30090 geometry
30092 Enables fixing lens geometry. This is the default.
30093 subpixel
30095 Enables fixing chromatic aberrations.
30096 vig_geo
30098 Enables fixing lens vignetting and lens geometry.
30099 vig_subpixel
30101 Enables fixing lens vignetting and chromatic aberrations.
30102 distortion
30104 Enables fixing both lens geometry and chromatic aberrations.
30105 all Enables all possible corrections.
30107 focal_length
30109 The focal length of the image/video (zoom; expected constant for
30110 video). For example, a 18--55mm lens has focal length range of
30111 [18--55], so a value in that range should be chosen when using that
30112 lens. Default 18.
30113 aperture
30115 The aperture of the image/video (expected constant for video). Note
30116 that aperture is only used for vignetting correction. Default 3.5.
30117 focus_distance
30119 The focus distance of the image/video (expected constant for
30120 video). Note that focus distance is only used for vignetting and
30121 only slightly affects the vignetting correction process. If
30122 unknown, leave it at the default value (which is 1000).
30123 scale
30125 The scale factor which is applied after transformation. After
30126 correction the video is no longer necessarily rectangular. This
30127 parameter controls how much of the resulting image is visible. The
30128 value 0 means that a value will be chosen automatically such that
30129 there is little or no unmapped area in the output image. 1.0 means
30130 that no additional scaling is done. Lower values may result in more
30131 of the corrected image being visible, while higher values may avoid
30132 unmapped areas in the output.
30133 target_geometry
30135 The target geometry of the output image/video. The following values
30136 are valid options:
30137 rectilinear (default)
30139 fisheye
30140 panoramic
30141 equirectangular
30142 fisheye_orthographic
30143 fisheye_stereographic
30144 fisheye_equisolid
30145 fisheye_thoby
30146 reverse
30147 Apply the reverse of image correction (instead of correcting
30148 distortion, apply it).
30149 interpolation
30151 The type of interpolation used when correcting distortion. The
30152 following values are valid options:
30153 nearest
30155 linear (default)
30156 lanczos
30157 Examples
30159 • Apply lens correction with make "Canon", camera model "Canon EOS
30161 100D", and lens model "Canon EF-S 18-55mm f/3.5-5.6 IS STM" with
30162 focal length of "18" and aperture of "8.0".
30163 ffmpeg -i input.mov -vf lensfun=make=Canon:model="Canon EOS 100D":lens_model="Canon EF-S 18-55mm f/3.5-5.6 IS STM":focal_length=18:aperture=8 -c:v h264 -b:v 8000k output.mov
30165 • Apply the same as before, but only for the first 5 seconds of
30167 video.
30168 ffmpeg -i input.mov -vf lensfun=make=Canon:model="Canon EOS 100D":lens_model="Canon EF-S 18-55mm f/3.5-5.6 IS STM":focal_length=18:aperture=8:enable='lte(t\,5)' -c:v h264 -b:v 8000k output.mov
30170 libvmaf
30172 Obtain the VMAF (Video Multi-Method Assessment Fusion) score between
30173 two input videos.
30174 The first input is the encoded video, and the second input is the
30176 reference video.
30177 The obtained VMAF score is printed through the logging system.
30179 It requires Netflix's vmaf library (libvmaf) as a pre-requisite. After
30181 installing the library it can be enabled using: "./configure
30182 --enable-libvmaf". If no model path is specified it uses the default
30183 model: "vmaf_v0.6.1.pkl".
30184 The filter has following options:
30186 model_path
30188 Set the model path which is to be used for SVM. Default value:
30189 "/usr/local/share/model/vmaf_v0.6.1.pkl"
30190 log_path
30192 Set the file path to be used to store logs.
30193 log_fmt
30195 Set the format of the log file (csv, json or xml).
30196 enable_transform
30198 This option can enable/disable the "score_transform" applied to the
30199 final predicted VMAF score, if you have specified score_transform
30200 option in the input parameter file passed to "run_vmaf_training.py"
30201 Default value: "false"
30202 phone_model
30204 Invokes the phone model which will generate VMAF scores higher than
30205 in the regular model, which is more suitable for laptop, TV, etc.
30206 viewing conditions. Default value: "false"
30207 psnr
30209 Enables computing psnr along with vmaf. Default value: "false"
30210 ssim
30212 Enables computing ssim along with vmaf. Default value: "false"
30213 ms_ssim
30215 Enables computing ms_ssim along with vmaf. Default value: "false"
30216 pool
30218 Set the pool method to be used for computing vmaf. Options are
30219 "min", "harmonic_mean" or "mean" (default).
30220 n_threads
30222 Set number of threads to be used when computing vmaf. Default
30223 value: 0, which makes use of all available logical processors.
30224 n_subsample
30226 Set interval for frame subsampling used when computing vmaf.
30227 Default value: 1
30228 enable_conf_interval
30230 Enables confidence interval. Default value: "false"
30231 This filter also supports the framesync options.
30233 Examples
30235 • On the below examples the input file main.mpg being processed is
30237 compared with the reference file ref.mpg.
30238 ffmpeg -i main.mpg -i ref.mpg -lavfi libvmaf -f null -
30240 • Example with options:
30242 ffmpeg -i main.mpg -i ref.mpg -lavfi libvmaf="psnr=1:log_fmt=json" -f null -
30244 • Example with options and different containers:
30246 ffmpeg -i main.mpg -i ref.mkv -lavfi "[0:v]settb=AVTB,setpts=PTS-STARTPTS[main];[1:v]settb=AVTB,setpts=PTS-STARTPTS[ref];[main][ref]libvmaf=psnr=1:log_fmt=json" -f null -
30248 limitdiff
30250 Apply limited difference filter using second and optionally third video
30251 stream.
30252 The filter accepts the following options:
30254 threshold
30256 Set the threshold to use when allowing certain differences between
30257 video streams. Any absolute difference value lower or exact than
30258 this threshold will pick pixel components from first video stream.
30259 elasticity
30261 Set the elasticity of soft thresholding when processing video
30262 streams. This value multiplied with first one sets second
30263 threshold. Any absolute difference value greater or exact than
30264 second threshold will pick pixel components from second video
30265 stream. For values between those two threshold linear interpolation
30266 between first and second video stream will be used.
30267 reference
30269 Enable the reference (third) video stream processing. By default is
30270 disabled. If set, this video stream will be used for calculating
30271 absolute difference with first video stream.
30272 planes
30274 Specify which planes will be processed. Defaults to all available.
30275 Commands
30277 This filter supports the all above options as commands except option
30279 reference.
30280 limiter
30282 Limits the pixel components values to the specified range [min, max].
30283 The filter accepts the following options:
30285 min Lower bound. Defaults to the lowest allowed value for the input.
30287 max Upper bound. Defaults to the highest allowed value for the input.
30289 planes
30291 Specify which planes will be processed. Defaults to all available.
30292 Commands
30294 This filter supports the all above options as commands.
30296 loop
30298 Loop video frames.
30299 The filter accepts the following options:
30301 loop
30303 Set the number of loops. Setting this value to -1 will result in
30304 infinite loops. Default is 0.
30305 size
30307 Set maximal size in number of frames. Default is 0.
30308 start
30310 Set first frame of loop. Default is 0.
30311 Examples
30313 • Loop single first frame infinitely:
30315 loop=loop=-1:size=1:start=0
30317 • Loop single first frame 10 times:
30319 loop=loop=10:size=1:start=0
30321 • Loop 10 first frames 5 times:
30323 loop=loop=5:size=10:start=0
30325 lut1d
30327 Apply a 1D LUT to an input video.
30328 The filter accepts the following options:
30330 file
30332 Set the 1D LUT file name.
30333 Currently supported formats:
30335 cube
30337 Iridas
30338 csp cineSpace
30340 interp
30342 Select interpolation mode.
30343 Available values are:
30345 nearest
30347 Use values from the nearest defined point.
30348 linear
30350 Interpolate values using the linear interpolation.
30351 cosine
30353 Interpolate values using the cosine interpolation.
30354 cubic
30356 Interpolate values using the cubic interpolation.
30357 spline
30359 Interpolate values using the spline interpolation.
30360 Commands
30362 This filter supports the all above options as commands.
30364 lut3d
30366 Apply a 3D LUT to an input video.
30367 The filter accepts the following options:
30369 file
30371 Set the 3D LUT file name.
30372 Currently supported formats:
30374 3dl AfterEffects
30376 cube
30378 Iridas
30379 dat DaVinci
30381 m3d Pandora
30383 csp cineSpace
30385 interp
30387 Select interpolation mode.
30388 Available values are:
30390 nearest
30392 Use values from the nearest defined point.
30393 trilinear
30395 Interpolate values using the 8 points defining a cube.
30396 tetrahedral
30398 Interpolate values using a tetrahedron.
30399 pyramid
30401 Interpolate values using a pyramid.
30402 prism
30404 Interpolate values using a prism.
30405 Commands
30407 This filter supports the "interp" option as commands.
30409 lumakey
30411 Turn certain luma values into transparency.
30412 The filter accepts the following options:
30414 threshold
30416 Set the luma which will be used as base for transparency. Default
30417 value is 0.
30418 tolerance
30420 Set the range of luma values to be keyed out. Default value is
30421 0.01.
30422 softness
30424 Set the range of softness. Default value is 0. Use this to control
30425 gradual transition from zero to full transparency.
30426 Commands
30428 This filter supports same commands as options. The command accepts the
30430 same syntax of the corresponding option.
30431 If the specified expression is not valid, it is kept at its current
30433 value.
30434 lut, lutrgb, lutyuv
30436 Compute a look-up table for binding each pixel component input value to
30437 an output value, and apply it to the input video.
30438 lutyuv applies a lookup table to a YUV input video, lutrgb to an RGB
30440 input video.
30441 These filters accept the following parameters:
30443 c0 set first pixel component expression
30445 c1 set second pixel component expression
30447 c2 set third pixel component expression
30449 c3 set fourth pixel component expression, corresponds to the alpha
30451 component
30452 r set red component expression
30454 g set green component expression
30456 b set blue component expression
30458 a alpha component expression
30460 y set Y/luminance component expression
30462 u set U/Cb component expression
30464 v set V/Cr component expression
30466 Each of them specifies the expression to use for computing the lookup
30468 table for the corresponding pixel component values.
30469 The exact component associated to each of the c* options depends on the
30471 format in input.
30472 The lut filter requires either YUV or RGB pixel formats in input,
30474 lutrgb requires RGB pixel formats in input, and lutyuv requires YUV.
30475 The expressions can contain the following constants and functions:
30477 w
30479 h The input width and height.
30480 val The input value for the pixel component.
30482 clipval
30484 The input value, clipped to the minval-maxval range.
30485 maxval
30487 The maximum value for the pixel component.
30488 minval
30490 The minimum value for the pixel component.
30491 negval
30493 The negated value for the pixel component value, clipped to the
30494 minval-maxval range; it corresponds to the expression
30495 "maxval-clipval+minval".
30496 clip(val)
30498 The computed value in val, clipped to the minval-maxval range.
30499 gammaval(gamma)
30501 The computed gamma correction value of the pixel component value,
30502 clipped to the minval-maxval range. It corresponds to the
30503 expression
30504 "pow((clipval-minval)/(maxval-minval)\,gamma)*(maxval-minval)+minval"
30505 All expressions default to "val".
30507 Commands
30509 This filter supports same commands as options.
30511 Examples
30513 • Negate input video:
30515 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
30517 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
30518 The above is the same as:
30520 lutrgb="r=negval:g=negval:b=negval"
30522 lutyuv="y=negval:u=negval:v=negval"
30523 • Negate luminance:
30525 lutyuv=y=negval
30527 • Remove chroma components, turning the video into a graytone image:
30529 lutyuv="u=128:v=128"
30531 • Apply a luma burning effect:
30533 lutyuv="y=2*val"
30535 • Remove green and blue components:
30537 lutrgb="g=0:b=0"
30539 • Set a constant alpha channel value on input:
30541 format=rgba,lutrgb=a="maxval-minval/2"
30543 • Correct luminance gamma by a factor of 0.5:
30545 lutyuv=y=gammaval(0.5)
30547 • Discard least significant bits of luma:
30549 lutyuv=y='bitand(val, 128+64+32)'
30551 • Technicolor like effect:
30553 lutyuv=u='(val-maxval/2)*2+maxval/2':v='(val-maxval/2)*2+maxval/2'
30555 lut2, tlut2
30557 The "lut2" filter takes two input streams and outputs one stream.
30558 The "tlut2" (time lut2) filter takes two consecutive frames from one
30560 single stream.
30561 This filter accepts the following parameters:
30563 c0 set first pixel component expression
30565 c1 set second pixel component expression
30567 c2 set third pixel component expression
30569 c3 set fourth pixel component expression, corresponds to the alpha
30571 component
30572 d set output bit depth, only available for "lut2" filter. By default
30574 is 0, which means bit depth is automatically picked from first
30575 input format.
30576 The "lut2" filter also supports the framesync options.
30578 Each of them specifies the expression to use for computing the lookup
30580 table for the corresponding pixel component values.
30581 The exact component associated to each of the c* options depends on the
30583 format in inputs.
30584 The expressions can contain the following constants:
30586 w
30588 h The input width and height.
30589 x The first input value for the pixel component.
30591 y The second input value for the pixel component.
30593 bdx The first input video bit depth.
30595 bdy The second input video bit depth.
30597 All expressions default to "x".
30599 Commands
30601 This filter supports the all above options as commands except option
30603 "d".
30604 Examples
30606 • Highlight differences between two RGB video streams:
30608 lut2='ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,0,pow(2,bdx)-1)'
30610 • Highlight differences between two YUV video streams:
30612 lut2='ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,pow(2,bdx-1),pow(2,bdx)-1):ifnot(x-y,pow(2,bdx-1),pow(2,bdx)-1)'
30614 • Show max difference between two video streams:
30616 lut2='if(lt(x,y),0,if(gt(x,y),pow(2,bdx)-1,pow(2,bdx-1))):if(lt(x,y),0,if(gt(x,y),pow(2,bdx)-1,pow(2,bdx-1))):if(lt(x,y),0,if(gt(x,y),pow(2,bdx)-1,pow(2,bdx-1)))'
30618 maskedclamp
30620 Clamp the first input stream with the second input and third input
30621 stream.
30622 Returns the value of first stream to be between second input stream -
30624 "undershoot" and third input stream + "overshoot".
30625 This filter accepts the following options:
30627 undershoot
30629 Default value is 0.
30630 overshoot
30632 Default value is 0.
30633 planes
30635 Set which planes will be processed as bitmap, unprocessed planes
30636 will be copied from first stream. By default value 0xf, all planes
30637 will be processed.
30638 Commands
30640 This filter supports the all above options as commands.
30642 maskedmax
30644 Merge the second and third input stream into output stream using
30645 absolute differences between second input stream and first input stream
30646 and absolute difference between third input stream and first input
30647 stream. The picked value will be from second input stream if second
30648 absolute difference is greater than first one or from third input
30649 stream otherwise.
30650 This filter accepts the following options:
30652 planes
30654 Set which planes will be processed as bitmap, unprocessed planes
30655 will be copied from first stream. By default value 0xf, all planes
30656 will be processed.
30657 Commands
30659 This filter supports the all above options as commands.
30661 maskedmerge
30663 Merge the first input stream with the second input stream using per
30664 pixel weights in the third input stream.
30665 A value of 0 in the third stream pixel component means that pixel
30667 component from first stream is returned unchanged, while maximum value
30668 (eg. 255 for 8-bit videos) means that pixel component from second
30669 stream is returned unchanged. Intermediate values define the amount of
30670 merging between both input stream's pixel components.
30671 This filter accepts the following options:
30673 planes
30675 Set which planes will be processed as bitmap, unprocessed planes
30676 will be copied from first stream. By default value 0xf, all planes
30677 will be processed.
30678 Commands
30680 This filter supports the all above options as commands.
30682 maskedmin
30684 Merge the second and third input stream into output stream using
30685 absolute differences between second input stream and first input stream
30686 and absolute difference between third input stream and first input
30687 stream. The picked value will be from second input stream if second
30688 absolute difference is less than first one or from third input stream
30689 otherwise.
30690 This filter accepts the following options:
30692 planes
30694 Set which planes will be processed as bitmap, unprocessed planes
30695 will be copied from first stream. By default value 0xf, all planes
30696 will be processed.
30697 Commands
30699 This filter supports the all above options as commands.
30701 maskedthreshold
30703 Pick pixels comparing absolute difference of two video streams with
30704 fixed threshold.
30705 If absolute difference between pixel component of first and second
30707 video stream is equal or lower than user supplied threshold than pixel
30708 component from first video stream is picked, otherwise pixel component
30709 from second video stream is picked.
30710 This filter accepts the following options:
30712 threshold
30714 Set threshold used when picking pixels from absolute difference
30715 from two input video streams.
30716 planes
30718 Set which planes will be processed as bitmap, unprocessed planes
30719 will be copied from second stream. By default value 0xf, all
30720 planes will be processed.
30721 Commands
30723 This filter supports the all above options as commands.
30725 maskfun
30727 Create mask from input video.
30728 For example it is useful to create motion masks after "tblend" filter.
30730 This filter accepts the following options:
30732 low Set low threshold. Any pixel component lower or exact than this
30734 value will be set to 0.
30735 high
30737 Set high threshold. Any pixel component higher than this value will
30738 be set to max value allowed for current pixel format.
30739 planes
30741 Set planes to filter, by default all available planes are filtered.
30742 fill
30744 Fill all frame pixels with this value.
30745 sum Set max average pixel value for frame. If sum of all pixel
30747 components is higher that this average, output frame will be
30748 completely filled with value set by fill option. Typically useful
30749 for scene changes when used in combination with "tblend" filter.
30750 Commands
30752 This filter supports the all above options as commands.
30754 mcdeint
30756 Apply motion-compensation deinterlacing.
30757 It needs one field per frame as input and must thus be used together
30759 with yadif=1/3 or equivalent.
30760 This filter is only available in ffmpeg version 4.4 or earlier.
30762 This filter accepts the following options:
30764 mode
30766 Set the deinterlacing mode.
30767 It accepts one of the following values:
30769 fast
30771 medium
30772 slow
30773 use iterative motion estimation
30774 extra_slow
30776 like slow, but use multiple reference frames.
30777 Default value is fast.
30779 parity
30781 Set the picture field parity assumed for the input video. It must
30782 be one of the following values:
30783 0, tff
30785 assume top field first
30786 1, bff
30788 assume bottom field first
30789 Default value is bff.
30791 qp Set per-block quantization parameter (QP) used by the internal
30793 encoder.
30794 Higher values should result in a smoother motion vector field but
30796 less optimal individual vectors. Default value is 1.
30797 median
30799 Pick median pixel from certain rectangle defined by radius.
30800 This filter accepts the following options:
30802 radius
30804 Set horizontal radius size. Default value is 1. Allowed range is
30805 integer from 1 to 127.
30806 planes
30808 Set which planes to process. Default is 15, which is all available
30809 planes.
30810 radiusV
30812 Set vertical radius size. Default value is 0. Allowed range is
30813 integer from 0 to 127. If it is 0, value will be picked from
30814 horizontal "radius" option.
30815 percentile
30817 Set median percentile. Default value is 0.5. Default value of 0.5
30818 will pick always median values, while 0 will pick minimum values,
30819 and 1 maximum values.
30820 Commands
30822 This filter supports same commands as options. The command accepts the
30824 same syntax of the corresponding option.
30825 If the specified expression is not valid, it is kept at its current
30827 value.
30828 mergeplanes
30830 Merge color channel components from several video streams.
30831 The filter accepts up to 4 input streams, and merge selected input
30833 planes to the output video.
30834 This filter accepts the following options:
30836 mapping
30838 Set input to output plane mapping. Default is 0.
30839 The mappings is specified as a bitmap. It should be specified as a
30841 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
30842 mapping for the first plane of the output stream. 'A' sets the
30843 number of the input stream to use (from 0 to 3), and 'a' the plane
30844 number of the corresponding input to use (from 0 to 3). The rest of
30845 the mappings is similar, 'Bb' describes the mapping for the output
30846 stream second plane, 'Cc' describes the mapping for the output
30847 stream third plane and 'Dd' describes the mapping for the output
30848 stream fourth plane.
30849 format
30851 Set output pixel format. Default is "yuva444p".
30852 Examples
30854 • Merge three gray video streams of same width and height into single
30856 video stream:
30857 [a0][a1][a2]mergeplanes=0x001020:yuv444p
30859 • Merge 1st yuv444p stream and 2nd gray video stream into yuva444p
30861 video stream:
30862 [a0][a1]mergeplanes=0x00010210:yuva444p
30864 • Swap Y and A plane in yuva444p stream:
30866 format=yuva444p,mergeplanes=0x03010200:yuva444p
30868 • Swap U and V plane in yuv420p stream:
30870 format=yuv420p,mergeplanes=0x000201:yuv420p
30872 • Cast a rgb24 clip to yuv444p:
30874 format=rgb24,mergeplanes=0x000102:yuv444p
30876 mestimate
30878 Estimate and export motion vectors using block matching algorithms.
30879 Motion vectors are stored in frame side data to be used by other
30880 filters.
30881 This filter accepts the following options:
30883 method
30885 Specify the motion estimation method. Accepts one of the following
30886 values:
30887 esa Exhaustive search algorithm.
30889 tss Three step search algorithm.
30891 tdls
30893 Two dimensional logarithmic search algorithm.
30894 ntss
30896 New three step search algorithm.
30897 fss Four step search algorithm.
30899 ds Diamond search algorithm.
30901 hexbs
30903 Hexagon-based search algorithm.
30904 epzs
30906 Enhanced predictive zonal search algorithm.
30907 umh Uneven multi-hexagon search algorithm.
30909 Default value is esa.
30911 mb_size
30913 Macroblock size. Default 16.
30914 search_param
30916 Search parameter. Default 7.
30917 midequalizer
30919 Apply Midway Image Equalization effect using two video streams.
30920 Midway Image Equalization adjusts a pair of images to have the same
30922 histogram, while maintaining their dynamics as much as possible. It's
30923 useful for e.g. matching exposures from a pair of stereo cameras.
30924 This filter has two inputs and one output, which must be of same pixel
30926 format, but may be of different sizes. The output of filter is first
30927 input adjusted with midway histogram of both inputs.
30928 This filter accepts the following option:
30930 planes
30932 Set which planes to process. Default is 15, which is all available
30933 planes.
30934 minterpolate
30936 Convert the video to specified frame rate using motion interpolation.
30937 This filter accepts the following options:
30939 fps Specify the output frame rate. This can be rational e.g.
30941 "60000/1001". Frames are dropped if fps is lower than source fps.
30942 Default 60.
30943 mi_mode
30945 Motion interpolation mode. Following values are accepted:
30946 dup Duplicate previous or next frame for interpolating new ones.
30948 blend
30950 Blend source frames. Interpolated frame is mean of previous and
30951 next frames.
30952 mci Motion compensated interpolation. Following options are
30954 effective when this mode is selected:
30955 mc_mode
30957 Motion compensation mode. Following values are accepted:
30958 obmc
30960 Overlapped block motion compensation.
30961 aobmc
30963 Adaptive overlapped block motion compensation. Window
30964 weighting coefficients are controlled adaptively
30965 according to the reliabilities of the neighboring
30966 motion vectors to reduce oversmoothing.
30967 Default mode is obmc.
30969 me_mode
30971 Motion estimation mode. Following values are accepted:
30972 bidir
30974 Bidirectional motion estimation. Motion vectors are
30975 estimated for each source frame in both forward and
30976 backward directions.
30977 bilat
30979 Bilateral motion estimation. Motion vectors are
30980 estimated directly for interpolated frame.
30981 Default mode is bilat.
30983 me The algorithm to be used for motion estimation. Following
30985 values are accepted:
30986 esa Exhaustive search algorithm.
30988 tss Three step search algorithm.
30990 tdls
30992 Two dimensional logarithmic search algorithm.
30993 ntss
30995 New three step search algorithm.
30996 fss Four step search algorithm.
30998 ds Diamond search algorithm.
31000 hexbs
31002 Hexagon-based search algorithm.
31003 epzs
31005 Enhanced predictive zonal search algorithm.
31006 umh Uneven multi-hexagon search algorithm.
31008 Default algorithm is epzs.
31010 mb_size
31012 Macroblock size. Default 16.
31013 search_param
31015 Motion estimation search parameter. Default 32.
31016 vsbmc
31018 Enable variable-size block motion compensation. Motion
31019 estimation is applied with smaller block sizes at object
31020 boundaries in order to make the them less blur. Default is
31021 0 (disabled).
31022 scd Scene change detection method. Scene change leads motion vectors to
31024 be in random direction. Scene change detection replace interpolated
31025 frames by duplicate ones. May not be needed for other modes.
31026 Following values are accepted:
31027 none
31029 Disable scene change detection.
31030 fdiff
31032 Frame difference. Corresponding pixel values are compared and
31033 if it satisfies scd_threshold scene change is detected.
31034 Default method is fdiff.
31036 scd_threshold
31038 Scene change detection threshold. Default is 10..
31039 mix
31041 Mix several video input streams into one video stream.
31042 A description of the accepted options follows.
31044 inputs
31046 The number of inputs. If unspecified, it defaults to 2.
31047 weights
31049 Specify weight of each input video stream as sequence. Each weight
31050 is separated by space. If number of weights is smaller than number
31051 of frames last specified weight will be used for all remaining
31052 unset weights.
31053 scale
31055 Specify scale, if it is set it will be multiplied with sum of each
31056 weight multiplied with pixel values to give final destination pixel
31057 value. By default scale is auto scaled to sum of weights.
31058 duration
31060 Specify how end of stream is determined.
31061 longest
31063 The duration of the longest input. (default)
31064 shortest
31066 The duration of the shortest input.
31067 first
31069 The duration of the first input.
31070 Commands
31072 This filter supports the following commands:
31074 weights
31076 scale
31077 Syntax is same as option with same name.
31078 monochrome
31080 Convert video to gray using custom color filter.
31081 A description of the accepted options follows.
31083 cb Set the chroma blue spot. Allowed range is from -1 to 1. Default
31085 value is 0.
31086 cr Set the chroma red spot. Allowed range is from -1 to 1. Default
31088 value is 0.
31089 size
31091 Set the color filter size. Allowed range is from .1 to 10. Default
31092 value is 1.
31093 high
31095 Set the highlights strength. Allowed range is from 0 to 1. Default
31096 value is 0.
31097 Commands
31099 This filter supports the all above options as commands.
31101 morpho
31103 This filter allows to apply main morphological grayscale transforms,
31104 erode and dilate with arbitrary structures set in second input stream.
31105 Unlike naive implementation and much slower performance in erosion and
31107 dilation filters, when speed is critical "morpho" filter should be used
31108 instead.
31109 A description of accepted options follows,
31111 mode
31113 Set morphological transform to apply, can be:
31114 erode
31116 dilate
31117 open
31118 close
31119 gradient
31120 tophat
31121 blackhat
31122 Default is "erode".
31124 planes
31126 Set planes to filter, by default all planes except alpha are
31127 filtered.
31128 structure
31130 Set which structure video frames will be processed from second
31131 input stream, can be first or all. Default is all.
31132 The "morpho" filter also supports the framesync options.
31134 Commands
31136 This filter supports same commands as options.
31138 mpdecimate
31140 Drop frames that do not differ greatly from the previous frame in order
31141 to reduce frame rate.
31142 The main use of this filter is for very-low-bitrate encoding (e.g.
31144 streaming over dialup modem), but it could in theory be used for fixing
31145 movies that were inverse-telecined incorrectly.
31146 A description of the accepted options follows.
31148 max Set the maximum number of consecutive frames which can be dropped
31150 (if positive), or the minimum interval between dropped frames (if
31151 negative). If the value is 0, the frame is dropped disregarding the
31152 number of previous sequentially dropped frames.
31153 Default value is 0.
31155 hi
31157 lo
31158 frac
31159 Set the dropping threshold values.
31160 Values for hi and lo are for 8x8 pixel blocks and represent actual
31162 pixel value differences, so a threshold of 64 corresponds to 1 unit
31163 of difference for each pixel, or the same spread out differently
31164 over the block.
31165 A frame is a candidate for dropping if no 8x8 blocks differ by more
31167 than a threshold of hi, and if no more than frac blocks (1 meaning
31168 the whole image) differ by more than a threshold of lo.
31169 Default value for hi is 64*12, default value for lo is 64*5, and
31171 default value for frac is 0.33.
31172 msad
31174 Obtain the MSAD (Mean Sum of Absolute Differences) between two input
31175 videos.
31176 This filter takes two input videos.
31178 Both input videos must have the same resolution and pixel format for
31180 this filter to work correctly. Also it assumes that both inputs have
31181 the same number of frames, which are compared one by one.
31182 The obtained per component, average, min and max MSAD is printed
31184 through the logging system.
31185 The filter stores the calculated MSAD of each frame in frame metadata.
31187 In the below example the input file main.mpg being processed is
31189 compared with the reference file ref.mpg.
31190 ffmpeg -i main.mpg -i ref.mpg -lavfi msad -f null -
31192 negate
31194 Negate (invert) the input video.
31195 It accepts the following option:
31197 components
31199 Set components to negate.
31200 Available values for components are:
31202 y
31204 u
31205 v
31206 a
31207 r
31208 g
31209 b
31210 negate_alpha
31211 With value 1, it negates the alpha component, if present. Default
31212 value is 0.
31213 Commands
31215 This filter supports same commands as options.
31217 nlmeans
31219 Denoise frames using Non-Local Means algorithm.
31220 Each pixel is adjusted by looking for other pixels with similar
31222 contexts. This context similarity is defined by comparing their
31223 surrounding patches of size pxp. Patches are searched in an area of rxr
31224 around the pixel.
31225 Note that the research area defines centers for patches, which means
31227 some patches will be made of pixels outside that research area.
31228 The filter accepts the following options.
31230 s Set denoising strength. Default is 1.0. Must be in range [1.0,
31232 30.0].
31233 p Set patch size. Default is 7. Must be odd number in range [0, 99].
31235 pc Same as p but for chroma planes.
31237 The default value is 0 and means automatic.
31239 r Set research size. Default is 15. Must be odd number in range [0,
31241 99].
31242 rc Same as r but for chroma planes.
31244 The default value is 0 and means automatic.
31246 nnedi
31248 Deinterlace video using neural network edge directed interpolation.
31249 This filter accepts the following options:
31251 weights
31253 Mandatory option, without binary file filter can not work.
31254 Currently file can be found here:
31255 https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
31256 deint
31258 Set which frames to deinterlace, by default it is "all". Can be
31259 "all" or "interlaced".
31260 field
31262 Set mode of operation.
31263 Can be one of the following:
31265 af Use frame flags, both fields.
31267 a Use frame flags, single field.
31269 t Use top field only.
31271 b Use bottom field only.
31273 tf Use both fields, top first.
31275 bf Use both fields, bottom first.
31277 planes
31279 Set which planes to process, by default filter process all frames.
31280 nsize
31282 Set size of local neighborhood around each pixel, used by the
31283 predictor neural network.
31284 Can be one of the following:
31286 s8x6
31288 s16x6
31289 s32x6
31290 s48x6
31291 s8x4
31292 s16x4
31293 s32x4
31294 nns Set the number of neurons in predictor neural network. Can be one
31295 of the following:
31296 n16
31298 n32
31299 n64
31300 n128
31301 n256
31302 qual
31303 Controls the number of different neural network predictions that
31304 are blended together to compute the final output value. Can be
31305 "fast", default or "slow".
31306 etype
31308 Set which set of weights to use in the predictor. Can be one of
31309 the following:
31310 a, abs
31312 weights trained to minimize absolute error
31313 s, mse
31315 weights trained to minimize squared error
31316 pscrn
31318 Controls whether or not the prescreener neural network is used to
31319 decide which pixels should be processed by the predictor neural
31320 network and which can be handled by simple cubic interpolation.
31321 The prescreener is trained to know whether cubic interpolation will
31322 be sufficient for a pixel or whether it should be predicted by the
31323 predictor nn. The computational complexity of the prescreener nn
31324 is much less than that of the predictor nn. Since most pixels can
31325 be handled by cubic interpolation, using the prescreener generally
31326 results in much faster processing. The prescreener is pretty
31327 accurate, so the difference between using it and not using it is
31328 almost always unnoticeable.
31329 Can be one of the following:
31331 none
31333 original
31334 new
31335 new2
31336 new3
31337 Default is "new".
31339 Commands
31341 This filter supports same commands as options, excluding weights
31343 option.
31344 noformat
31346 Force libavfilter not to use any of the specified pixel formats for the
31347 input to the next filter.
31348 It accepts the following parameters:
31350 pix_fmts
31352 A '|'-separated list of pixel format names, such as
31353 pix_fmts=yuv420p|monow|rgb24".
31354 Examples
31356 • Force libavfilter to use a format different from yuv420p for the
31358 input to the vflip filter:
31359 noformat=pix_fmts=yuv420p,vflip
31361 • Convert the input video to any of the formats not contained in the
31363 list:
31364 noformat=yuv420p|yuv444p|yuv410p
31366 noise
31368 Add noise on video input frame.
31369 The filter accepts the following options:
31371 all_seed
31373 c0_seed
31374 c1_seed
31375 c2_seed
31376 c3_seed
31377 Set noise seed for specific pixel component or all pixel components
31378 in case of all_seed. Default value is 123457.
31379 all_strength, alls
31381 c0_strength, c0s
31382 c1_strength, c1s
31383 c2_strength, c2s
31384 c3_strength, c3s
31385 Set noise strength for specific pixel component or all pixel
31386 components in case all_strength. Default value is 0. Allowed range
31387 is [0, 100].
31388 all_flags, allf
31390 c0_flags, c0f
31391 c1_flags, c1f
31392 c2_flags, c2f
31393 c3_flags, c3f
31394 Set pixel component flags or set flags for all components if
31395 all_flags. Available values for component flags are:
31396 a averaged temporal noise (smoother)
31398 p mix random noise with a (semi)regular pattern
31400 t temporal noise (noise pattern changes between frames)
31402 u uniform noise (gaussian otherwise)
31404 Examples
31406 Add temporal and uniform noise to input video:
31408 noise=alls=20:allf=t+u
31410 normalize
31412 Normalize RGB video (aka histogram stretching, contrast stretching).
31413 See: https://en.wikipedia.org/wiki/Normalization_(image_processing)
31414 For each channel of each frame, the filter computes the input range and
31416 maps it linearly to the user-specified output range. The output range
31417 defaults to the full dynamic range from pure black to pure white.
31418 Temporal smoothing can be used on the input range to reduce flickering
31420 (rapid changes in brightness) caused when small dark or bright objects
31421 enter or leave the scene. This is similar to the auto-exposure
31422 (automatic gain control) on a video camera, and, like a video camera,
31423 it may cause a period of over- or under-exposure of the video.
31424 The R,G,B channels can be normalized independently, which may cause
31426 some color shifting, or linked together as a single channel, which
31427 prevents color shifting. Linked normalization preserves hue.
31428 Independent normalization does not, so it can be used to remove some
31429 color casts. Independent and linked normalization can be combined in
31430 any ratio.
31431 The normalize filter accepts the following options:
31433 blackpt
31435 whitept
31436 Colors which define the output range. The minimum input value is
31437 mapped to the blackpt. The maximum input value is mapped to the
31438 whitept. The defaults are black and white respectively. Specifying
31439 white for blackpt and black for whitept will give color-inverted,
31440 normalized video. Shades of grey can be used to reduce the dynamic
31441 range (contrast). Specifying saturated colors here can create some
31442 interesting effects.
31443 smoothing
31445 The number of previous frames to use for temporal smoothing. The
31446 input range of each channel is smoothed using a rolling average
31447 over the current frame and the smoothing previous frames. The
31448 default is 0 (no temporal smoothing).
31449 independence
31451 Controls the ratio of independent (color shifting) channel
31452 normalization to linked (color preserving) normalization. 0.0 is
31453 fully linked, 1.0 is fully independent. Defaults to 1.0 (fully
31454 independent).
31455 strength
31457 Overall strength of the filter. 1.0 is full strength. 0.0 is a
31458 rather expensive no-op. Defaults to 1.0 (full strength).
31459 Commands
31461 This filter supports same commands as options, excluding smoothing
31463 option. The command accepts the same syntax of the corresponding
31464 option.
31465 If the specified expression is not valid, it is kept at its current
31467 value.
31468 Examples
31470 Stretch video contrast to use the full dynamic range, with no temporal
31472 smoothing; may flicker depending on the source content:
31473 normalize=blackpt=black:whitept=white:smoothing=0
31475 As above, but with 50 frames of temporal smoothing; flicker should be
31477 reduced, depending on the source content:
31478 normalize=blackpt=black:whitept=white:smoothing=50
31480 As above, but with hue-preserving linked channel normalization:
31482 normalize=blackpt=black:whitept=white:smoothing=50:independence=0
31484 As above, but with half strength:
31486 normalize=blackpt=black:whitept=white:smoothing=50:independence=0:strength=0.5
31488 Map the darkest input color to red, the brightest input color to cyan:
31490 normalize=blackpt=red:whitept=cyan
31492 null
31494 Pass the video source unchanged to the output.
31495 ocr
31497 Optical Character Recognition
31498 This filter uses Tesseract for optical character recognition. To enable
31500 compilation of this filter, you need to configure FFmpeg with
31501 "--enable-libtesseract".
31502 It accepts the following options:
31504 datapath
31506 Set datapath to tesseract data. Default is to use whatever was set
31507 at installation.
31508 language
31510 Set language, default is "eng".
31511 whitelist
31513 Set character whitelist.
31514 blacklist
31516 Set character blacklist.
31517 The filter exports recognized text as the frame metadata
31519 "lavfi.ocr.text". The filter exports confidence of recognized words as
31520 the frame metadata "lavfi.ocr.confidence".
31521 ocv
31523 Apply a video transform using libopencv.
31524 To enable this filter, install the libopencv library and headers and
31526 configure FFmpeg with "--enable-libopencv".
31527 It accepts the following parameters:
31529 filter_name
31531 The name of the libopencv filter to apply.
31532 filter_params
31534 The parameters to pass to the libopencv filter. If not specified,
31535 the default values are assumed.
31536 Refer to the official libopencv documentation for more precise
31538 information:
31539 <http://docs.opencv.org/master/modules/imgproc/doc/filtering.html>
31540 Several libopencv filters are supported; see the following subsections.
31542 dilate
31544 Dilate an image by using a specific structuring element. It
31546 corresponds to the libopencv function "cvDilate".
31547 It accepts the parameters: struct_el|nb_iterations.
31549 struct_el represents a structuring element, and has the syntax:
31551 colsxrows+anchor_xxanchor_y/shape
31552 cols and rows represent the number of columns and rows of the
31554 structuring element, anchor_x and anchor_y the anchor point, and shape
31555 the shape for the structuring element. shape must be "rect", "cross",
31556 "ellipse", or "custom".
31557 If the value for shape is "custom", it must be followed by a string of
31559 the form "=filename". The file with name filename is assumed to
31560 represent a binary image, with each printable character corresponding
31561 to a bright pixel. When a custom shape is used, cols and rows are
31562 ignored, the number or columns and rows of the read file are assumed
31563 instead.
31564 The default value for struct_el is "3x3+0x0/rect".
31566 nb_iterations specifies the number of times the transform is applied to
31568 the image, and defaults to 1.
31569 Some examples:
31571 # Use the default values
31573 ocv=dilate
31574 # Dilate using a structuring element with a 5x5 cross, iterating two times
31576 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
31577 # Read the shape from the file diamond.shape, iterating two times.
31579 # The file diamond.shape may contain a pattern of characters like this
31580 # *
31581 # ***
31582 # *****
31583 # ***
31584 # *
31585 # The specified columns and rows are ignored
31586 # but the anchor point coordinates are not
31587 ocv=dilate:0x0+2x2/custom=diamond.shape|2
31588 erode
31590 Erode an image by using a specific structuring element. It corresponds
31592 to the libopencv function "cvErode".
31593 It accepts the parameters: struct_el:nb_iterations, with the same
31595 syntax and semantics as the dilate filter.
31596 smooth
31598 Smooth the input video.
31600 The filter takes the following parameters:
31602 type|param1|param2|param3|param4.
31603 type is the type of smooth filter to apply, and must be one of the
31605 following values: "blur", "blur_no_scale", "median", "gaussian", or
31606 "bilateral". The default value is "gaussian".
31607 The meaning of param1, param2, param3, and param4 depends on the smooth
31609 type. param1 and param2 accept integer positive values or 0. param3 and
31610 param4 accept floating point values.
31611 The default value for param1 is 3. The default value for the other
31613 parameters is 0.
31614 These parameters correspond to the parameters assigned to the libopencv
31616 function "cvSmooth".
31617 oscilloscope
31619 2D Video Oscilloscope.
31620 Useful to measure spatial impulse, step responses, chroma delays, etc.
31622 It accepts the following parameters:
31624 x Set scope center x position.
31626 y Set scope center y position.
31628 s Set scope size, relative to frame diagonal.
31630 t Set scope tilt/rotation.
31632 o Set trace opacity.
31634 tx Set trace center x position.
31636 ty Set trace center y position.
31638 tw Set trace width, relative to width of frame.
31640 th Set trace height, relative to height of frame.
31642 c Set which components to trace. By default it traces first three
31644 components.
31645 g Draw trace grid. By default is enabled.
31647 st Draw some statistics. By default is enabled.
31649 sc Draw scope. By default is enabled.
31651 Commands
31653 This filter supports same commands as options. The command accepts the
31655 same syntax of the corresponding option.
31656 If the specified expression is not valid, it is kept at its current
31658 value.
31659 Examples
31661 • Inspect full first row of video frame.
31663 oscilloscope=x=0.5:y=0:s=1
31665 • Inspect full last row of video frame.
31667 oscilloscope=x=0.5:y=1:s=1
31669 • Inspect full 5th line of video frame of height 1080.
31671 oscilloscope=x=0.5:y=5/1080:s=1
31673 • Inspect full last column of video frame.
31675 oscilloscope=x=1:y=0.5:s=1:t=1
31677 overlay
31679 Overlay one video on top of another.
31680 It takes two inputs and has one output. The first input is the "main"
31682 video on which the second input is overlaid.
31683 It accepts the following parameters:
31685 A description of the accepted options follows.
31687 x
31689 y Set the expression for the x and y coordinates of the overlaid
31690 video on the main video. Default value is "0" for both expressions.
31691 In case the expression is invalid, it is set to a huge value
31692 (meaning that the overlay will not be displayed within the output
31693 visible area).
31694 eof_action
31696 See framesync.
31697 eval
31699 Set when the expressions for x, and y are evaluated.
31700 It accepts the following values:
31702 init
31704 only evaluate expressions once during the filter initialization
31705 or when a command is processed
31706 frame
31708 evaluate expressions for each incoming frame
31709 Default value is frame.
31711 shortest
31713 See framesync.
31714 format
31716 Set the format for the output video.
31717 It accepts the following values:
31719 yuv420
31721 force YUV420 output
31722 yuv420p10
31724 force YUV420p10 output
31725 yuv422
31727 force YUV422 output
31728 yuv422p10
31730 force YUV422p10 output
31731 yuv444
31733 force YUV444 output
31734 rgb force packed RGB output
31736 gbrp
31738 force planar RGB output
31739 auto
31741 automatically pick format
31742 Default value is yuv420.
31744 repeatlast
31746 See framesync.
31747 alpha
31749 Set format of alpha of the overlaid video, it can be straight or
31750 premultiplied. Default is straight.
31751 The x, and y expressions can contain the following parameters.
31753 main_w, W
31755 main_h, H
31756 The main input width and height.
31757 overlay_w, w
31759 overlay_h, h
31760 The overlay input width and height.
31761 x
31763 y The computed values for x and y. They are evaluated for each new
31764 frame.
31765 hsub
31767 vsub
31768 horizontal and vertical chroma subsample values of the output
31769 format. For example for the pixel format "yuv422p" hsub is 2 and
31770 vsub is 1.
31771 n the number of input frame, starting from 0
31773 pos the position in the file of the input frame, NAN if unknown
31775 t The timestamp, expressed in seconds. It's NAN if the input
31777 timestamp is unknown.
31778 This filter also supports the framesync options.
31780 Note that the n, pos, t variables are available only when evaluation is
31782 done per frame, and will evaluate to NAN when eval is set to init.
31783 Be aware that frames are taken from each input video in timestamp
31785 order, hence, if their initial timestamps differ, it is a good idea to
31786 pass the two inputs through a setpts=PTS-STARTPTS filter to have them
31787 begin in the same zero timestamp, as the example for the movie filter
31788 does.
31789 You can chain together more overlays but you should test the efficiency
31791 of such approach.
31792 Commands
31794 This filter supports the following commands:
31796 x
31798 y Modify the x and y of the overlay input. The command accepts the
31799 same syntax of the corresponding option.
31800 If the specified expression is not valid, it is kept at its current
31802 value.
31803 Examples
31805 • Draw the overlay at 10 pixels from the bottom right corner of the
31807 main video:
31808 overlay=main_w-overlay_w-10:main_h-overlay_h-10
31810 Using named options the example above becomes:
31812 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
31814 • Insert a transparent PNG logo in the bottom left corner of the
31816 input, using the ffmpeg tool with the "-filter_complex" option:
31817 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
31819 • Insert 2 different transparent PNG logos (second logo on bottom
31821 right corner) using the ffmpeg tool:
31822 ffmpeg -i input -i logo1 -i logo2 -filter_complex 'overlay=x=10:y=H-h-10,overlay=x=W-w-10:y=H-h-10' output
31824 • Add a transparent color layer on top of the main video; "WxH" must
31826 specify the size of the main input to the overlay filter:
31827 color=color=red@.3:size=WxH [over]; [in][over] overlay [out]
31829 • Play an original video and a filtered version (here with the
31831 deshake filter) side by side using the ffplay tool:
31832 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
31834 The above command is the same as:
31836 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
31838 • Make a sliding overlay appearing from the left to the right top
31840 part of the screen starting since time 2:
31841 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
31843 • Compose output by putting two input videos side to side:
31845 ffmpeg -i left.avi -i right.avi -filter_complex "
31847 nullsrc=size=200x100 [background];
31848 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
31849 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
31850 [background][left] overlay=shortest=1 [background+left];
31851 [background+left][right] overlay=shortest=1:x=100 [left+right]
31852 "
31853 • Mask 10-20 seconds of a video by applying the delogo filter to a
31855 section
31856 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
31858 -vf '[in]split[split_main][split_delogo];[split_delogo]trim=start=360:end=371,delogo=0:0:640:480[delogoed];[split_main][delogoed]overlay=eof_action=pass[out]'
31859 masked.avi
31860 • Chain several overlays in cascade:
31862 nullsrc=s=200x200 [bg];
31864 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
31865 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
31866 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
31867 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
31868 [in3] null, [mid2] overlay=100:100 [out0]
31869 overlay_cuda
31871 Overlay one video on top of another.
31872 This is the CUDA variant of the overlay filter. It only accepts CUDA
31874 frames. The underlying input pixel formats have to match.
31875 It takes two inputs and has one output. The first input is the "main"
31877 video on which the second input is overlaid.
31878 It accepts the following parameters:
31880 x
31882 y Set expressions for the x and y coordinates of the overlaid video
31883 on the main video.
31884 They can contain the following parameters:
31886 main_w, W
31888 main_h, H
31889 The main input width and height.
31890 overlay_w, w
31892 overlay_h, h
31893 The overlay input width and height.
31894 x
31896 y The computed values for x and y. They are evaluated for each
31897 new frame.
31898 n The ordinal index of the main input frame, starting from 0.
31900 pos The byte offset position in the file of the main input frame,
31902 NAN if unknown.
31903 t The timestamp of the main input frame, expressed in seconds,
31905 NAN if unknown.
31906 Default value is "0" for both expressions.
31908 eval
31910 Set when the expressions for x and y are evaluated.
31911 It accepts the following values:
31913 init
31915 Evaluate expressions once during filter initialization or when
31916 a command is processed.
31917 frame
31919 Evaluate expressions for each incoming frame
31920 Default value is frame.
31922 eof_action
31924 See framesync.
31925 shortest
31927 See framesync.
31928 repeatlast
31930 See framesync.
31931 This filter also supports the framesync options.
31933 owdenoise
31935 Apply Overcomplete Wavelet denoiser.
31936 The filter accepts the following options:
31938 depth
31940 Set depth.
31941 Larger depth values will denoise lower frequency components more,
31943 but slow down filtering.
31944 Must be an int in the range 8-16, default is 8.
31946 luma_strength, ls
31948 Set luma strength.
31949 Must be a double value in the range 0-1000, default is 1.0.
31951 chroma_strength, cs
31953 Set chroma strength.
31954 Must be a double value in the range 0-1000, default is 1.0.
31956 pad
31958 Add paddings to the input image, and place the original input at the
31959 provided x, y coordinates.
31960 It accepts the following parameters:
31962 width, w
31964 height, h
31965 Specify an expression for the size of the output image with the
31966 paddings added. If the value for width or height is 0, the
31967 corresponding input size is used for the output.
31968 The width expression can reference the value set by the height
31970 expression, and vice versa.
31971 The default value of width and height is 0.
31973 x
31975 y Specify the offsets to place the input image at within the padded
31976 area, with respect to the top/left border of the output image.
31977 The x expression can reference the value set by the y expression,
31979 and vice versa.
31980 The default value of x and y is 0.
31982 If x or y evaluate to a negative number, they'll be changed so the
31984 input image is centered on the padded area.
31985 color
31987 Specify the color of the padded area. For the syntax of this
31988 option, check the "Color" section in the ffmpeg-utils manual.
31989 The default value of color is "black".
31991 eval
31993 Specify when to evaluate width, height, x and y expression.
31994 It accepts the following values:
31996 init
31998 Only evaluate expressions once during the filter initialization
31999 or when a command is processed.
32000 frame
32002 Evaluate expressions for each incoming frame.
32003 Default value is init.
32005 aspect
32007 Pad to aspect instead to a resolution.
32008 The value for the width, height, x, and y options are expressions
32010 containing the following constants:
32011 in_w
32013 in_h
32014 The input video width and height.
32015 iw
32017 ih These are the same as in_w and in_h.
32018 out_w
32020 out_h
32021 The output width and height (the size of the padded area), as
32022 specified by the width and height expressions.
32023 ow
32025 oh These are the same as out_w and out_h.
32026 x
32028 y The x and y offsets as specified by the x and y expressions, or NAN
32029 if not yet specified.
32030 a same as iw / ih
32032 sar input sample aspect ratio
32034 dar input display aspect ratio, it is the same as (iw / ih) * sar
32036 hsub
32038 vsub
32039 The horizontal and vertical chroma subsample values. For example
32040 for the pixel format "yuv422p" hsub is 2 and vsub is 1.
32041 Examples
32043 • Add paddings with the color "violet" to the input video. The output
32045 video size is 640x480, and the top-left corner of the input video
32046 is placed at column 0, row 40
32047 pad=640:480:0:40:violet
32049 The example above is equivalent to the following command:
32051 pad=width=640:height=480:x=0:y=40:color=violet
32053 • Pad the input to get an output with dimensions increased by 3/2,
32055 and put the input video at the center of the padded area:
32056 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
32058 • Pad the input to get a squared output with size equal to the
32060 maximum value between the input width and height, and put the input
32061 video at the center of the padded area:
32062 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
32064 • Pad the input to get a final w/h ratio of 16:9:
32066 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
32068 • In case of anamorphic video, in order to set the output display
32070 aspect correctly, it is necessary to use sar in the expression,
32071 according to the relation:
32072 (ih * X / ih) * sar = output_dar
32074 X = output_dar / sar
32075 Thus the previous example needs to be modified to:
32077 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
32079 • Double the output size and put the input video in the bottom-right
32081 corner of the output padded area:
32082 pad="2*iw:2*ih:ow-iw:oh-ih"
32084 palettegen
32086 Generate one palette for a whole video stream.
32087 It accepts the following options:
32089 max_colors
32091 Set the maximum number of colors to quantize in the palette. Note:
32092 the palette will still contain 256 colors; the unused palette
32093 entries will be black.
32094 reserve_transparent
32096 Create a palette of 255 colors maximum and reserve the last one for
32097 transparency. Reserving the transparency color is useful for GIF
32098 optimization. If not set, the maximum of colors in the palette
32099 will be 256. You probably want to disable this option for a
32100 standalone image. Set by default.
32101 transparency_color
32103 Set the color that will be used as background for transparency.
32104 stats_mode
32106 Set statistics mode.
32107 It accepts the following values:
32109 full
32111 Compute full frame histograms.
32112 diff
32114 Compute histograms only for the part that differs from previous
32115 frame. This might be relevant to give more importance to the
32116 moving part of your input if the background is static.
32117 single
32119 Compute new histogram for each frame.
32120 Default value is full.
32122 use_alpha
32124 Create a palette of colors with alpha components. Setting this,
32125 will automatically disable 'reserve_transparent'.
32126 The filter also exports the frame metadata "lavfi.color_quant_ratio"
32128 ("nb_color_in / nb_color_out") which you can use to evaluate the degree
32129 of color quantization of the palette. This information is also visible
32130 at info logging level.
32131 Examples
32133 • Generate a representative palette of a given video using ffmpeg:
32135 ffmpeg -i input.mkv -vf palettegen palette.png
32137 paletteuse
32139 Use a palette to downsample an input video stream.
32140 The filter takes two inputs: one video stream and a palette. The
32142 palette must be a 256 pixels image.
32143 It accepts the following options:
32145 dither
32147 Select dithering mode. Available algorithms are:
32148 bayer
32150 Ordered 8x8 bayer dithering (deterministic)
32151 heckbert
32153 Dithering as defined by Paul Heckbert in 1982 (simple error
32154 diffusion). Note: this dithering is sometimes considered
32155 "wrong" and is included as a reference.
32156 floyd_steinberg
32158 Floyd and Steingberg dithering (error diffusion)
32159 sierra2
32161 Frankie Sierra dithering v2 (error diffusion)
32162 sierra2_4a
32164 Frankie Sierra dithering v2 "Lite" (error diffusion)
32165 Default is sierra2_4a.
32167 bayer_scale
32169 When bayer dithering is selected, this option defines the scale of
32170 the pattern (how much the crosshatch pattern is visible). A low
32171 value means more visible pattern for less banding, and higher value
32172 means less visible pattern at the cost of more banding.
32173 The option must be an integer value in the range [0,5]. Default is
32175 2.
32176 diff_mode
32178 If set, define the zone to process
32179 rectangle
32181 Only the changing rectangle will be reprocessed. This is
32182 similar to GIF cropping/offsetting compression mechanism. This
32183 option can be useful for speed if only a part of the image is
32184 changing, and has use cases such as limiting the scope of the
32185 error diffusal dither to the rectangle that bounds the moving
32186 scene (it leads to more deterministic output if the scene
32187 doesn't change much, and as a result less moving noise and
32188 better GIF compression).
32189 Default is none.
32191 new Take new palette for each output frame.
32193 alpha_threshold
32195 Sets the alpha threshold for transparency. Alpha values above this
32196 threshold will be treated as completely opaque, and values below
32197 this threshold will be treated as completely transparent.
32198 The option must be an integer value in the range [0,255]. Default
32200 is 128.
32201 use_alpha
32203 Apply the palette by taking alpha values into account. Only useful
32204 with palettes that are containing multiple colors with alpha
32205 components. Setting this will automatically disable
32206 'alpha_treshold'.
32207 Examples
32209 • Use a palette (generated for example with palettegen) to encode a
32211 GIF using ffmpeg:
32212 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
32214 perspective
32216 Correct perspective of video not recorded perpendicular to the screen.
32217 A description of the accepted parameters follows.
32219 x0
32221 y0
32222 x1
32223 y1
32224 x2
32225 y2
32226 x3
32227 y3 Set coordinates expression for top left, top right, bottom left and
32228 bottom right corners. Default values are "0:0:W:0:0:H:W:H" with
32229 which perspective will remain unchanged. If the "sense" option is
32230 set to "source", then the specified points will be sent to the
32231 corners of the destination. If the "sense" option is set to
32232 "destination", then the corners of the source will be sent to the
32233 specified coordinates.
32234 The expressions can use the following variables:
32236 W
32238 H the width and height of video frame.
32239 in Input frame count.
32241 on Output frame count.
32243 interpolation
32245 Set interpolation for perspective correction.
32246 It accepts the following values:
32248 linear
32250 cubic
32251 Default value is linear.
32253 sense
32255 Set interpretation of coordinate options.
32256 It accepts the following values:
32258 0, source
32260 Send point in the source specified by the given coordinates to
32261 the corners of the destination.
32262 1, destination
32264 Send the corners of the source to the point in the destination
32265 specified by the given coordinates.
32266 Default value is source.
32268 eval
32270 Set when the expressions for coordinates x0,y0,...x3,y3 are
32271 evaluated.
32272 It accepts the following values:
32274 init
32276 only evaluate expressions once during the filter initialization
32277 or when a command is processed
32278 frame
32280 evaluate expressions for each incoming frame
32281 Default value is init.
32283 phase
32285 Delay interlaced video by one field time so that the field order
32286 changes.
32287 The intended use is to fix PAL movies that have been captured with the
32289 opposite field order to the film-to-video transfer.
32290 A description of the accepted parameters follows.
32292 mode
32294 Set phase mode.
32295 It accepts the following values:
32297 t Capture field order top-first, transfer bottom-first. Filter
32299 will delay the bottom field.
32300 b Capture field order bottom-first, transfer top-first. Filter
32302 will delay the top field.
32303 p Capture and transfer with the same field order. This mode only
32305 exists for the documentation of the other options to refer to,
32306 but if you actually select it, the filter will faithfully do
32307 nothing.
32308 a Capture field order determined automatically by field flags,
32310 transfer opposite. Filter selects among t and b modes on a
32311 frame by frame basis using field flags. If no field information
32312 is available, then this works just like u.
32313 u Capture unknown or varying, transfer opposite. Filter selects
32315 among t and b on a frame by frame basis by analyzing the images
32316 and selecting the alternative that produces best match between
32317 the fields.
32318 T Capture top-first, transfer unknown or varying. Filter selects
32320 among t and p using image analysis.
32321 B Capture bottom-first, transfer unknown or varying. Filter
32323 selects among b and p using image analysis.
32324 A Capture determined by field flags, transfer unknown or varying.
32326 Filter selects among t, b and p using field flags and image
32327 analysis. If no field information is available, then this works
32328 just like U. This is the default mode.
32329 U Both capture and transfer unknown or varying. Filter selects
32331 among t, b and p using image analysis only.
32332 Commands
32334 This filter supports the all above options as commands.
32336 photosensitivity
32338 Reduce various flashes in video, so to help users with epilepsy.
32339 It accepts the following options:
32341 frames, f
32343 Set how many frames to use when filtering. Default is 30.
32344 threshold, t
32346 Set detection threshold factor. Default is 1. Lower is stricter.
32347 skip
32349 Set how many pixels to skip when sampling frames. Default is 1.
32350 Allowed range is from 1 to 1024.
32351 bypass
32353 Leave frames unchanged. Default is disabled.
32354 pixdesctest
32356 Pixel format descriptor test filter, mainly useful for internal
32357 testing. The output video should be equal to the input video.
32358 For example:
32360 format=monow, pixdesctest
32362 can be used to test the monowhite pixel format descriptor definition.
32364 pixscope
32366 Display sample values of color channels. Mainly useful for checking
32367 color and levels. Minimum supported resolution is 640x480.
32368 The filters accept the following options:
32370 x Set scope X position, relative offset on X axis.
32372 y Set scope Y position, relative offset on Y axis.
32374 w Set scope width.
32376 h Set scope height.
32378 o Set window opacity. This window also holds statistics about pixel
32380 area.
32381 wx Set window X position, relative offset on X axis.
32383 wy Set window Y position, relative offset on Y axis.
32385 Commands
32387 This filter supports same commands as options.
32389 pp
32391 Enable the specified chain of postprocessing subfilters using
32392 libpostproc. This library should be automatically selected with a GPL
32393 build ("--enable-gpl"). Subfilters must be separated by '/' and can be
32394 disabled by prepending a '-'. Each subfilter and some options have a
32395 short and a long name that can be used interchangeably, i.e. dr/dering
32396 are the same.
32397 The filters accept the following options:
32399 subfilters
32401 Set postprocessing subfilters string.
32402 All subfilters share common options to determine their scope:
32404 a/autoq
32406 Honor the quality commands for this subfilter.
32407 c/chrom
32409 Do chrominance filtering, too (default).
32410 y/nochrom
32412 Do luminance filtering only (no chrominance).
32413 n/noluma
32415 Do chrominance filtering only (no luminance).
32416 These options can be appended after the subfilter name, separated by a
32418 '|'.
32419 Available subfilters are:
32421 hb/hdeblock[|difference[|flatness]]
32423 Horizontal deblocking filter
32424 difference
32426 Difference factor where higher values mean more deblocking
32427 (default: 32).
32428 flatness
32430 Flatness threshold where lower values mean more deblocking
32431 (default: 39).
32432 vb/vdeblock[|difference[|flatness]]
32434 Vertical deblocking filter
32435 difference
32437 Difference factor where higher values mean more deblocking
32438 (default: 32).
32439 flatness
32441 Flatness threshold where lower values mean more deblocking
32442 (default: 39).
32443 ha/hadeblock[|difference[|flatness]]
32445 Accurate horizontal deblocking filter
32446 difference
32448 Difference factor where higher values mean more deblocking
32449 (default: 32).
32450 flatness
32452 Flatness threshold where lower values mean more deblocking
32453 (default: 39).
32454 va/vadeblock[|difference[|flatness]]
32456 Accurate vertical deblocking filter
32457 difference
32459 Difference factor where higher values mean more deblocking
32460 (default: 32).
32461 flatness
32463 Flatness threshold where lower values mean more deblocking
32464 (default: 39).
32465 The horizontal and vertical deblocking filters share the difference and
32467 flatness values so you cannot set different horizontal and vertical
32468 thresholds.
32469 h1/x1hdeblock
32471 Experimental horizontal deblocking filter
32472 v1/x1vdeblock
32474 Experimental vertical deblocking filter
32475 dr/dering
32477 Deringing filter
32478 tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise
32480 reducer
32481 threshold1
32482 larger -> stronger filtering
32483 threshold2
32485 larger -> stronger filtering
32486 threshold3
32488 larger -> stronger filtering
32489 al/autolevels[:f/fullyrange], automatic brightness / contrast
32491 correction
32492 f/fullyrange
32493 Stretch luminance to "0-255".
32494 lb/linblenddeint
32496 Linear blend deinterlacing filter that deinterlaces the given block
32497 by filtering all lines with a "(1 2 1)" filter.
32498 li/linipoldeint
32500 Linear interpolating deinterlacing filter that deinterlaces the
32501 given block by linearly interpolating every second line.
32502 ci/cubicipoldeint
32504 Cubic interpolating deinterlacing filter deinterlaces the given
32505 block by cubically interpolating every second line.
32506 md/mediandeint
32508 Median deinterlacing filter that deinterlaces the given block by
32509 applying a median filter to every second line.
32510 fd/ffmpegdeint
32512 FFmpeg deinterlacing filter that deinterlaces the given block by
32513 filtering every second line with a "(-1 4 2 4 -1)" filter.
32514 l5/lowpass5
32516 Vertically applied FIR lowpass deinterlacing filter that
32517 deinterlaces the given block by filtering all lines with a "(-1 2 6
32518 2 -1)" filter.
32519 fq/forceQuant[|quantizer]
32521 Overrides the quantizer table from the input with the constant
32522 quantizer you specify.
32523 quantizer
32525 Quantizer to use
32526 de/default
32528 Default pp filter combination ("hb|a,vb|a,dr|a")
32529 fa/fast
32531 Fast pp filter combination ("h1|a,v1|a,dr|a")
32532 ac High quality pp filter combination ("ha|a|128|7,va|a,dr|a")
32534 Examples
32536 • Apply horizontal and vertical deblocking, deringing and automatic
32538 brightness/contrast:
32539 pp=hb/vb/dr/al
32541 • Apply default filters without brightness/contrast correction:
32543 pp=de/-al
32545 • Apply default filters and temporal denoiser:
32547 pp=default/tmpnoise|1|2|3
32549 • Apply deblocking on luminance only, and switch vertical deblocking
32551 on or off automatically depending on available CPU time:
32552 pp=hb|y/vb|a
32554 pp7
32556 Apply Postprocessing filter 7. It is variant of the spp filter, similar
32557 to spp = 6 with 7 point DCT, where only the center sample is used after
32558 IDCT.
32559 The filter accepts the following options:
32561 qp Force a constant quantization parameter. It accepts an integer in
32563 range 0 to 63. If not set, the filter will use the QP from the
32564 video stream (if available).
32565 mode
32567 Set thresholding mode. Available modes are:
32568 hard
32570 Set hard thresholding.
32571 soft
32573 Set soft thresholding (better de-ringing effect, but likely
32574 blurrier).
32575 medium
32577 Set medium thresholding (good results, default).
32578 premultiply
32580 Apply alpha premultiply effect to input video stream using first plane
32581 of second stream as alpha.
32582 Both streams must have same dimensions and same pixel format.
32584 The filter accepts the following option:
32586 planes
32588 Set which planes will be processed, unprocessed planes will be
32589 copied. By default value 0xf, all planes will be processed.
32590 inplace
32592 Do not require 2nd input for processing, instead use alpha plane
32593 from input stream.
32594 prewitt
32596 Apply prewitt operator to input video stream.
32597 The filter accepts the following option:
32599 planes
32601 Set which planes will be processed, unprocessed planes will be
32602 copied. By default value 0xf, all planes will be processed.
32603 scale
32605 Set value which will be multiplied with filtered result.
32606 delta
32608 Set value which will be added to filtered result.
32609 Commands
32611 This filter supports the all above options as commands.
32613 pseudocolor
32615 Alter frame colors in video with pseudocolors.
32616 This filter accepts the following options:
32618 c0 set pixel first component expression
32620 c1 set pixel second component expression
32622 c2 set pixel third component expression
32624 c3 set pixel fourth component expression, corresponds to the alpha
32626 component
32627 index, i
32629 set component to use as base for altering colors
32630 preset, p
32632 Pick one of built-in LUTs. By default is set to none.
32633 Available LUTs:
32635 magma
32637 inferno
32638 plasma
32639 viridis
32640 turbo
32641 cividis
32642 range1
32643 range2
32644 shadows
32645 highlights
32646 solar
32647 nominal
32648 preferred
32649 total
32650 opacity
32651 Set opacity of output colors. Allowed range is from 0 to 1.
32652 Default value is set to 1.
32653 Each of the expression options specifies the expression to use for
32655 computing the lookup table for the corresponding pixel component
32656 values.
32657 The expressions can contain the following constants and functions:
32659 w
32661 h The input width and height.
32662 val The input value for the pixel component.
32664 ymin, umin, vmin, amin
32666 The minimum allowed component value.
32667 ymax, umax, vmax, amax
32669 The maximum allowed component value.
32670 All expressions default to "val".
32672 Commands
32674 This filter supports the all above options as commands.
32676 Examples
32678 • Change too high luma values to gradient:
32680 pseudocolor="'if(between(val,ymax,amax),lerp(ymin,ymax,(val-ymax)/(amax-ymax)),-1):if(between(val,ymax,amax),lerp(umax,umin,(val-ymax)/(amax-ymax)),-1):if(between(val,ymax,amax),lerp(vmin,vmax,(val-ymax)/(amax-ymax)),-1):-1'"
32682 psnr
32684 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
32685 Ratio) between two input videos.
32686 This filter takes in input two input videos, the first input is
32688 considered the "main" source and is passed unchanged to the output. The
32689 second input is used as a "reference" video for computing the PSNR.
32690 Both video inputs must have the same resolution and pixel format for
32692 this filter to work correctly. Also it assumes that both inputs have
32693 the same number of frames, which are compared one by one.
32694 The obtained average PSNR is printed through the logging system.
32696 The filter stores the accumulated MSE (mean squared error) of each
32698 frame, and at the end of the processing it is averaged across all
32699 frames equally, and the following formula is applied to obtain the
32700 PSNR:
32701 PSNR = 10*log10(MAX^2/MSE)
32703 Where MAX is the average of the maximum values of each component of the
32705 image.
32706 The description of the accepted parameters follows.
32708 stats_file, f
32710 If specified the filter will use the named file to save the PSNR of
32711 each individual frame. When filename equals "-" the data is sent to
32712 standard output.
32713 stats_version
32715 Specifies which version of the stats file format to use. Details of
32716 each format are written below. Default value is 1.
32717 stats_add_max
32719 Determines whether the max value is output to the stats log.
32720 Default value is 0. Requires stats_version >= 2. If this is set
32721 and stats_version < 2, the filter will return an error.
32722 This filter also supports the framesync options.
32724 The file printed if stats_file is selected, contains a sequence of
32726 key/value pairs of the form key:value for each compared couple of
32727 frames.
32728 If a stats_version greater than 1 is specified, a header line precedes
32730 the list of per-frame-pair stats, with key value pairs following the
32731 frame format with the following parameters:
32732 psnr_log_version
32734 The version of the log file format. Will match stats_version.
32735 fields
32737 A comma separated list of the per-frame-pair parameters included in
32738 the log.
32739 A description of each shown per-frame-pair parameter follows:
32741 n sequential number of the input frame, starting from 1
32743 mse_avg
32745 Mean Square Error pixel-by-pixel average difference of the compared
32746 frames, averaged over all the image components.
32747 mse_y, mse_u, mse_v, mse_r, mse_g, mse_b, mse_a
32749 Mean Square Error pixel-by-pixel average difference of the compared
32750 frames for the component specified by the suffix.
32751 psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
32753 Peak Signal to Noise ratio of the compared frames for the component
32754 specified by the suffix.
32755 max_avg, max_y, max_u, max_v
32757 Maximum allowed value for each channel, and average over all
32758 channels.
32759 Examples
32761 • For example:
32763 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
32765 [main][ref] psnr="stats_file=stats.log" [out]
32766 On this example the input file being processed is compared with the
32768 reference file ref_movie.mpg. The PSNR of each individual frame is
32769 stored in stats.log.
32770 • Another example with different containers:
32772 ffmpeg -i main.mpg -i ref.mkv -lavfi "[0:v]settb=AVTB,setpts=PTS-STARTPTS[main];[1:v]settb=AVTB,setpts=PTS-STARTPTS[ref];[main][ref]psnr" -f null -
32774 pullup
32776 Pulldown reversal (inverse telecine) filter, capable of handling mixed
32777 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps
32778 progressive content.
32779 The pullup filter is designed to take advantage of future context in
32781 making its decisions. This filter is stateless in the sense that it
32782 does not lock onto a pattern to follow, but it instead looks forward to
32783 the following fields in order to identify matches and rebuild
32784 progressive frames.
32785 To produce content with an even framerate, insert the fps filter after
32787 pullup, use "fps=24000/1001" if the input frame rate is 29.97fps,
32788 "fps=24" for 30fps and the (rare) telecined 25fps input.
32789 The filter accepts the following options:
32791 jl
32793 jr
32794 jt
32795 jb These options set the amount of "junk" to ignore at the left,
32796 right, top, and bottom of the image, respectively. Left and right
32797 are in units of 8 pixels, while top and bottom are in units of 2
32798 lines. The default is 8 pixels on each side.
32799 sb Set the strict breaks. Setting this option to 1 will reduce the
32801 chances of filter generating an occasional mismatched frame, but it
32802 may also cause an excessive number of frames to be dropped during
32803 high motion sequences. Conversely, setting it to -1 will make
32804 filter match fields more easily. This may help processing of video
32805 where there is slight blurring between the fields, but may also
32806 cause there to be interlaced frames in the output. Default value
32807 is 0.
32808 mp Set the metric plane to use. It accepts the following values:
32810 l Use luma plane.
32812 u Use chroma blue plane.
32814 v Use chroma red plane.
32816 This option may be set to use chroma plane instead of the default
32818 luma plane for doing filter's computations. This may improve
32819 accuracy on very clean source material, but more likely will
32820 decrease accuracy, especially if there is chroma noise (rainbow
32821 effect) or any grayscale video. The main purpose of setting mp to
32822 a chroma plane is to reduce CPU load and make pullup usable in
32823 realtime on slow machines.
32824 For best results (without duplicated frames in the output file) it is
32826 necessary to change the output frame rate. For example, to inverse
32827 telecine NTSC input:
32828 ffmpeg -i input -vf pullup -r 24000/1001 ...
32830 qp
32832 Change video quantization parameters (QP).
32833 The filter accepts the following option:
32835 qp Set expression for quantization parameter.
32837 The expression is evaluated through the eval API and can contain, among
32839 others, the following constants:
32840 known
32842 1 if index is not 129, 0 otherwise.
32843 qp Sequential index starting from -129 to 128.
32845 Examples
32847 • Some equation like:
32849 qp=2+2*sin(PI*qp)
32851 random
32853 Flush video frames from internal cache of frames into a random order.
32854 No frame is discarded. Inspired by frei0r nervous filter.
32855 frames
32857 Set size in number of frames of internal cache, in range from 2 to
32858 512. Default is 30.
32859 seed
32861 Set seed for random number generator, must be an integer included
32862 between 0 and "UINT32_MAX". If not specified, or if explicitly set
32863 to less than 0, the filter will try to use a good random seed on a
32864 best effort basis.
32865 readeia608
32867 Read closed captioning (EIA-608) information from the top lines of a
32868 video frame.
32869 This filter adds frame metadata for "lavfi.readeia608.X.cc" and
32871 "lavfi.readeia608.X.line", where "X" is the number of the identified
32872 line with EIA-608 data (starting from 0). A description of each
32873 metadata value follows:
32874 lavfi.readeia608.X.cc
32876 The two bytes stored as EIA-608 data (printed in hexadecimal).
32877 lavfi.readeia608.X.line
32879 The number of the line on which the EIA-608 data was identified and
32880 read.
32881 This filter accepts the following options:
32883 scan_min
32885 Set the line to start scanning for EIA-608 data. Default is 0.
32886 scan_max
32888 Set the line to end scanning for EIA-608 data. Default is 29.
32889 spw Set the ratio of width reserved for sync code detection. Default
32891 is 0.27. Allowed range is "[0.1 - 0.7]".
32892 chp Enable checking the parity bit. In the event of a parity error, the
32894 filter will output 0x00 for that character. Default is false.
32895 lp Lowpass lines prior to further processing. Default is enabled.
32897 Commands
32899 This filter supports the all above options as commands.
32901 Examples
32903 • Output a csv with presentation time and the first two lines of
32905 identified EIA-608 captioning data.
32906 ffprobe -f lavfi -i movie=captioned_video.mov,readeia608 -show_entries frame=pts_time:frame_tags=lavfi.readeia608.0.cc,lavfi.readeia608.1.cc -of csv
32908 readvitc
32910 Read vertical interval timecode (VITC) information from the top lines
32911 of a video frame.
32912 The filter adds frame metadata key "lavfi.readvitc.tc_str" with the
32914 timecode value, if a valid timecode has been detected. Further metadata
32915 key "lavfi.readvitc.found" is set to 0/1 depending on whether timecode
32916 data has been found or not.
32917 This filter accepts the following options:
32919 scan_max
32921 Set the maximum number of lines to scan for VITC data. If the value
32922 is set to "-1" the full video frame is scanned. Default is 45.
32923 thr_b
32925 Set the luma threshold for black. Accepts float numbers in the
32926 range [0.0,1.0], default value is 0.2. The value must be equal or
32927 less than "thr_w".
32928 thr_w
32930 Set the luma threshold for white. Accepts float numbers in the
32931 range [0.0,1.0], default value is 0.6. The value must be equal or
32932 greater than "thr_b".
32933 Examples
32935 • Detect and draw VITC data onto the video frame; if no valid VITC is
32937 detected, draw "--:--:--:--" as a placeholder:
32938 ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--}:x=(w-tw)/2:y=400-ascent'
32940 remap
32942 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
32943 Destination pixel at position (X, Y) will be picked from source (x, y)
32945 position where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are
32946 out of range, zero value for pixel will be used for destination pixel.
32947 Xmap and Ymap input video streams must be of same dimensions. Output
32949 video stream will have Xmap/Ymap video stream dimensions. Xmap and
32950 Ymap input video streams are 16bit depth, single channel.
32951 format
32953 Specify pixel format of output from this filter. Can be "color" or
32954 "gray". Default is "color".
32955 fill
32957 Specify the color of the unmapped pixels. For the syntax of this
32958 option, check the "Color" section in the ffmpeg-utils manual.
32959 Default color is "black".
32960 removegrain
32962 The removegrain filter is a spatial denoiser for progressive video.
32963 m0 Set mode for the first plane.
32965 m1 Set mode for the second plane.
32967 m2 Set mode for the third plane.
32969 m3 Set mode for the fourth plane.
32971 Range of mode is from 0 to 24. Description of each mode follows:
32973 0 Leave input plane unchanged. Default.
32975 1 Clips the pixel with the minimum and maximum of the 8 neighbour
32977 pixels.
32978 2 Clips the pixel with the second minimum and maximum of the 8
32980 neighbour pixels.
32981 3 Clips the pixel with the third minimum and maximum of the 8
32983 neighbour pixels.
32984 4 Clips the pixel with the fourth minimum and maximum of the 8
32986 neighbour pixels. This is equivalent to a median filter.
32987 5 Line-sensitive clipping giving the minimal change.
32989 6 Line-sensitive clipping, intermediate.
32991 7 Line-sensitive clipping, intermediate.
32993 8 Line-sensitive clipping, intermediate.
32995 9 Line-sensitive clipping on a line where the neighbours pixels are
32997 the closest.
32998 10 Replaces the target pixel with the closest neighbour.
33000 11 [1 2 1] horizontal and vertical kernel blur.
33002 12 Same as mode 11.
33004 13 Bob mode, interpolates top field from the line where the neighbours
33006 pixels are the closest.
33007 14 Bob mode, interpolates bottom field from the line where the
33009 neighbours pixels are the closest.
33010 15 Bob mode, interpolates top field. Same as 13 but with a more
33012 complicated interpolation formula.
33013 16 Bob mode, interpolates bottom field. Same as 14 but with a more
33015 complicated interpolation formula.
33016 17 Clips the pixel with the minimum and maximum of respectively the
33018 maximum and minimum of each pair of opposite neighbour pixels.
33019 18 Line-sensitive clipping using opposite neighbours whose greatest
33021 distance from the current pixel is minimal.
33022 19 Replaces the pixel with the average of its 8 neighbours.
33024 20 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
33026 21 Clips pixels using the averages of opposite neighbour.
33028 22 Same as mode 21 but simpler and faster.
33030 23 Small edge and halo removal, but reputed useless.
33032 24 Similar as 23.
33034 removelogo
33036 Suppress a TV station logo, using an image file to determine which
33037 pixels comprise the logo. It works by filling in the pixels that
33038 comprise the logo with neighboring pixels.
33039 The filter accepts the following options:
33041 filename, f
33043 Set the filter bitmap file, which can be any image format supported
33044 by libavformat. The width and height of the image file must match
33045 those of the video stream being processed.
33046 Pixels in the provided bitmap image with a value of zero are not
33048 considered part of the logo, non-zero pixels are considered part of the
33049 logo. If you use white (255) for the logo and black (0) for the rest,
33050 you will be safe. For making the filter bitmap, it is recommended to
33051 take a screen capture of a black frame with the logo visible, and then
33052 using a threshold filter followed by the erode filter once or twice.
33053 If needed, little splotches can be fixed manually. Remember that if
33055 logo pixels are not covered, the filter quality will be much reduced.
33056 Marking too many pixels as part of the logo does not hurt as much, but
33057 it will increase the amount of blurring needed to cover over the image
33058 and will destroy more information than necessary, and extra pixels will
33059 slow things down on a large logo.
33060 repeatfields
33062 This filter uses the repeat_field flag from the Video ES headers and
33063 hard repeats fields based on its value.
33064 reverse
33066 Reverse a video clip.
33067 Warning: This filter requires memory to buffer the entire clip, so
33069 trimming is suggested.
33070 Examples
33072 • Take the first 5 seconds of a clip, and reverse it.
33074 trim=end=5,reverse
33076 rgbashift
33078 Shift R/G/B/A pixels horizontally and/or vertically.
33079 The filter accepts the following options:
33081 rh Set amount to shift red horizontally.
33083 rv Set amount to shift red vertically.
33085 gh Set amount to shift green horizontally.
33087 gv Set amount to shift green vertically.
33089 bh Set amount to shift blue horizontally.
33091 bv Set amount to shift blue vertically.
33093 ah Set amount to shift alpha horizontally.
33095 av Set amount to shift alpha vertically.
33097 edge
33099 Set edge mode, can be smear, default, or warp.
33100 Commands
33102 This filter supports the all above options as commands.
33104 roberts
33106 Apply roberts cross operator to input video stream.
33107 The filter accepts the following option:
33109 planes
33111 Set which planes will be processed, unprocessed planes will be
33112 copied. By default value 0xf, all planes will be processed.
33113 scale
33115 Set value which will be multiplied with filtered result.
33116 delta
33118 Set value which will be added to filtered result.
33119 Commands
33121 This filter supports the all above options as commands.
33123 rotate
33125 Rotate video by an arbitrary angle expressed in radians.
33126 The filter accepts the following options:
33128 A description of the optional parameters follows.
33130 angle, a
33132 Set an expression for the angle by which to rotate the input video
33133 clockwise, expressed as a number of radians. A negative value will
33134 result in a counter-clockwise rotation. By default it is set to
33135 "0".
33136 This expression is evaluated for each frame.
33138 out_w, ow
33140 Set the output width expression, default value is "iw". This
33141 expression is evaluated just once during configuration.
33142 out_h, oh
33144 Set the output height expression, default value is "ih". This
33145 expression is evaluated just once during configuration.
33146 bilinear
33148 Enable bilinear interpolation if set to 1, a value of 0 disables
33149 it. Default value is 1.
33150 fillcolor, c
33152 Set the color used to fill the output area not covered by the
33153 rotated image. For the general syntax of this option, check the
33154 "Color" section in the ffmpeg-utils manual. If the special value
33155 "none" is selected then no background is printed (useful for
33156 example if the background is never shown).
33157 Default value is "black".
33159 The expressions for the angle and the output size can contain the
33161 following constants and functions:
33162 n sequential number of the input frame, starting from 0. It is always
33164 NAN before the first frame is filtered.
33165 t time in seconds of the input frame, it is set to 0 when the filter
33167 is configured. It is always NAN before the first frame is filtered.
33168 hsub
33170 vsub
33171 horizontal and vertical chroma subsample values. For example for
33172 the pixel format "yuv422p" hsub is 2 and vsub is 1.
33173 in_w, iw
33175 in_h, ih
33176 the input video width and height
33177 out_w, ow
33179 out_h, oh
33180 the output width and height, that is the size of the padded area as
33181 specified by the width and height expressions
33182 rotw(a)
33184 roth(a)
33185 the minimal width/height required for completely containing the
33186 input video rotated by a radians.
33187 These are only available when computing the out_w and out_h
33189 expressions.
33190 Examples
33192 • Rotate the input by PI/6 radians clockwise:
33194 rotate=PI/6
33196 • Rotate the input by PI/6 radians counter-clockwise:
33198 rotate=-PI/6
33200 • Rotate the input by 45 degrees clockwise:
33202 rotate=45*PI/180
33204 • Apply a constant rotation with period T, starting from an angle of
33206 PI/3:
33207 rotate=PI/3+2*PI*t/T
33209 • Make the input video rotation oscillating with a period of T
33211 seconds and an amplitude of A radians:
33212 rotate=A*sin(2*PI/T*t)
33214 • Rotate the video, output size is chosen so that the whole rotating
33216 input video is always completely contained in the output:
33217 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
33219 • Rotate the video, reduce the output size so that no background is
33221 ever shown:
33222 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
33224 Commands
33226 The filter supports the following commands:
33228 a, angle
33230 Set the angle expression. The command accepts the same syntax of
33231 the corresponding option.
33232 If the specified expression is not valid, it is kept at its current
33234 value.
33235 sab
33237 Apply Shape Adaptive Blur.
33238 The filter accepts the following options:
33240 luma_radius, lr
33242 Set luma blur filter strength, must be a value in range 0.1-4.0,
33243 default value is 1.0. A greater value will result in a more blurred
33244 image, and in slower processing.
33245 luma_pre_filter_radius, lpfr
33247 Set luma pre-filter radius, must be a value in the 0.1-2.0 range,
33248 default value is 1.0.
33249 luma_strength, ls
33251 Set luma maximum difference between pixels to still be considered,
33252 must be a value in the 0.1-100.0 range, default value is 1.0.
33253 chroma_radius, cr
33255 Set chroma blur filter strength, must be a value in range -0.9-4.0.
33256 A greater value will result in a more blurred image, and in slower
33257 processing.
33258 chroma_pre_filter_radius, cpfr
33260 Set chroma pre-filter radius, must be a value in the -0.9-2.0
33261 range.
33262 chroma_strength, cs
33264 Set chroma maximum difference between pixels to still be
33265 considered, must be a value in the -0.9-100.0 range.
33266 Each chroma option value, if not explicitly specified, is set to the
33268 corresponding luma option value.
33269 scale
33271 Scale (resize) the input video, using the libswscale library.
33272 The scale filter forces the output display aspect ratio to be the same
33274 of the input, by changing the output sample aspect ratio.
33275 If the input image format is different from the format requested by the
33277 next filter, the scale filter will convert the input to the requested
33278 format.
33279 Options
33281 The filter accepts the following options, or any of the options
33283 supported by the libswscale scaler.
33284 See the ffmpeg-scaler manual for the complete list of scaler options.
33286 width, w
33288 height, h
33289 Set the output video dimension expression. Default value is the
33290 input dimension.
33291 If the width or w value is 0, the input width is used for the
33293 output. If the height or h value is 0, the input height is used for
33294 the output.
33295 If one and only one of the values is -n with n >= 1, the scale
33297 filter will use a value that maintains the aspect ratio of the
33298 input image, calculated from the other specified dimension. After
33299 that it will, however, make sure that the calculated dimension is
33300 divisible by n and adjust the value if necessary.
33301 If both values are -n with n >= 1, the behavior will be identical
33303 to both values being set to 0 as previously detailed.
33304 See below for the list of accepted constants for use in the
33306 dimension expression.
33307 eval
33309 Specify when to evaluate width and height expression. It accepts
33310 the following values:
33311 init
33313 Only evaluate expressions once during the filter initialization
33314 or when a command is processed.
33315 frame
33317 Evaluate expressions for each incoming frame.
33318 Default value is init.
33320 interl
33322 Set the interlacing mode. It accepts the following values:
33323 1 Force interlaced aware scaling.
33325 0 Do not apply interlaced scaling.
33327 -1 Select interlaced aware scaling depending on whether the source
33329 frames are flagged as interlaced or not.
33330 Default value is 0.
33332 flags
33334 Set libswscale scaling flags. See the ffmpeg-scaler manual for the
33335 complete list of values. If not explicitly specified the filter
33336 applies the default flags.
33337 param0, param1
33339 Set libswscale input parameters for scaling algorithms that need
33340 them. See the ffmpeg-scaler manual for the complete documentation.
33341 If not explicitly specified the filter applies empty parameters.
33342 size, s
33344 Set the video size. For the syntax of this option, check the "Video
33345 size" section in the ffmpeg-utils manual.
33346 in_color_matrix
33348 out_color_matrix
33349 Set in/output YCbCr color space type.
33350 This allows the autodetected value to be overridden as well as
33352 allows forcing a specific value used for the output and encoder.
33353 If not specified, the color space type depends on the pixel format.
33355 Possible values:
33357 auto
33359 Choose automatically.
33360 bt709
33362 Format conforming to International Telecommunication Union
33363 (ITU) Recommendation BT.709.
33364 fcc Set color space conforming to the United States Federal
33366 Communications Commission (FCC) Code of Federal Regulations
33367 (CFR) Title 47 (2003) 73.682 (a).
33368 bt601
33370 bt470
33371 smpte170m
33372 Set color space conforming to:
33373 • ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
33375 • ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
33377 • Society of Motion Picture and Television Engineers (SMPTE)
33379 ST 170:2004
33380 smpte240m
33382 Set color space conforming to SMPTE ST 240:1999.
33383 bt2020
33385 Set color space conforming to ITU-R BT.2020 non-constant
33386 luminance system.
33387 in_range
33389 out_range
33390 Set in/output YCbCr sample range.
33391 This allows the autodetected value to be overridden as well as
33393 allows forcing a specific value used for the output and encoder. If
33394 not specified, the range depends on the pixel format. Possible
33395 values:
33396 auto/unknown
33398 Choose automatically.
33399 jpeg/full/pc
33401 Set full range (0-255 in case of 8-bit luma).
33402 mpeg/limited/tv
33404 Set "MPEG" range (16-235 in case of 8-bit luma).
33405 force_original_aspect_ratio
33407 Enable decreasing or increasing output video width or height if
33408 necessary to keep the original aspect ratio. Possible values:
33409 disable
33411 Scale the video as specified and disable this feature.
33412 decrease
33414 The output video dimensions will automatically be decreased if
33415 needed.
33416 increase
33418 The output video dimensions will automatically be increased if
33419 needed.
33420 One useful instance of this option is that when you know a specific
33422 device's maximum allowed resolution, you can use this to limit the
33423 output video to that, while retaining the aspect ratio. For
33424 example, device A allows 1280x720 playback, and your video is
33425 1920x800. Using this option (set it to decrease) and specifying
33426 1280x720 to the command line makes the output 1280x533.
33427 Please note that this is a different thing than specifying -1 for w
33429 or h, you still need to specify the output resolution for this
33430 option to work.
33431 force_divisible_by
33433 Ensures that both the output dimensions, width and height, are
33434 divisible by the given integer when used together with
33435 force_original_aspect_ratio. This works similar to using "-n" in
33436 the w and h options.
33437 This option respects the value set for force_original_aspect_ratio,
33439 increasing or decreasing the resolution accordingly. The video's
33440 aspect ratio may be slightly modified.
33441 This option can be handy if you need to have a video fit within or
33443 exceed a defined resolution using force_original_aspect_ratio but
33444 also have encoder restrictions on width or height divisibility.
33445 The values of the w and h options are expressions containing the
33447 following constants:
33448 in_w
33450 in_h
33451 The input width and height
33452 iw
33454 ih These are the same as in_w and in_h.
33455 out_w
33457 out_h
33458 The output (scaled) width and height
33459 ow
33461 oh These are the same as out_w and out_h
33462 a The same as iw / ih
33464 sar input sample aspect ratio
33466 dar The input display aspect ratio. Calculated from "(iw / ih) * sar".
33468 hsub
33470 vsub
33471 horizontal and vertical input chroma subsample values. For example
33472 for the pixel format "yuv422p" hsub is 2 and vsub is 1.
33473 ohsub
33475 ovsub
33476 horizontal and vertical output chroma subsample values. For example
33477 for the pixel format "yuv422p" hsub is 2 and vsub is 1.
33478 n The (sequential) number of the input frame, starting from 0. Only
33480 available with "eval=frame".
33481 t The presentation timestamp of the input frame, expressed as a
33483 number of seconds. Only available with "eval=frame".
33484 pos The position (byte offset) of the frame in the input stream, or NaN
33486 if this information is unavailable and/or meaningless (for example
33487 in case of synthetic video). Only available with "eval=frame".
33488 Examples
33490 • Scale the input video to a size of 200x100
33492 scale=w=200:h=100
33494 This is equivalent to:
33496 scale=200:100
33498 or:
33500 scale=200x100
33502 • Specify a size abbreviation for the output size:
33504 scale=qcif
33506 which can also be written as:
33508 scale=size=qcif
33510 • Scale the input to 2x:
33512 scale=w=2*iw:h=2*ih
33514 • The above is the same as:
33516 scale=2*in_w:2*in_h
33518 • Scale the input to 2x with forced interlaced scaling:
33520 scale=2*iw:2*ih:interl=1
33522 • Scale the input to half size:
33524 scale=w=iw/2:h=ih/2
33526 • Increase the width, and set the height to the same size:
33528 scale=3/2*iw:ow
33530 • Seek Greek harmony:
33532 scale=iw:1/PHI*iw
33534 scale=ih*PHI:ih
33535 • Increase the height, and set the width to 3/2 of the height:
33537 scale=w=3/2*oh:h=3/5*ih
33539 • Increase the size, making the size a multiple of the chroma
33541 subsample values:
33542 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
33544 • Increase the width to a maximum of 500 pixels, keeping the same
33546 aspect ratio as the input:
33547 scale=w='min(500\, iw*3/2):h=-1'
33549 • Make pixels square by combining scale and setsar:
33551 scale='trunc(ih*dar):ih',setsar=1/1
33553 • Make pixels square by combining scale and setsar, making sure the
33555 resulting resolution is even (required by some codecs):
33556 scale='trunc(ih*dar/2)*2:trunc(ih/2)*2',setsar=1/1
33558 Commands
33560 This filter supports the following commands:
33562 width, w
33564 height, h
33565 Set the output video dimension expression. The command accepts the
33566 same syntax of the corresponding option.
33567 If the specified expression is not valid, it is kept at its current
33569 value.
33570 scale_cuda
33572 Scale (resize) and convert (pixel format) the input video, using
33573 accelerated CUDA kernels. Setting the output width and height works in
33574 the same way as for the scale filter.
33575 The filter accepts the following options:
33577 w
33579 h Set the output video dimension expression. Default value is the
33580 input dimension.
33581 Allows for the same expressions as the scale filter.
33583 interp_algo
33585 Sets the algorithm used for scaling:
33586 nearest
33588 Nearest neighbour
33589 Used by default if input parameters match the desired output.
33591 bilinear
33593 Bilinear
33594 bicubic
33596 Bicubic
33597 This is the default.
33599 lanczos
33601 Lanczos
33602 format
33604 Controls the output pixel format. By default, or if none is
33605 specified, the input pixel format is used.
33606 The filter does not support converting between YUV and RGB pixel
33608 formats.
33609 passthrough
33611 If set to 0, every frame is processed, even if no conversion is
33612 neccesary. This mode can be useful to use the filter as a buffer
33613 for a downstream frame-consumer that exhausts the limited decoder
33614 frame pool.
33615 If set to 1, frames are passed through as-is if they match the
33617 desired output parameters. This is the default behaviour.
33618 param
33620 Algorithm-Specific parameter.
33621 Affects the curves of the bicubic algorithm.
33623 force_original_aspect_ratio
33625 force_divisible_by
33626 Work the same as the identical scale filter options.
33627 Examples
33629 • Scale input to 720p, keeping aspect ratio and ensuring the output
33631 is yuv420p.
33632 scale_cuda=-2:720:format=yuv420p
33634 • Upscale to 4K using nearest neighbour algorithm.
33636 scale_cuda=4096:2160:interp_algo=nearest
33638 • Don't do any conversion or scaling, but copy all input frames into
33640 newly allocated ones. This can be useful to deal with a filter and
33641 encode chain that otherwise exhausts the decoders frame pool.
33642 scale_cuda=passthrough=0
33644 scale_npp
33646 Use the NVIDIA Performance Primitives (libnpp) to perform scaling
33647 and/or pixel format conversion on CUDA video frames. Setting the output
33648 width and height works in the same way as for the scale filter.
33649 The following additional options are accepted:
33651 format
33653 The pixel format of the output CUDA frames. If set to the string
33654 "same" (the default), the input format will be kept. Note that
33655 automatic format negotiation and conversion is not yet supported
33656 for hardware frames
33657 interp_algo
33659 The interpolation algorithm used for resizing. One of the
33660 following:
33661 nn Nearest neighbour.
33663 linear
33665 cubic
33666 cubic2p_bspline
33667 2-parameter cubic (B=1, C=0)
33668 cubic2p_catmullrom
33670 2-parameter cubic (B=0, C=1/2)
33671 cubic2p_b05c03
33673 2-parameter cubic (B=1/2, C=3/10)
33674 super
33676 Supersampling
33677 lanczos
33679 force_original_aspect_ratio
33680 Enable decreasing or increasing output video width or height if
33681 necessary to keep the original aspect ratio. Possible values:
33682 disable
33684 Scale the video as specified and disable this feature.
33685 decrease
33687 The output video dimensions will automatically be decreased if
33688 needed.
33689 increase
33691 The output video dimensions will automatically be increased if
33692 needed.
33693 One useful instance of this option is that when you know a specific
33695 device's maximum allowed resolution, you can use this to limit the
33696 output video to that, while retaining the aspect ratio. For
33697 example, device A allows 1280x720 playback, and your video is
33698 1920x800. Using this option (set it to decrease) and specifying
33699 1280x720 to the command line makes the output 1280x533.
33700 Please note that this is a different thing than specifying -1 for w
33702 or h, you still need to specify the output resolution for this
33703 option to work.
33704 force_divisible_by
33706 Ensures that both the output dimensions, width and height, are
33707 divisible by the given integer when used together with
33708 force_original_aspect_ratio. This works similar to using "-n" in
33709 the w and h options.
33710 This option respects the value set for force_original_aspect_ratio,
33712 increasing or decreasing the resolution accordingly. The video's
33713 aspect ratio may be slightly modified.
33714 This option can be handy if you need to have a video fit within or
33716 exceed a defined resolution using force_original_aspect_ratio but
33717 also have encoder restrictions on width or height divisibility.
33718 eval
33720 Specify when to evaluate width and height expression. It accepts
33721 the following values:
33722 init
33724 Only evaluate expressions once during the filter initialization
33725 or when a command is processed.
33726 frame
33728 Evaluate expressions for each incoming frame.
33729 The values of the w and h options are expressions containing the
33731 following constants:
33732 in_w
33734 in_h
33735 The input width and height
33736 iw
33738 ih These are the same as in_w and in_h.
33739 out_w
33741 out_h
33742 The output (scaled) width and height
33743 ow
33745 oh These are the same as out_w and out_h
33746 a The same as iw / ih
33748 sar input sample aspect ratio
33750 dar The input display aspect ratio. Calculated from "(iw / ih) * sar".
33752 n The (sequential) number of the input frame, starting from 0. Only
33754 available with "eval=frame".
33755 t The presentation timestamp of the input frame, expressed as a
33757 number of seconds. Only available with "eval=frame".
33758 pos The position (byte offset) of the frame in the input stream, or NaN
33760 if this information is unavailable and/or meaningless (for example
33761 in case of synthetic video). Only available with "eval=frame".
33762 scale2ref
33764 Scale (resize) the input video, based on a reference video.
33765 See the scale filter for available options, scale2ref supports the same
33767 but uses the reference video instead of the main input as basis.
33768 scale2ref also supports the following additional constants for the w
33769 and h options:
33770 main_w
33772 main_h
33773 The main input video's width and height
33774 main_a
33776 The same as main_w / main_h
33777 main_sar
33779 The main input video's sample aspect ratio
33780 main_dar, mdar
33782 The main input video's display aspect ratio. Calculated from
33783 "(main_w / main_h) * main_sar".
33784 main_hsub
33786 main_vsub
33787 The main input video's horizontal and vertical chroma subsample
33788 values. For example for the pixel format "yuv422p" hsub is 2 and
33789 vsub is 1.
33790 main_n
33792 The (sequential) number of the main input frame, starting from 0.
33793 Only available with "eval=frame".
33794 main_t
33796 The presentation timestamp of the main input frame, expressed as a
33797 number of seconds. Only available with "eval=frame".
33798 main_pos
33800 The position (byte offset) of the frame in the main input stream,
33801 or NaN if this information is unavailable and/or meaningless (for
33802 example in case of synthetic video). Only available with
33803 "eval=frame".
33804 Examples
33806 • Scale a subtitle stream (b) to match the main video (a) in size
33808 before overlaying
33809 'scale2ref[b][a];[a][b]overlay'
33811 • Scale a logo to 1/10th the height of a video, while preserving its
33813 display aspect ratio.
33814 [logo-in][video-in]scale2ref=w=oh*mdar:h=ih/10[logo-out][video-out]
33816 Commands
33818 This filter supports the following commands:
33820 width, w
33822 height, h
33823 Set the output video dimension expression. The command accepts the
33824 same syntax of the corresponding option.
33825 If the specified expression is not valid, it is kept at its current
33827 value.
33828 scale2ref_npp
33830 Use the NVIDIA Performance Primitives (libnpp) to scale (resize) the
33831 input video, based on a reference video.
33832 See the scale_npp filter for available options, scale2ref_npp supports
33834 the same but uses the reference video instead of the main input as
33835 basis. scale2ref_npp also supports the following additional constants
33836 for the w and h options:
33837 main_w
33839 main_h
33840 The main input video's width and height
33841 main_a
33843 The same as main_w / main_h
33844 main_sar
33846 The main input video's sample aspect ratio
33847 main_dar, mdar
33849 The main input video's display aspect ratio. Calculated from
33850 "(main_w / main_h) * main_sar".
33851 main_n
33853 The (sequential) number of the main input frame, starting from 0.
33854 Only available with "eval=frame".
33855 main_t
33857 The presentation timestamp of the main input frame, expressed as a
33858 number of seconds. Only available with "eval=frame".
33859 main_pos
33861 The position (byte offset) of the frame in the main input stream,
33862 or NaN if this information is unavailable and/or meaningless (for
33863 example in case of synthetic video). Only available with
33864 "eval=frame".
33865 Examples
33867 • Scale a subtitle stream (b) to match the main video (a) in size
33869 before overlaying
33870 'scale2ref_npp[b][a];[a][b]overlay_cuda'
33872 • Scale a logo to 1/10th the height of a video, while preserving its
33874 display aspect ratio.
33875 [logo-in][video-in]scale2ref_npp=w=oh*mdar:h=ih/10[logo-out][video-out]
33877 scharr
33879 Apply scharr operator to input video stream.
33880 The filter accepts the following option:
33882 planes
33884 Set which planes will be processed, unprocessed planes will be
33885 copied. By default value 0xf, all planes will be processed.
33886 scale
33888 Set value which will be multiplied with filtered result.
33889 delta
33891 Set value which will be added to filtered result.
33892 Commands
33894 This filter supports the all above options as commands.
33896 scroll
33898 Scroll input video horizontally and/or vertically by constant speed.
33899 The filter accepts the following options:
33901 horizontal, h
33903 Set the horizontal scrolling speed. Default is 0. Allowed range is
33904 from -1 to 1. Negative values changes scrolling direction.
33905 vertical, v
33907 Set the vertical scrolling speed. Default is 0. Allowed range is
33908 from -1 to 1. Negative values changes scrolling direction.
33909 hpos
33911 Set the initial horizontal scrolling position. Default is 0.
33912 Allowed range is from 0 to 1.
33913 vpos
33915 Set the initial vertical scrolling position. Default is 0. Allowed
33916 range is from 0 to 1.
33917 Commands
33919 This filter supports the following commands:
33921 horizontal, h
33923 Set the horizontal scrolling speed.
33924 vertical, v
33926 Set the vertical scrolling speed.
33927 scdet
33929 Detect video scene change.
33930 This filter sets frame metadata with mafd between frame, the scene
33932 score, and forward the frame to the next filter, so they can use these
33933 metadata to detect scene change or others.
33934 In addition, this filter logs a message and sets frame metadata when it
33936 detects a scene change by threshold.
33937 "lavfi.scd.mafd" metadata keys are set with mafd for every frame.
33939 "lavfi.scd.score" metadata keys are set with scene change score for
33941 every frame to detect scene change.
33942 "lavfi.scd.time" metadata keys are set with current filtered frame time
33944 which detect scene change with threshold.
33945 The filter accepts the following options:
33947 threshold, t
33949 Set the scene change detection threshold as a percentage of maximum
33950 change. Good values are in the "[8.0, 14.0]" range. The range for
33951 threshold is "[0., 100.]".
33952 Default value is 10..
33954 sc_pass, s
33956 Set the flag to pass scene change frames to the next filter.
33957 Default value is 0 You can enable it if you want to get snapshot of
33958 scene change frames only.
33959 selectivecolor
33961 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of
33962 colors (such as "reds", "yellows", "greens", "cyans", ...). The
33963 adjustment range is defined by the "purity" of the color (that is, how
33964 saturated it already is).
33965 This filter is similar to the Adobe Photoshop Selective Color tool.
33967 The filter accepts the following options:
33969 correction_method
33971 Select color correction method.
33972 Available values are:
33974 absolute
33976 Specified adjustments are applied "as-is" (added/subtracted to
33977 original pixel component value).
33978 relative
33980 Specified adjustments are relative to the original component
33981 value.
33982 Default is "absolute".
33984 reds
33986 Adjustments for red pixels (pixels where the red component is the
33987 maximum)
33988 yellows
33990 Adjustments for yellow pixels (pixels where the blue component is
33991 the minimum)
33992 greens
33994 Adjustments for green pixels (pixels where the green component is
33995 the maximum)
33996 cyans
33998 Adjustments for cyan pixels (pixels where the red component is the
33999 minimum)
34000 blues
34002 Adjustments for blue pixels (pixels where the blue component is the
34003 maximum)
34004 magentas
34006 Adjustments for magenta pixels (pixels where the green component is
34007 the minimum)
34008 whites
34010 Adjustments for white pixels (pixels where all components are
34011 greater than 128)
34012 neutrals
34014 Adjustments for all pixels except pure black and pure white
34015 blacks
34017 Adjustments for black pixels (pixels where all components are
34018 lesser than 128)
34019 psfile
34021 Specify a Photoshop selective color file (".asv") to import the
34022 settings from.
34023 All the adjustment settings (reds, yellows, ...) accept up to 4 space
34025 separated floating point adjustment values in the [-1,1] range,
34026 respectively to adjust the amount of cyan, magenta, yellow and black
34027 for the pixels of its range.
34028 Examples
34030 • Increase cyan by 50% and reduce yellow by 33% in every green areas,
34032 and increase magenta by 27% in blue areas:
34033 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
34035 • Use a Photoshop selective color preset:
34037 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
34039 separatefields
34041 The "separatefields" takes a frame-based video input and splits each
34042 frame into its components fields, producing a new half height clip with
34043 twice the frame rate and twice the frame count.
34044 This filter use field-dominance information in frame to decide which of
34046 each pair of fields to place first in the output. If it gets it wrong
34047 use setfield filter before "separatefields" filter.
34048 setdar, setsar
34050 The "setdar" filter sets the Display Aspect Ratio for the filter output
34051 video.
34052 This is done by changing the specified Sample (aka Pixel) Aspect Ratio,
34054 according to the following equation:
34055 <DAR> = <HORIZONTAL_RESOLUTION> / <VERTICAL_RESOLUTION> * <SAR>
34057 Keep in mind that the "setdar" filter does not modify the pixel
34059 dimensions of the video frame. Also, the display aspect ratio set by
34060 this filter may be changed by later filters in the filterchain, e.g. in
34061 case of scaling or if another "setdar" or a "setsar" filter is applied.
34062 The "setsar" filter sets the Sample (aka Pixel) Aspect Ratio for the
34064 filter output video.
34065 Note that as a consequence of the application of this filter, the
34067 output display aspect ratio will change according to the equation
34068 above.
34069 Keep in mind that the sample aspect ratio set by the "setsar" filter
34071 may be changed by later filters in the filterchain, e.g. if another
34072 "setsar" or a "setdar" filter is applied.
34073 It accepts the following parameters:
34075 r, ratio, dar ("setdar" only), sar ("setsar" only)
34077 Set the aspect ratio used by the filter.
34078 The parameter can be a floating point number string, an expression,
34080 or a string of the form num:den, where num and den are the
34081 numerator and denominator of the aspect ratio. If the parameter is
34082 not specified, it is assumed the value "0". In case the form
34083 "num:den" is used, the ":" character should be escaped.
34084 max Set the maximum integer value to use for expressing numerator and
34086 denominator when reducing the expressed aspect ratio to a rational.
34087 Default value is 100.
34088 The parameter sar is an expression containing the following constants:
34090 E, PI, PHI
34092 These are approximated values for the mathematical constants e
34093 (Euler's number), pi (Greek pi), and phi (the golden ratio).
34094 w, h
34096 The input width and height.
34097 a These are the same as w / h.
34099 sar The input sample aspect ratio.
34101 dar The input display aspect ratio. It is the same as (w / h) * sar.
34103 hsub, vsub
34105 Horizontal and vertical chroma subsample values. For example, for
34106 the pixel format "yuv422p" hsub is 2 and vsub is 1.
34107 Examples
34109 • To change the display aspect ratio to 16:9, specify one of the
34111 following:
34112 setdar=dar=1.77777
34114 setdar=dar=16/9
34115 • To change the sample aspect ratio to 10:11, specify:
34117 setsar=sar=10/11
34119 • To set a display aspect ratio of 16:9, and specify a maximum
34121 integer value of 1000 in the aspect ratio reduction, use the
34122 command:
34123 setdar=ratio=16/9:max=1000
34125 setfield
34127 Force field for the output video frame.
34128 The "setfield" filter marks the interlace type field for the output
34130 frames. It does not change the input frame, but only sets the
34131 corresponding property, which affects how the frame is treated by
34132 following filters (e.g. "fieldorder" or "yadif").
34133 The filter accepts the following options:
34135 mode
34137 Available values are:
34138 auto
34140 Keep the same field property.
34141 bff Mark the frame as bottom-field-first.
34143 tff Mark the frame as top-field-first.
34145 prog
34147 Mark the frame as progressive.
34148 setparams
34150 Force frame parameter for the output video frame.
34151 The "setparams" filter marks interlace and color range for the output
34153 frames. It does not change the input frame, but only sets the
34154 corresponding property, which affects how the frame is treated by
34155 filters/encoders.
34156 field_mode
34158 Available values are:
34159 auto
34161 Keep the same field property (default).
34162 bff Mark the frame as bottom-field-first.
34164 tff Mark the frame as top-field-first.
34166 prog
34168 Mark the frame as progressive.
34169 range
34171 Available values are:
34172 auto
34174 Keep the same color range property (default).
34175 unspecified, unknown
34177 Mark the frame as unspecified color range.
34178 limited, tv, mpeg
34180 Mark the frame as limited range.
34181 full, pc, jpeg
34183 Mark the frame as full range.
34184 color_primaries
34186 Set the color primaries. Available values are:
34187 auto
34189 Keep the same color primaries property (default).
34190 bt709
34192 unknown
34193 bt470m
34194 bt470bg
34195 smpte170m
34196 smpte240m
34197 film
34198 bt2020
34199 smpte428
34200 smpte431
34201 smpte432
34202 jedec-p22
34203 color_trc
34204 Set the color transfer. Available values are:
34205 auto
34207 Keep the same color trc property (default).
34208 bt709
34210 unknown
34211 bt470m
34212 bt470bg
34213 smpte170m
34214 smpte240m
34215 linear
34216 log100
34217 log316
34218 iec61966-2-4
34219 bt1361e
34220 iec61966-2-1
34221 bt2020-10
34222 bt2020-12
34223 smpte2084
34224 smpte428
34225 arib-std-b67
34226 colorspace
34227 Set the colorspace. Available values are:
34228 auto
34230 Keep the same colorspace property (default).
34231 gbr
34233 bt709
34234 unknown
34235 fcc
34236 bt470bg
34237 smpte170m
34238 smpte240m
34239 ycgco
34240 bt2020nc
34241 bt2020c
34242 smpte2085
34243 chroma-derived-nc
34244 chroma-derived-c
34245 ictcp
34246 sharpen_npp
34248 Use the NVIDIA Performance Primitives (libnpp) to perform image
34249 sharpening with border control.
34250 The following additional options are accepted:
34252 border_type
34254 Type of sampling to be used ad frame borders. One of the following:
34255 replicate
34257 Replicate pixel values.
34258 shear
34260 Apply shear transform to input video.
34261 This filter supports the following options:
34263 shx Shear factor in X-direction. Default value is 0. Allowed range is
34265 from -2 to 2.
34266 shy Shear factor in Y-direction. Default value is 0. Allowed range is
34268 from -2 to 2.
34269 fillcolor, c
34271 Set the color used to fill the output area not covered by the
34272 transformed video. For the general syntax of this option, check the
34273 "Color" section in the ffmpeg-utils manual. If the special value
34274 "none" is selected then no background is printed (useful for
34275 example if the background is never shown).
34276 Default value is "black".
34278 interp
34280 Set interpolation type. Can be "bilinear" or "nearest". Default is
34281 "bilinear".
34282 Commands
34284 This filter supports the all above options as commands.
34286 showinfo
34288 Show a line containing various information for each input video frame.
34289 The input video is not modified.
34290 This filter supports the following options:
34292 checksum
34294 Calculate checksums of each plane. By default enabled.
34295 The shown line contains a sequence of key/value pairs of the form
34297 key:value.
34298 The following values are shown in the output:
34300 n The (sequential) number of the input frame, starting from 0.
34302 pts The Presentation TimeStamp of the input frame, expressed as a
34304 number of time base units. The time base unit depends on the filter
34305 input pad.
34306 pts_time
34308 The Presentation TimeStamp of the input frame, expressed as a
34309 number of seconds.
34310 pos The position of the frame in the input stream, or -1 if this
34312 information is unavailable and/or meaningless (for example in case
34313 of synthetic video).
34314 fmt The pixel format name.
34316 sar The sample aspect ratio of the input frame, expressed in the form
34318 num/den.
34319 s The size of the input frame. For the syntax of this option, check
34321 the "Video size" section in the ffmpeg-utils manual.
34322 i The type of interlaced mode ("P" for "progressive", "T" for top
34324 field first, "B" for bottom field first).
34325 iskey
34327 This is 1 if the frame is a key frame, 0 otherwise.
34328 type
34330 The picture type of the input frame ("I" for an I-frame, "P" for a
34331 P-frame, "B" for a B-frame, or "?" for an unknown type). Also
34332 refer to the documentation of the "AVPictureType" enum and of the
34333 "av_get_picture_type_char" function defined in libavutil/avutil.h.
34334 checksum
34336 The Adler-32 checksum (printed in hexadecimal) of all the planes of
34337 the input frame.
34338 plane_checksum
34340 The Adler-32 checksum (printed in hexadecimal) of each plane of the
34341 input frame, expressed in the form "[c0 c1 c2 c3]".
34342 mean
34344 The mean value of pixels in each plane of the input frame,
34345 expressed in the form "[mean0 mean1 mean2 mean3]".
34346 stdev
34348 The standard deviation of pixel values in each plane of the input
34349 frame, expressed in the form "[stdev0 stdev1 stdev2 stdev3]".
34350 showpalette
34352 Displays the 256 colors palette of each frame. This filter is only
34353 relevant for pal8 pixel format frames.
34354 It accepts the following option:
34356 s Set the size of the box used to represent one palette color entry.
34358 Default is 30 (for a "30x30" pixel box).
34359 shuffleframes
34361 Reorder and/or duplicate and/or drop video frames.
34362 It accepts the following parameters:
34364 mapping
34366 Set the destination indexes of input frames. This is space or '|'
34367 separated list of indexes that maps input frames to output frames.
34368 Number of indexes also sets maximal value that each index may have.
34369 '-1' index have special meaning and that is to drop frame.
34370 The first frame has the index 0. The default is to keep the input
34372 unchanged.
34373 Examples
34375 • Swap second and third frame of every three frames of the input:
34377 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
34379 • Swap 10th and 1st frame of every ten frames of the input:
34381 ffmpeg -i INPUT -vf "shuffleframes=9 1 2 3 4 5 6 7 8 0" OUTPUT
34383 shufflepixels
34385 Reorder pixels in video frames.
34386 This filter accepts the following options:
34388 direction, d
34390 Set shuffle direction. Can be forward or inverse direction.
34391 Default direction is forward.
34392 mode, m
34394 Set shuffle mode. Can be horizontal, vertical or block mode.
34395 width, w
34397 height, h
34398 Set shuffle block_size. In case of horizontal shuffle mode only
34399 width part of size is used, and in case of vertical shuffle mode
34400 only height part of size is used.
34401 seed, s
34403 Set random seed used with shuffling pixels. Mainly useful to set to
34404 be able to reverse filtering process to get original input. For
34405 example, to reverse forward shuffle you need to use same parameters
34406 and exact same seed and to set direction to inverse.
34407 shuffleplanes
34409 Reorder and/or duplicate video planes.
34410 It accepts the following parameters:
34412 map0
34414 The index of the input plane to be used as the first output plane.
34415 map1
34417 The index of the input plane to be used as the second output plane.
34418 map2
34420 The index of the input plane to be used as the third output plane.
34421 map3
34423 The index of the input plane to be used as the fourth output plane.
34424 The first plane has the index 0. The default is to keep the input
34426 unchanged.
34427 Examples
34429 • Swap the second and third planes of the input:
34431 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
34433 signalstats
34435 Evaluate various visual metrics that assist in determining issues
34436 associated with the digitization of analog video media.
34437 By default the filter will log these metadata values:
34439 YMIN
34441 Display the minimal Y value contained within the input frame.
34442 Expressed in range of [0-255].
34443 YLOW
34445 Display the Y value at the 10% percentile within the input frame.
34446 Expressed in range of [0-255].
34447 YAVG
34449 Display the average Y value within the input frame. Expressed in
34450 range of [0-255].
34451 YHIGH
34453 Display the Y value at the 90% percentile within the input frame.
34454 Expressed in range of [0-255].
34455 YMAX
34457 Display the maximum Y value contained within the input frame.
34458 Expressed in range of [0-255].
34459 UMIN
34461 Display the minimal U value contained within the input frame.
34462 Expressed in range of [0-255].
34463 ULOW
34465 Display the U value at the 10% percentile within the input frame.
34466 Expressed in range of [0-255].
34467 UAVG
34469 Display the average U value within the input frame. Expressed in
34470 range of [0-255].
34471 UHIGH
34473 Display the U value at the 90% percentile within the input frame.
34474 Expressed in range of [0-255].
34475 UMAX
34477 Display the maximum U value contained within the input frame.
34478 Expressed in range of [0-255].
34479 VMIN
34481 Display the minimal V value contained within the input frame.
34482 Expressed in range of [0-255].
34483 VLOW
34485 Display the V value at the 10% percentile within the input frame.
34486 Expressed in range of [0-255].
34487 VAVG
34489 Display the average V value within the input frame. Expressed in
34490 range of [0-255].
34491 VHIGH
34493 Display the V value at the 90% percentile within the input frame.
34494 Expressed in range of [0-255].
34495 VMAX
34497 Display the maximum V value contained within the input frame.
34498 Expressed in range of [0-255].
34499 SATMIN
34501 Display the minimal saturation value contained within the input
34502 frame. Expressed in range of [0-~181.02].
34503 SATLOW
34505 Display the saturation value at the 10% percentile within the input
34506 frame. Expressed in range of [0-~181.02].
34507 SATAVG
34509 Display the average saturation value within the input frame.
34510 Expressed in range of [0-~181.02].
34511 SATHIGH
34513 Display the saturation value at the 90% percentile within the input
34514 frame. Expressed in range of [0-~181.02].
34515 SATMAX
34517 Display the maximum saturation value contained within the input
34518 frame. Expressed in range of [0-~181.02].
34519 HUEMED
34521 Display the median value for hue within the input frame. Expressed
34522 in range of [0-360].
34523 HUEAVG
34525 Display the average value for hue within the input frame. Expressed
34526 in range of [0-360].
34527 YDIF
34529 Display the average of sample value difference between all values
34530 of the Y plane in the current frame and corresponding values of the
34531 previous input frame. Expressed in range of [0-255].
34532 UDIF
34534 Display the average of sample value difference between all values
34535 of the U plane in the current frame and corresponding values of the
34536 previous input frame. Expressed in range of [0-255].
34537 VDIF
34539 Display the average of sample value difference between all values
34540 of the V plane in the current frame and corresponding values of the
34541 previous input frame. Expressed in range of [0-255].
34542 YBITDEPTH
34544 Display bit depth of Y plane in current frame. Expressed in range
34545 of [0-16].
34546 UBITDEPTH
34548 Display bit depth of U plane in current frame. Expressed in range
34549 of [0-16].
34550 VBITDEPTH
34552 Display bit depth of V plane in current frame. Expressed in range
34553 of [0-16].
34554 The filter accepts the following options:
34556 stat
34558 out stat specify an additional form of image analysis. out output
34559 video with the specified type of pixel highlighted.
34560 Both options accept the following values:
34562 tout
34564 Identify temporal outliers pixels. A temporal outlier is a
34565 pixel unlike the neighboring pixels of the same field. Examples
34566 of temporal outliers include the results of video dropouts,
34567 head clogs, or tape tracking issues.
34568 vrep
34570 Identify vertical line repetition. Vertical line repetition
34571 includes similar rows of pixels within a frame. In born-digital
34572 video vertical line repetition is common, but this pattern is
34573 uncommon in video digitized from an analog source. When it
34574 occurs in video that results from the digitization of an analog
34575 source it can indicate concealment from a dropout compensator.
34576 brng
34578 Identify pixels that fall outside of legal broadcast range.
34579 color, c
34581 Set the highlight color for the out option. The default color is
34582 yellow.
34583 Examples
34585 • Output data of various video metrics:
34587 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
34589 • Output specific data about the minimum and maximum values of the Y
34591 plane per frame:
34592 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
34594 • Playback video while highlighting pixels that are outside of
34596 broadcast range in red.
34597 ffplay example.mov -vf signalstats="out=brng:color=red"
34599 • Playback video with signalstats metadata drawn over the frame.
34601 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
34603 The contents of signalstat_drawtext.txt used in the command are:
34605 time %{pts:hms}
34607 Y (%{metadata:lavfi.signalstats.YMIN}-%{metadata:lavfi.signalstats.YMAX})
34608 U (%{metadata:lavfi.signalstats.UMIN}-%{metadata:lavfi.signalstats.UMAX})
34609 V (%{metadata:lavfi.signalstats.VMIN}-%{metadata:lavfi.signalstats.VMAX})
34610 saturation maximum: %{metadata:lavfi.signalstats.SATMAX}
34611 signature
34613 Calculates the MPEG-7 Video Signature. The filter can handle more than
34614 one input. In this case the matching between the inputs can be
34615 calculated additionally. The filter always passes through the first
34616 input. The signature of each stream can be written into a file.
34617 It accepts the following options:
34619 detectmode
34621 Enable or disable the matching process.
34622 Available values are:
34624 off Disable the calculation of a matching (default).
34626 full
34628 Calculate the matching for the whole video and output whether
34629 the whole video matches or only parts.
34630 fast
34632 Calculate only until a matching is found or the video ends.
34633 Should be faster in some cases.
34634 nb_inputs
34636 Set the number of inputs. The option value must be a non negative
34637 integer. Default value is 1.
34638 filename
34640 Set the path to which the output is written. If there is more than
34641 one input, the path must be a prototype, i.e. must contain %d or
34642 %0nd (where n is a positive integer), that will be replaced with
34643 the input number. If no filename is specified, no output will be
34644 written. This is the default.
34645 format
34647 Choose the output format.
34648 Available values are:
34650 binary
34652 Use the specified binary representation (default).
34653 xml Use the specified xml representation.
34655 th_d
34657 Set threshold to detect one word as similar. The option value must
34658 be an integer greater than zero. The default value is 9000.
34659 th_dc
34661 Set threshold to detect all words as similar. The option value must
34662 be an integer greater than zero. The default value is 60000.
34663 th_xh
34665 Set threshold to detect frames as similar. The option value must be
34666 an integer greater than zero. The default value is 116.
34667 th_di
34669 Set the minimum length of a sequence in frames to recognize it as
34670 matching sequence. The option value must be a non negative integer
34671 value. The default value is 0.
34672 th_it
34674 Set the minimum relation, that matching frames to all frames must
34675 have. The option value must be a double value between 0 and 1. The
34676 default value is 0.5.
34677 Examples
34679 • To calculate the signature of an input video and store it in
34681 signature.bin:
34682 ffmpeg -i input.mkv -vf signature=filename=signature.bin -map 0:v -f null -
34684 • To detect whether two videos match and store the signatures in XML
34686 format in signature0.xml and signature1.xml:
34687 ffmpeg -i input1.mkv -i input2.mkv -filter_complex "[0:v][1:v] signature=nb_inputs=2:detectmode=full:format=xml:filename=signature%d.xml" -map :v -f null -
34689 smartblur
34691 Blur the input video without impacting the outlines.
34692 It accepts the following options:
34694 luma_radius, lr
34696 Set the luma radius. The option value must be a float number in the
34697 range [0.1,5.0] that specifies the variance of the gaussian filter
34698 used to blur the image (slower if larger). Default value is 1.0.
34699 luma_strength, ls
34701 Set the luma strength. The option value must be a float number in
34702 the range [-1.0,1.0] that configures the blurring. A value included
34703 in [0.0,1.0] will blur the image whereas a value included in
34704 [-1.0,0.0] will sharpen the image. Default value is 1.0.
34705 luma_threshold, lt
34707 Set the luma threshold used as a coefficient to determine whether a
34708 pixel should be blurred or not. The option value must be an integer
34709 in the range [-30,30]. A value of 0 will filter all the image, a
34710 value included in [0,30] will filter flat areas and a value
34711 included in [-30,0] will filter edges. Default value is 0.
34712 chroma_radius, cr
34714 Set the chroma radius. The option value must be a float number in
34715 the range [0.1,5.0] that specifies the variance of the gaussian
34716 filter used to blur the image (slower if larger). Default value is
34717 luma_radius.
34718 chroma_strength, cs
34720 Set the chroma strength. The option value must be a float number in
34721 the range [-1.0,1.0] that configures the blurring. A value included
34722 in [0.0,1.0] will blur the image whereas a value included in
34723 [-1.0,0.0] will sharpen the image. Default value is luma_strength.
34724 chroma_threshold, ct
34726 Set the chroma threshold used as a coefficient to determine whether
34727 a pixel should be blurred or not. The option value must be an
34728 integer in the range [-30,30]. A value of 0 will filter all the
34729 image, a value included in [0,30] will filter flat areas and a
34730 value included in [-30,0] will filter edges. Default value is
34731 luma_threshold.
34732 If a chroma option is not explicitly set, the corresponding luma value
34734 is set.
34735 sobel
34737 Apply sobel operator to input video stream.
34738 The filter accepts the following option:
34740 planes
34742 Set which planes will be processed, unprocessed planes will be
34743 copied. By default value 0xf, all planes will be processed.
34744 scale
34746 Set value which will be multiplied with filtered result.
34747 delta
34749 Set value which will be added to filtered result.
34750 Commands
34752 This filter supports the all above options as commands.
34754 spp
34756 Apply a simple postprocessing filter that compresses and decompresses
34757 the image at several (or - in the case of quality level 6 - all) shifts
34758 and average the results.
34759 The filter accepts the following options:
34761 quality
34763 Set quality. This option defines the number of levels for
34764 averaging. It accepts an integer in the range 0-6. If set to 0, the
34765 filter will have no effect. A value of 6 means the higher quality.
34766 For each increment of that value the speed drops by a factor of
34767 approximately 2. Default value is 3.
34768 qp Force a constant quantization parameter. If not set, the filter
34770 will use the QP from the video stream (if available).
34771 mode
34773 Set thresholding mode. Available modes are:
34774 hard
34776 Set hard thresholding (default).
34777 soft
34779 Set soft thresholding (better de-ringing effect, but likely
34780 blurrier).
34781 use_bframe_qp
34783 Enable the use of the QP from the B-Frames if set to 1. Using this
34784 option may cause flicker since the B-Frames have often larger QP.
34785 Default is 0 (not enabled).
34786 Commands
34788 This filter supports the following commands:
34790 quality, level
34792 Set quality level. The value "max" can be used to set the maximum
34793 level, currently 6.
34794 sr
34796 Scale the input by applying one of the super-resolution methods based
34797 on convolutional neural networks. Supported models:
34798 • Super-Resolution Convolutional Neural Network model (SRCNN). See
34800 <https://arxiv.org/abs/1501.00092>.
34801 • Efficient Sub-Pixel Convolutional Neural Network model (ESPCN).
34803 See <https://arxiv.org/abs/1609.05158>.
34804 Training scripts as well as scripts for model file (.pb) saving can be
34806 found at <https://github.com/XueweiMeng/sr/tree/sr_dnn_native>.
34807 Original repository is at
34808 <https://github.com/HighVoltageRocknRoll/sr.git>.
34809 Native model files (.model) can be generated from TensorFlow model
34811 files (.pb) by using tools/python/convert.py
34812 The filter accepts the following options:
34814 dnn_backend
34816 Specify which DNN backend to use for model loading and execution.
34817 This option accepts the following values:
34818 native
34820 Native implementation of DNN loading and execution.
34821 tensorflow
34823 TensorFlow backend. To enable this backend you need to install
34824 the TensorFlow for C library (see
34825 <https://www.tensorflow.org/install/lang_c>) and configure
34826 FFmpeg with "--enable-libtensorflow"
34827 Default value is native.
34829 model
34831 Set path to model file specifying network architecture and its
34832 parameters. Note that different backends use different file
34833 formats. TensorFlow backend can load files for both formats, while
34834 native backend can load files for only its format.
34835 scale_factor
34837 Set scale factor for SRCNN model. Allowed values are 2, 3 and 4.
34838 Default value is 2. Scale factor is necessary for SRCNN model,
34839 because it accepts input upscaled using bicubic upscaling with
34840 proper scale factor.
34841 To get full functionality (such as async execution), please use the
34843 dnn_processing filter.
34844 ssim
34846 Obtain the SSIM (Structural SImilarity Metric) between two input
34847 videos.
34848 This filter takes in input two input videos, the first input is
34850 considered the "main" source and is passed unchanged to the output. The
34851 second input is used as a "reference" video for computing the SSIM.
34852 Both video inputs must have the same resolution and pixel format for
34854 this filter to work correctly. Also it assumes that both inputs have
34855 the same number of frames, which are compared one by one.
34856 The filter stores the calculated SSIM of each frame.
34858 The description of the accepted parameters follows.
34860 stats_file, f
34862 If specified the filter will use the named file to save the SSIM of
34863 each individual frame. When filename equals "-" the data is sent to
34864 standard output.
34865 The file printed if stats_file is selected, contains a sequence of
34867 key/value pairs of the form key:value for each compared couple of
34868 frames.
34869 A description of each shown parameter follows:
34871 n sequential number of the input frame, starting from 1
34873 Y, U, V, R, G, B
34875 SSIM of the compared frames for the component specified by the
34876 suffix.
34877 All SSIM of the compared frames for the whole frame.
34879 dB Same as above but in dB representation.
34881 This filter also supports the framesync options.
34883 Examples
34885 • For example:
34887 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
34889 [main][ref] ssim="stats_file=stats.log" [out]
34890 On this example the input file being processed is compared with the
34892 reference file ref_movie.mpg. The SSIM of each individual frame is
34893 stored in stats.log.
34894 • Another example with both psnr and ssim at same time:
34896 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
34898 • Another example with different containers:
34900 ffmpeg -i main.mpg -i ref.mkv -lavfi "[0:v]settb=AVTB,setpts=PTS-STARTPTS[main];[1:v]settb=AVTB,setpts=PTS-STARTPTS[ref];[main][ref]ssim" -f null -
34902 stereo3d
34904 Convert between different stereoscopic image formats.
34905 The filters accept the following options:
34907 in Set stereoscopic image format of input.
34909 Available values for input image formats are:
34911 sbsl
34913 side by side parallel (left eye left, right eye right)
34914 sbsr
34916 side by side crosseye (right eye left, left eye right)
34917 sbs2l
34919 side by side parallel with half width resolution (left eye
34920 left, right eye right)
34921 sbs2r
34923 side by side crosseye with half width resolution (right eye
34924 left, left eye right)
34925 abl
34927 tbl above-below (left eye above, right eye below)
34928 abr
34930 tbr above-below (right eye above, left eye below)
34931 ab2l
34933 tb2l
34934 above-below with half height resolution (left eye above, right
34935 eye below)
34936 ab2r
34938 tb2r
34939 above-below with half height resolution (right eye above, left
34940 eye below)
34941 al alternating frames (left eye first, right eye second)
34943 ar alternating frames (right eye first, left eye second)
34945 irl interleaved rows (left eye has top row, right eye starts on
34947 next row)
34948 irr interleaved rows (right eye has top row, left eye starts on
34950 next row)
34951 icl interleaved columns, left eye first
34953 icr interleaved columns, right eye first
34955 Default value is sbsl.
34957 out Set stereoscopic image format of output.
34959 sbsl
34961 side by side parallel (left eye left, right eye right)
34962 sbsr
34964 side by side crosseye (right eye left, left eye right)
34965 sbs2l
34967 side by side parallel with half width resolution (left eye
34968 left, right eye right)
34969 sbs2r
34971 side by side crosseye with half width resolution (right eye
34972 left, left eye right)
34973 abl
34975 tbl above-below (left eye above, right eye below)
34976 abr
34978 tbr above-below (right eye above, left eye below)
34979 ab2l
34981 tb2l
34982 above-below with half height resolution (left eye above, right
34983 eye below)
34984 ab2r
34986 tb2r
34987 above-below with half height resolution (right eye above, left
34988 eye below)
34989 al alternating frames (left eye first, right eye second)
34991 ar alternating frames (right eye first, left eye second)
34993 irl interleaved rows (left eye has top row, right eye starts on
34995 next row)
34996 irr interleaved rows (right eye has top row, left eye starts on
34998 next row)
34999 arbg
35001 anaglyph red/blue gray (red filter on left eye, blue filter on
35002 right eye)
35003 argg
35005 anaglyph red/green gray (red filter on left eye, green filter
35006 on right eye)
35007 arcg
35009 anaglyph red/cyan gray (red filter on left eye, cyan filter on
35010 right eye)
35011 arch
35013 anaglyph red/cyan half colored (red filter on left eye, cyan
35014 filter on right eye)
35015 arcc
35017 anaglyph red/cyan color (red filter on left eye, cyan filter on
35018 right eye)
35019 arcd
35021 anaglyph red/cyan color optimized with the least squares
35022 projection of dubois (red filter on left eye, cyan filter on
35023 right eye)
35024 agmg
35026 anaglyph green/magenta gray (green filter on left eye, magenta
35027 filter on right eye)
35028 agmh
35030 anaglyph green/magenta half colored (green filter on left eye,
35031 magenta filter on right eye)
35032 agmc
35034 anaglyph green/magenta colored (green filter on left eye,
35035 magenta filter on right eye)
35036 agmd
35038 anaglyph green/magenta color optimized with the least squares
35039 projection of dubois (green filter on left eye, magenta filter
35040 on right eye)
35041 aybg
35043 anaglyph yellow/blue gray (yellow filter on left eye, blue
35044 filter on right eye)
35045 aybh
35047 anaglyph yellow/blue half colored (yellow filter on left eye,
35048 blue filter on right eye)
35049 aybc
35051 anaglyph yellow/blue colored (yellow filter on left eye, blue
35052 filter on right eye)
35053 aybd
35055 anaglyph yellow/blue color optimized with the least squares
35056 projection of dubois (yellow filter on left eye, blue filter on
35057 right eye)
35058 ml mono output (left eye only)
35060 mr mono output (right eye only)
35062 chl checkerboard, left eye first
35064 chr checkerboard, right eye first
35066 icl interleaved columns, left eye first
35068 icr interleaved columns, right eye first
35070 hdmi
35072 HDMI frame pack
35073 Default value is arcd.
35075 Examples
35077 • Convert input video from side by side parallel to anaglyph
35079 yellow/blue dubois:
35080 stereo3d=sbsl:aybd
35082 • Convert input video from above below (left eye above, right eye
35084 below) to side by side crosseye.
35085 stereo3d=abl:sbsr
35087 streamselect, astreamselect
35089 Select video or audio streams.
35090 The filter accepts the following options:
35092 inputs
35094 Set number of inputs. Default is 2.
35095 map Set input indexes to remap to outputs.
35097 Commands
35099 The "streamselect" and "astreamselect" filter supports the following
35101 commands:
35102 map Set input indexes to remap to outputs.
35104 Examples
35106 • Select first 5 seconds 1st stream and rest of time 2nd stream:
35108 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
35110 • Same as above, but for audio:
35112 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
35114 subtitles
35116 Draw subtitles on top of input video using the libass library.
35117 To enable compilation of this filter you need to configure FFmpeg with
35119 "--enable-libass". This filter also requires a build with libavcodec
35120 and libavformat to convert the passed subtitles file to ASS (Advanced
35121 Substation Alpha) subtitles format.
35122 The filter accepts the following options:
35124 filename, f
35126 Set the filename of the subtitle file to read. It must be
35127 specified.
35128 original_size
35130 Specify the size of the original video, the video for which the ASS
35131 file was composed. For the syntax of this option, check the "Video
35132 size" section in the ffmpeg-utils manual. Due to a misdesign in
35133 ASS aspect ratio arithmetic, this is necessary to correctly scale
35134 the fonts if the aspect ratio has been changed.
35135 fontsdir
35137 Set a directory path containing fonts that can be used by the
35138 filter. These fonts will be used in addition to whatever the font
35139 provider uses.
35140 alpha
35142 Process alpha channel, by default alpha channel is untouched.
35143 charenc
35145 Set subtitles input character encoding. "subtitles" filter only.
35146 Only useful if not UTF-8.
35147 stream_index, si
35149 Set subtitles stream index. "subtitles" filter only.
35150 force_style
35152 Override default style or script info parameters of the subtitles.
35153 It accepts a string containing ASS style format "KEY=VALUE" couples
35154 separated by ",".
35155 If the first key is not specified, it is assumed that the first value
35157 specifies the filename.
35158 For example, to render the file sub.srt on top of the input video, use
35160 the command:
35161 subtitles=sub.srt
35163 which is equivalent to:
35165 subtitles=filename=sub.srt
35167 To render the default subtitles stream from file video.mkv, use:
35169 subtitles=video.mkv
35171 To render the second subtitles stream from that file, use:
35173 subtitles=video.mkv:si=1
35175 To make the subtitles stream from sub.srt appear in 80% transparent
35177 blue "DejaVu Serif", use:
35178 subtitles=sub.srt:force_style='Fontname=DejaVu Serif,PrimaryColour=&HCCFF0000'
35180 super2xsai
35182 Scale the input by 2x and smooth using the Super2xSaI (Scale and
35183 Interpolate) pixel art scaling algorithm.
35184 Useful for enlarging pixel art images without reducing sharpness.
35186 swaprect
35188 Swap two rectangular objects in video.
35189 This filter accepts the following options:
35191 w Set object width.
35193 h Set object height.
35195 x1 Set 1st rect x coordinate.
35197 y1 Set 1st rect y coordinate.
35199 x2 Set 2nd rect x coordinate.
35201 y2 Set 2nd rect y coordinate.
35203 All expressions are evaluated once for each frame.
35205 The all options are expressions containing the following constants:
35207 w
35209 h The input width and height.
35210 a same as w / h
35212 sar input sample aspect ratio
35214 dar input display aspect ratio, it is the same as (w / h) * sar
35216 n The number of the input frame, starting from 0.
35218 t The timestamp expressed in seconds. It's NAN if the input timestamp
35220 is unknown.
35221 pos the position in the file of the input frame, NAN if unknown
35223 Commands
35225 This filter supports the all above options as commands.
35227 swapuv
35229 Swap U & V plane.
35230 tblend
35232 Blend successive video frames.
35233 See blend
35235 telecine
35237 Apply telecine process to the video.
35238 This filter accepts the following options:
35240 first_field
35242 top, t
35243 top field first
35244 bottom, b
35246 bottom field first The default value is "top".
35247 pattern
35249 A string of numbers representing the pulldown pattern you wish to
35250 apply. The default value is 23.
35251 Some typical patterns:
35253 NTSC output (30i):
35255 27.5p: 32222
35256 24p: 23 (classic)
35257 24p: 2332 (preferred)
35258 20p: 33
35259 18p: 334
35260 16p: 3444
35261 PAL output (25i):
35263 27.5p: 12222
35264 24p: 222222222223 ("Euro pulldown")
35265 16.67p: 33
35266 16p: 33333334
35267 thistogram
35269 Compute and draw a color distribution histogram for the input video
35270 across time.
35271 Unlike histogram video filter which only shows histogram of single
35273 input frame at certain time, this filter shows also past histograms of
35274 number of frames defined by "width" option.
35275 The computed histogram is a representation of the color component
35277 distribution in an image.
35278 The filter accepts the following options:
35280 width, w
35282 Set width of single color component output. Default value is 0.
35283 Value of 0 means width will be picked from input video. This also
35284 set number of passed histograms to keep. Allowed range is [0,
35285 8192].
35286 display_mode, d
35288 Set display mode. It accepts the following values:
35289 stack
35291 Per color component graphs are placed below each other.
35292 parade
35294 Per color component graphs are placed side by side.
35295 overlay
35297 Presents information identical to that in the "parade", except
35298 that the graphs representing color components are superimposed
35299 directly over one another.
35300 Default is "stack".
35302 levels_mode, m
35304 Set mode. Can be either "linear", or "logarithmic". Default is
35305 "linear".
35306 components, c
35308 Set what color components to display. Default is 7.
35309 bgopacity, b
35311 Set background opacity. Default is 0.9.
35312 envelope, e
35314 Show envelope. Default is disabled.
35315 ecolor, ec
35317 Set envelope color. Default is "gold".
35318 slide
35320 Set slide mode.
35321 Available values for slide is:
35323 frame
35325 Draw new frame when right border is reached.
35326 replace
35328 Replace old columns with new ones.
35329 scroll
35331 Scroll from right to left.
35332 rscroll
35334 Scroll from left to right.
35335 picture
35337 Draw single picture.
35338 Default is "replace".
35340 threshold
35342 Apply threshold effect to video stream.
35343 This filter needs four video streams to perform thresholding. First
35345 stream is stream we are filtering. Second stream is holding threshold
35346 values, third stream is holding min values, and last, fourth stream is
35347 holding max values.
35348 The filter accepts the following option:
35350 planes
35352 Set which planes will be processed, unprocessed planes will be
35353 copied. By default value 0xf, all planes will be processed.
35354 For example if first stream pixel's component value is less then
35356 threshold value of pixel component from 2nd threshold stream, third
35357 stream value will picked, otherwise fourth stream pixel component value
35358 will be picked.
35359 Using color source filter one can perform various types of
35361 thresholding:
35362 Commands
35364 This filter supports the all options as commands.
35366 Examples
35368 • Binary threshold, using gray color as threshold:
35370 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=black -f lavfi -i color=white -lavfi threshold output.avi
35372 • Inverted binary threshold, using gray color as threshold:
35374 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -f lavfi -i color=black -lavfi threshold output.avi
35376 • Truncate binary threshold, using gray color as threshold:
35378 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=gray -lavfi threshold output.avi
35380 • Threshold to zero, using gray color as threshold:
35382 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -i 320x240.avi -lavfi threshold output.avi
35384 • Inverted threshold to zero, using gray color as threshold:
35386 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=white -lavfi threshold output.avi
35388 thumbnail
35390 Select the most representative frame in a given sequence of consecutive
35391 frames.
35392 The filter accepts the following options:
35394 n Set the frames batch size to analyze; in a set of n frames, the
35396 filter will pick one of them, and then handle the next batch of n
35397 frames until the end. Default is 100.
35398 Since the filter keeps track of the whole frames sequence, a bigger n
35400 value will result in a higher memory usage, so a high value is not
35401 recommended.
35402 Examples
35404 • Extract one picture each 50 frames:
35406 thumbnail=50
35408 • Complete example of a thumbnail creation with ffmpeg:
35410 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
35412 tile
35414 Tile several successive frames together.
35415 The untile filter can do the reverse.
35417 The filter accepts the following options:
35419 layout
35421 Set the grid size (i.e. the number of lines and columns). For the
35422 syntax of this option, check the "Video size" section in the
35423 ffmpeg-utils manual.
35424 nb_frames
35426 Set the maximum number of frames to render in the given area. It
35427 must be less than or equal to wxh. The default value is 0, meaning
35428 all the area will be used.
35429 margin
35431 Set the outer border margin in pixels.
35432 padding
35434 Set the inner border thickness (i.e. the number of pixels between
35435 frames). For more advanced padding options (such as having
35436 different values for the edges), refer to the pad video filter.
35437 color
35439 Specify the color of the unused area. For the syntax of this
35440 option, check the "Color" section in the ffmpeg-utils manual. The
35441 default value of color is "black".
35442 overlap
35444 Set the number of frames to overlap when tiling several successive
35445 frames together. The value must be between 0 and nb_frames - 1.
35446 init_padding
35448 Set the number of frames to initially be empty before displaying
35449 first output frame. This controls how soon will one get first
35450 output frame. The value must be between 0 and nb_frames - 1.
35451 Examples
35453 • Produce 8x8 PNG tiles of all keyframes (-skip_frame nokey) in a
35455 movie:
35456 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
35458 The -vsync 0 is necessary to prevent ffmpeg from duplicating each
35460 output frame to accommodate the originally detected frame rate.
35461 • Display 5 pictures in an area of "3x2" frames, with 7 pixels
35463 between them, and 2 pixels of initial margin, using mixed flat and
35464 named options:
35465 tile=3x2:nb_frames=5:padding=7:margin=2
35467 tinterlace
35469 Perform various types of temporal field interlacing.
35470 Frames are counted starting from 1, so the first input frame is
35472 considered odd.
35473 The filter accepts the following options:
35475 mode
35477 Specify the mode of the interlacing. This option can also be
35478 specified as a value alone. See below for a list of values for this
35479 option.
35480 Available values are:
35482 merge, 0
35484 Move odd frames into the upper field, even into the lower
35485 field, generating a double height frame at half frame rate.
35486 ------> time
35488 Input:
35489 Frame 1 Frame 2 Frame 3 Frame 4
35490 11111 22222 33333 44444
35492 11111 22222 33333 44444
35493 11111 22222 33333 44444
35494 11111 22222 33333 44444
35495 Output:
35497 11111 33333
35498 22222 44444
35499 11111 33333
35500 22222 44444
35501 11111 33333
35502 22222 44444
35503 11111 33333
35504 22222 44444
35505 drop_even, 1
35507 Only output odd frames, even frames are dropped, generating a
35508 frame with unchanged height at half frame rate.
35509 ------> time
35511 Input:
35512 Frame 1 Frame 2 Frame 3 Frame 4
35513 11111 22222 33333 44444
35515 11111 22222 33333 44444
35516 11111 22222 33333 44444
35517 11111 22222 33333 44444
35518 Output:
35520 11111 33333
35521 11111 33333
35522 11111 33333
35523 11111 33333
35524 drop_odd, 2
35526 Only output even frames, odd frames are dropped, generating a
35527 frame with unchanged height at half frame rate.
35528 ------> time
35530 Input:
35531 Frame 1 Frame 2 Frame 3 Frame 4
35532 11111 22222 33333 44444
35534 11111 22222 33333 44444
35535 11111 22222 33333 44444
35536 11111 22222 33333 44444
35537 Output:
35539 22222 44444
35540 22222 44444
35541 22222 44444
35542 22222 44444
35543 pad, 3
35545 Expand each frame to full height, but pad alternate lines with
35546 black, generating a frame with double height at the same input
35547 frame rate.
35548 ------> time
35550 Input:
35551 Frame 1 Frame 2 Frame 3 Frame 4
35552 11111 22222 33333 44444
35554 11111 22222 33333 44444
35555 11111 22222 33333 44444
35556 11111 22222 33333 44444
35557 Output:
35559 11111 ..... 33333 .....
35560 ..... 22222 ..... 44444
35561 11111 ..... 33333 .....
35562 ..... 22222 ..... 44444
35563 11111 ..... 33333 .....
35564 ..... 22222 ..... 44444
35565 11111 ..... 33333 .....
35566 ..... 22222 ..... 44444
35567 interleave_top, 4
35569 Interleave the upper field from odd frames with the lower field
35570 from even frames, generating a frame with unchanged height at
35571 half frame rate.
35572 ------> time
35574 Input:
35575 Frame 1 Frame 2 Frame 3 Frame 4
35576 11111<- 22222 33333<- 44444
35578 11111 22222<- 33333 44444<-
35579 11111<- 22222 33333<- 44444
35580 11111 22222<- 33333 44444<-
35581 Output:
35583 11111 33333
35584 22222 44444
35585 11111 33333
35586 22222 44444
35587 interleave_bottom, 5
35589 Interleave the lower field from odd frames with the upper field
35590 from even frames, generating a frame with unchanged height at
35591 half frame rate.
35592 ------> time
35594 Input:
35595 Frame 1 Frame 2 Frame 3 Frame 4
35596 11111 22222<- 33333 44444<-
35598 11111<- 22222 33333<- 44444
35599 11111 22222<- 33333 44444<-
35600 11111<- 22222 33333<- 44444
35601 Output:
35603 22222 44444
35604 11111 33333
35605 22222 44444
35606 11111 33333
35607 interlacex2, 6
35609 Double frame rate with unchanged height. Frames are inserted
35610 each containing the second temporal field from the previous
35611 input frame and the first temporal field from the next input
35612 frame. This mode relies on the top_field_first flag. Useful for
35613 interlaced video displays with no field synchronisation.
35614 ------> time
35616 Input:
35617 Frame 1 Frame 2 Frame 3 Frame 4
35618 11111 22222 33333 44444
35620 11111 22222 33333 44444
35621 11111 22222 33333 44444
35622 11111 22222 33333 44444
35623 Output:
35625 11111 22222 22222 33333 33333 44444 44444
35626 11111 11111 22222 22222 33333 33333 44444
35627 11111 22222 22222 33333 33333 44444 44444
35628 11111 11111 22222 22222 33333 33333 44444
35629 mergex2, 7
35631 Move odd frames into the upper field, even into the lower
35632 field, generating a double height frame at same frame rate.
35633 ------> time
35635 Input:
35636 Frame 1 Frame 2 Frame 3 Frame 4
35637 11111 22222 33333 44444
35639 11111 22222 33333 44444
35640 11111 22222 33333 44444
35641 11111 22222 33333 44444
35642 Output:
35644 11111 33333 33333 55555
35645 22222 22222 44444 44444
35646 11111 33333 33333 55555
35647 22222 22222 44444 44444
35648 11111 33333 33333 55555
35649 22222 22222 44444 44444
35650 11111 33333 33333 55555
35651 22222 22222 44444 44444
35652 Numeric values are deprecated but are accepted for backward
35654 compatibility reasons.
35655 Default mode is "merge".
35657 flags
35659 Specify flags influencing the filter process.
35660 Available value for flags is:
35662 low_pass_filter, vlpf
35664 Enable linear vertical low-pass filtering in the filter.
35665 Vertical low-pass filtering is required when creating an
35666 interlaced destination from a progressive source which contains
35667 high-frequency vertical detail. Filtering will reduce interlace
35668 'twitter' and Moire patterning.
35669 complex_filter, cvlpf
35671 Enable complex vertical low-pass filtering. This will slightly
35672 less reduce interlace 'twitter' and Moire patterning but better
35673 retain detail and subjective sharpness impression.
35674 bypass_il
35676 Bypass already interlaced frames, only adjust the frame rate.
35677 Vertical low-pass filtering and bypassing already interlaced frames
35679 can only be enabled for mode interleave_top and interleave_bottom.
35680 tmedian
35682 Pick median pixels from several successive input video frames.
35683 The filter accepts the following options:
35685 radius
35687 Set radius of median filter. Default is 1. Allowed range is from 1
35688 to 127.
35689 planes
35691 Set which planes to filter. Default value is 15, by which all
35692 planes are processed.
35693 percentile
35695 Set median percentile. Default value is 0.5. Default value of 0.5
35696 will pick always median values, while 0 will pick minimum values,
35697 and 1 maximum values.
35698 Commands
35700 This filter supports all above options as commands, excluding option
35702 "radius".
35703 tmidequalizer
35705 Apply Temporal Midway Video Equalization effect.
35706 Midway Video Equalization adjusts a sequence of video frames to have
35708 the same histograms, while maintaining their dynamics as much as
35709 possible. It's useful for e.g. matching exposures from a video frames
35710 sequence.
35711 This filter accepts the following option:
35713 radius
35715 Set filtering radius. Default is 5. Allowed range is from 1 to 127.
35716 sigma
35718 Set filtering sigma. Default is 0.5. This controls strength of
35719 filtering. Setting this option to 0 effectively does nothing.
35720 planes
35722 Set which planes to process. Default is 15, which is all available
35723 planes.
35724 tmix
35726 Mix successive video frames.
35727 A description of the accepted options follows.
35729 frames
35731 The number of successive frames to mix. If unspecified, it defaults
35732 to 3.
35733 weights
35735 Specify weight of each input video frame. Each weight is separated
35736 by space. If number of weights is smaller than number of frames
35737 last specified weight will be used for all remaining unset weights.
35738 scale
35740 Specify scale, if it is set it will be multiplied with sum of each
35741 weight multiplied with pixel values to give final destination pixel
35742 value. By default scale is auto scaled to sum of weights.
35743 Examples
35745 • Average 7 successive frames:
35747 tmix=frames=7:weights="1 1 1 1 1 1 1"
35749 • Apply simple temporal convolution:
35751 tmix=frames=3:weights="-1 3 -1"
35753 • Similar as above but only showing temporal differences:
35755 tmix=frames=3:weights="-1 2 -1":scale=1
35757 Commands
35759 This filter supports the following commands:
35761 weights
35763 scale
35764 Syntax is same as option with same name.
35765 tonemap
35767 Tone map colors from different dynamic ranges.
35768 This filter expects data in single precision floating point, as it
35770 needs to operate on (and can output) out-of-range values. Another
35771 filter, such as zscale, is needed to convert the resulting frame to a
35772 usable format.
35773 The tonemapping algorithms implemented only work on linear light, so
35775 input data should be linearized beforehand (and possibly correctly
35776 tagged).
35777 ffmpeg -i INPUT -vf zscale=transfer=linear,tonemap=clip,zscale=transfer=bt709,format=yuv420p OUTPUT
35779 Options
35781 The filter accepts the following options.
35783 tonemap
35785 Set the tone map algorithm to use.
35786 Possible values are:
35788 none
35790 Do not apply any tone map, only desaturate overbright pixels.
35791 clip
35793 Hard-clip any out-of-range values. Use it for perfect color
35794 accuracy for in-range values, while distorting out-of-range
35795 values.
35796 linear
35798 Stretch the entire reference gamut to a linear multiple of the
35799 display.
35800 gamma
35802 Fit a logarithmic transfer between the tone curves.
35803 reinhard
35805 Preserve overall image brightness with a simple curve, using
35806 nonlinear contrast, which results in flattening details and
35807 degrading color accuracy.
35808 hable
35810 Preserve both dark and bright details better than reinhard, at
35811 the cost of slightly darkening everything. Use it when detail
35812 preservation is more important than color and brightness
35813 accuracy.
35814 mobius
35816 Smoothly map out-of-range values, while retaining contrast and
35817 colors for in-range material as much as possible. Use it when
35818 color accuracy is more important than detail preservation.
35819 Default is none.
35821 param
35823 Tune the tone mapping algorithm.
35824 This affects the following algorithms:
35826 none
35828 Ignored.
35829 linear
35831 Specifies the scale factor to use while stretching. Default to
35832 1.0.
35833 gamma
35835 Specifies the exponent of the function. Default to 1.8.
35836 clip
35838 Specify an extra linear coefficient to multiply into the signal
35839 before clipping. Default to 1.0.
35840 reinhard
35842 Specify the local contrast coefficient at the display peak.
35843 Default to 0.5, which means that in-gamut values will be about
35844 half as bright as when clipping.
35845 hable
35847 Ignored.
35848 mobius
35850 Specify the transition point from linear to mobius transform.
35851 Every value below this point is guaranteed to be mapped 1:1.
35852 The higher the value, the more accurate the result will be, at
35853 the cost of losing bright details. Default to 0.3, which due
35854 to the steep initial slope still preserves in-range colors
35855 fairly accurately.
35856 desat
35858 Apply desaturation for highlights that exceed this level of
35859 brightness. The higher the parameter, the more color information
35860 will be preserved. This setting helps prevent unnaturally blown-out
35861 colors for super-highlights, by (smoothly) turning into white
35862 instead. This makes images feel more natural, at the cost of
35863 reducing information about out-of-range colors.
35864 The default of 2.0 is somewhat conservative and will mostly just
35866 apply to skies or directly sunlit surfaces. A setting of 0.0
35867 disables this option.
35868 This option works only if the input frame has a supported color
35870 tag.
35871 peak
35873 Override signal/nominal/reference peak with this value. Useful when
35874 the embedded peak information in display metadata is not reliable
35875 or when tone mapping from a lower range to a higher range.
35876 tpad
35878 Temporarily pad video frames.
35879 The filter accepts the following options:
35881 start
35883 Specify number of delay frames before input video stream. Default
35884 is 0.
35885 stop
35887 Specify number of padding frames after input video stream. Set to
35888 -1 to pad indefinitely. Default is 0.
35889 start_mode
35891 Set kind of frames added to beginning of stream. Can be either add
35892 or clone. With add frames of solid-color are added. With clone
35893 frames are clones of first frame. Default is add.
35894 stop_mode
35896 Set kind of frames added to end of stream. Can be either add or
35897 clone. With add frames of solid-color are added. With clone
35898 frames are clones of last frame. Default is add.
35899 start_duration, stop_duration
35901 Specify the duration of the start/stop delay. See the Time duration
35902 section in the ffmpeg-utils(1) manual for the accepted syntax.
35903 These options override start and stop. Default is 0.
35904 color
35906 Specify the color of the padded area. For the syntax of this
35907 option, check the "Color" section in the ffmpeg-utils manual.
35908 The default value of color is "black".
35910 transpose
35912 Transpose rows with columns in the input video and optionally flip it.
35913 It accepts the following parameters:
35915 dir Specify the transposition direction.
35917 Can assume the following values:
35919 0, 4, cclock_flip
35921 Rotate by 90 degrees counterclockwise and vertically flip
35922 (default), that is:
35923 L.R L.l
35925 . . -> . .
35926 l.r R.r
35927 1, 5, clock
35929 Rotate by 90 degrees clockwise, that is:
35930 L.R l.L
35932 . . -> . .
35933 l.r r.R
35934 2, 6, cclock
35936 Rotate by 90 degrees counterclockwise, that is:
35937 L.R R.r
35939 . . -> . .
35940 l.r L.l
35941 3, 7, clock_flip
35943 Rotate by 90 degrees clockwise and vertically flip, that is:
35944 L.R r.R
35946 . . -> . .
35947 l.r l.L
35948 For values between 4-7, the transposition is only done if the input
35950 video geometry is portrait and not landscape. These values are
35951 deprecated, the "passthrough" option should be used instead.
35952 Numerical values are deprecated, and should be dropped in favor of
35954 symbolic constants.
35955 passthrough
35957 Do not apply the transposition if the input geometry matches the
35958 one specified by the specified value. It accepts the following
35959 values:
35960 none
35962 Always apply transposition.
35963 portrait
35965 Preserve portrait geometry (when height >= width).
35966 landscape
35968 Preserve landscape geometry (when width >= height).
35969 Default value is "none".
35971 For example to rotate by 90 degrees clockwise and preserve portrait
35973 layout:
35974 transpose=dir=1:passthrough=portrait
35976 The command above can also be specified as:
35978 transpose=1:portrait
35980 transpose_npp
35982 Transpose rows with columns in the input video and optionally flip it.
35983 For more in depth examples see the transpose video filter, which shares
35984 mostly the same options.
35985 It accepts the following parameters:
35987 dir Specify the transposition direction.
35989 Can assume the following values:
35991 cclock_flip
35993 Rotate by 90 degrees counterclockwise and vertically flip.
35994 (default)
35995 clock
35997 Rotate by 90 degrees clockwise.
35998 cclock
36000 Rotate by 90 degrees counterclockwise.
36001 clock_flip
36003 Rotate by 90 degrees clockwise and vertically flip.
36004 passthrough
36006 Do not apply the transposition if the input geometry matches the
36007 one specified by the specified value. It accepts the following
36008 values:
36009 none
36011 Always apply transposition. (default)
36012 portrait
36014 Preserve portrait geometry (when height >= width).
36015 landscape
36017 Preserve landscape geometry (when width >= height).
36018 trim
36020 Trim the input so that the output contains one continuous subpart of
36021 the input.
36022 It accepts the following parameters:
36024 start
36026 Specify the time of the start of the kept section, i.e. the frame
36027 with the timestamp start will be the first frame in the output.
36028 end Specify the time of the first frame that will be dropped, i.e. the
36030 frame immediately preceding the one with the timestamp end will be
36031 the last frame in the output.
36032 start_pts
36034 This is the same as start, except this option sets the start
36035 timestamp in timebase units instead of seconds.
36036 end_pts
36038 This is the same as end, except this option sets the end timestamp
36039 in timebase units instead of seconds.
36040 duration
36042 The maximum duration of the output in seconds.
36043 start_frame
36045 The number of the first frame that should be passed to the output.
36046 end_frame
36048 The number of the first frame that should be dropped.
36049 start, end, and duration are expressed as time duration specifications;
36051 see the Time duration section in the ffmpeg-utils(1) manual for the
36052 accepted syntax.
36053 Note that the first two sets of the start/end options and the duration
36055 option look at the frame timestamp, while the _frame variants simply
36056 count the frames that pass through the filter. Also note that this
36057 filter does not modify the timestamps. If you wish for the output
36058 timestamps to start at zero, insert a setpts filter after the trim
36059 filter.
36060 If multiple start or end options are set, this filter tries to be
36062 greedy and keep all the frames that match at least one of the specified
36063 constraints. To keep only the part that matches all the constraints at
36064 once, chain multiple trim filters.
36065 The defaults are such that all the input is kept. So it is possible to
36067 set e.g. just the end values to keep everything before the specified
36068 time.
36069 Examples:
36071 • Drop everything except the second minute of input:
36073 ffmpeg -i INPUT -vf trim=60:120
36075 • Keep only the first second:
36077 ffmpeg -i INPUT -vf trim=duration=1
36079 unpremultiply
36081 Apply alpha unpremultiply effect to input video stream using first
36082 plane of second stream as alpha.
36083 Both streams must have same dimensions and same pixel format.
36085 The filter accepts the following option:
36087 planes
36089 Set which planes will be processed, unprocessed planes will be
36090 copied. By default value 0xf, all planes will be processed.
36091 If the format has 1 or 2 components, then luma is bit 0. If the
36093 format has 3 or 4 components: for RGB formats bit 0 is green, bit 1
36094 is blue and bit 2 is red; for YUV formats bit 0 is luma, bit 1 is
36095 chroma-U and bit 2 is chroma-V. If present, the alpha channel is
36096 always the last bit.
36097 inplace
36099 Do not require 2nd input for processing, instead use alpha plane
36100 from input stream.
36101 unsharp
36103 Sharpen or blur the input video.
36104 It accepts the following parameters:
36106 luma_msize_x, lx
36108 Set the luma matrix horizontal size. It must be an odd integer
36109 between 3 and 23. The default value is 5.
36110 luma_msize_y, ly
36112 Set the luma matrix vertical size. It must be an odd integer
36113 between 3 and 23. The default value is 5.
36114 luma_amount, la
36116 Set the luma effect strength. It must be a floating point number,
36117 reasonable values lay between -1.5 and 1.5.
36118 Negative values will blur the input video, while positive values
36120 will sharpen it, a value of zero will disable the effect.
36121 Default value is 1.0.
36123 chroma_msize_x, cx
36125 Set the chroma matrix horizontal size. It must be an odd integer
36126 between 3 and 23. The default value is 5.
36127 chroma_msize_y, cy
36129 Set the chroma matrix vertical size. It must be an odd integer
36130 between 3 and 23. The default value is 5.
36131 chroma_amount, ca
36133 Set the chroma effect strength. It must be a floating point number,
36134 reasonable values lay between -1.5 and 1.5.
36135 Negative values will blur the input video, while positive values
36137 will sharpen it, a value of zero will disable the effect.
36138 Default value is 0.0.
36140 All parameters are optional and default to the equivalent of the string
36142 '5:5:1.0:5:5:0.0'.
36143 Examples
36145 • Apply strong luma sharpen effect:
36147 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
36149 • Apply a strong blur of both luma and chroma parameters:
36151 unsharp=7:7:-2:7:7:-2
36153 untile
36155 Decompose a video made of tiled images into the individual images.
36156 The frame rate of the output video is the frame rate of the input video
36158 multiplied by the number of tiles.
36159 This filter does the reverse of tile.
36161 The filter accepts the following options:
36163 layout
36165 Set the grid size (i.e. the number of lines and columns). For the
36166 syntax of this option, check the "Video size" section in the
36167 ffmpeg-utils manual.
36168 Examples
36170 • Produce a 1-second video from a still image file made of 25 frames
36172 stacked vertically, like an analogic film reel:
36173 ffmpeg -r 1 -i image.jpg -vf untile=1x25 movie.mkv
36175 uspp
36177 Apply ultra slow/simple postprocessing filter that compresses and
36178 decompresses the image at several (or - in the case of quality level 8
36179 - all) shifts and average the results.
36180 The way this differs from the behavior of spp is that uspp actually
36182 encodes & decodes each case with libavcodec Snow, whereas spp uses a
36183 simplified intra only 8x8 DCT similar to MJPEG.
36184 This filter is only available in ffmpeg version 4.4 or earlier.
36186 The filter accepts the following options:
36188 quality
36190 Set quality. This option defines the number of levels for
36191 averaging. It accepts an integer in the range 0-8. If set to 0, the
36192 filter will have no effect. A value of 8 means the higher quality.
36193 For each increment of that value the speed drops by a factor of
36194 approximately 2. Default value is 3.
36195 qp Force a constant quantization parameter. If not set, the filter
36197 will use the QP from the video stream (if available).
36198 v360
36200 Convert 360 videos between various formats.
36201 The filter accepts the following options:
36203 input
36205 output
36206 Set format of the input/output video.
36207 Available formats:
36209 e
36211 equirect
36212 Equirectangular projection.
36213 c3x2
36215 c6x1
36216 c1x6
36217 Cubemap with 3x2/6x1/1x6 layout.
36218 Format specific options:
36220 in_pad
36222 out_pad
36223 Set padding proportion for the input/output cubemap. Values
36224 in decimals.
36225 Example values:
36227 0 No padding.
36229 0.01
36231 1% of face is padding. For example, with 1920x1280
36232 resolution face size would be 640x640 and padding would
36233 be 3 pixels from each side. (640 * 0.01 = 6 pixels)
36234 Default value is @samp{0}. Maximum value is @samp{0.1}.
36236 fin_pad
36238 fout_pad
36239 Set fixed padding for the input/output cubemap. Values in
36240 pixels.
36241 Default value is @samp{0}. If greater than zero it
36243 overrides other padding options.
36244 in_forder
36246 out_forder
36247 Set order of faces for the input/output cubemap. Choose one
36248 direction for each position.
36249 Designation of directions:
36251 r right
36253 l left
36255 u up
36257 d down
36259 f forward
36261 b back
36263 Default value is @samp{rludfb}.
36265 in_frot
36267 out_frot
36268 Set rotation of faces for the input/output cubemap. Choose
36269 one angle for each position.
36270 Designation of angles:
36272 0 0 degrees clockwise
36274 1 90 degrees clockwise
36276 2 180 degrees clockwise
36278 3 270 degrees clockwise
36280 Default value is @samp{000000}.
36282 eac Equi-Angular Cubemap.
36284 flat
36286 gnomonic
36287 rectilinear
36288 Regular video.
36289 Format specific options:
36291 h_fov
36293 v_fov
36294 d_fov
36295 Set output horizontal/vertical/diagonal field of view.
36296 Values in degrees.
36297 If diagonal field of view is set it overrides horizontal
36299 and vertical field of view.
36300 ih_fov
36302 iv_fov
36303 id_fov
36304 Set input horizontal/vertical/diagonal field of view.
36305 Values in degrees.
36306 If diagonal field of view is set it overrides horizontal
36308 and vertical field of view.
36309 dfisheye
36311 Dual fisheye.
36312 Format specific options:
36314 h_fov
36316 v_fov
36317 d_fov
36318 Set output horizontal/vertical/diagonal field of view.
36319 Values in degrees.
36320 If diagonal field of view is set it overrides horizontal
36322 and vertical field of view.
36323 ih_fov
36325 iv_fov
36326 id_fov
36327 Set input horizontal/vertical/diagonal field of view.
36328 Values in degrees.
36329 If diagonal field of view is set it overrides horizontal
36331 and vertical field of view.
36332 barrel
36334 fb
36335 barrelsplit
36336 Facebook's 360 formats.
36337 sg Stereographic format.
36339 Format specific options:
36341 h_fov
36343 v_fov
36344 d_fov
36345 Set output horizontal/vertical/diagonal field of view.
36346 Values in degrees.
36347 If diagonal field of view is set it overrides horizontal
36349 and vertical field of view.
36350 ih_fov
36352 iv_fov
36353 id_fov
36354 Set input horizontal/vertical/diagonal field of view.
36355 Values in degrees.
36356 If diagonal field of view is set it overrides horizontal
36358 and vertical field of view.
36359 mercator
36361 Mercator format.
36362 ball
36364 Ball format, gives significant distortion toward the back.
36365 hammer
36367 Hammer-Aitoff map projection format.
36368 sinusoidal
36370 Sinusoidal map projection format.
36371 fisheye
36373 Fisheye projection.
36374 Format specific options:
36376 h_fov
36378 v_fov
36379 d_fov
36380 Set output horizontal/vertical/diagonal field of view.
36381 Values in degrees.
36382 If diagonal field of view is set it overrides horizontal
36384 and vertical field of view.
36385 ih_fov
36387 iv_fov
36388 id_fov
36389 Set input horizontal/vertical/diagonal field of view.
36390 Values in degrees.
36391 If diagonal field of view is set it overrides horizontal
36393 and vertical field of view.
36394 pannini
36396 Pannini projection.
36397 Format specific options:
36399 h_fov
36401 Set output pannini parameter.
36402 ih_fov
36404 Set input pannini parameter.
36405 cylindrical
36407 Cylindrical projection.
36408 Format specific options:
36410 h_fov
36412 v_fov
36413 d_fov
36414 Set output horizontal/vertical/diagonal field of view.
36415 Values in degrees.
36416 If diagonal field of view is set it overrides horizontal
36418 and vertical field of view.
36419 ih_fov
36421 iv_fov
36422 id_fov
36423 Set input horizontal/vertical/diagonal field of view.
36424 Values in degrees.
36425 If diagonal field of view is set it overrides horizontal
36427 and vertical field of view.
36428 perspective
36430 Perspective projection. (output only)
36431 Format specific options:
36433 v_fov
36435 Set perspective parameter.
36436 tetrahedron
36438 Tetrahedron projection.
36439 tsp Truncated square pyramid projection.
36441 he
36443 hequirect
36444 Half equirectangular projection.
36445 equisolid
36447 Equisolid format.
36448 Format specific options:
36450 h_fov
36452 v_fov
36453 d_fov
36454 Set output horizontal/vertical/diagonal field of view.
36455 Values in degrees.
36456 If diagonal field of view is set it overrides horizontal
36458 and vertical field of view.
36459 ih_fov
36461 iv_fov
36462 id_fov
36463 Set input horizontal/vertical/diagonal field of view.
36464 Values in degrees.
36465 If diagonal field of view is set it overrides horizontal
36467 and vertical field of view.
36468 og Orthographic format.
36470 Format specific options:
36472 h_fov
36474 v_fov
36475 d_fov
36476 Set output horizontal/vertical/diagonal field of view.
36477 Values in degrees.
36478 If diagonal field of view is set it overrides horizontal
36480 and vertical field of view.
36481 ih_fov
36483 iv_fov
36484 id_fov
36485 Set input horizontal/vertical/diagonal field of view.
36486 Values in degrees.
36487 If diagonal field of view is set it overrides horizontal
36489 and vertical field of view.
36490 octahedron
36492 Octahedron projection.
36493 cylindricalea
36495 Cylindrical Equal Area projection.
36496 interp
36498 Set interpolation method.Note: more complex interpolation methods
36499 require much more memory to run.
36500 Available methods:
36502 near
36504 nearest
36505 Nearest neighbour.
36506 line
36508 linear
36509 Bilinear interpolation.
36510 lagrange9
36512 Lagrange9 interpolation.
36513 cube
36515 cubic
36516 Bicubic interpolation.
36517 lanc
36519 lanczos
36520 Lanczos interpolation.
36521 sp16
36523 spline16
36524 Spline16 interpolation.
36525 gauss
36527 gaussian
36528 Gaussian interpolation.
36529 mitchell
36531 Mitchell interpolation.
36532 Default value is @samp{line}.
36534 w
36536 h Set the output video resolution.
36537 Default resolution depends on formats.
36539 in_stereo
36541 out_stereo
36542 Set the input/output stereo format.
36543 2d 2D mono
36545 sbs Side by side
36547 tb Top bottom
36549 Default value is @samp{2d} for input and output format.
36551 yaw
36553 pitch
36554 roll
36555 Set rotation for the output video. Values in degrees.
36556 rorder
36558 Set rotation order for the output video. Choose one item for each
36559 position.
36560 y, Y
36562 yaw
36563 p, P
36565 pitch
36566 r, R
36568 roll
36569 Default value is @samp{ypr}.
36571 h_flip
36573 v_flip
36574 d_flip
36575 Flip the output video horizontally(swaps
36576 left-right)/vertically(swaps up-down)/in-depth(swaps back-forward).
36577 Boolean values.
36578 ih_flip
36580 iv_flip
36581 Set if input video is flipped horizontally/vertically. Boolean
36582 values.
36583 in_trans
36585 Set if input video is transposed. Boolean value, by default
36586 disabled.
36587 out_trans
36589 Set if output video needs to be transposed. Boolean value, by
36590 default disabled.
36591 h_offset
36593 v_offset
36594 Set output horizontal/vertical off-axis offset. Default is set to
36595 0. Allowed range is from -1 to 1.
36596 alpha_mask
36598 Build mask in alpha plane for all unmapped pixels by marking them
36599 fully transparent. Boolean value, by default disabled.
36600 reset_rot
36602 Reset rotation of output video. Boolean value, by default disabled.
36603 Examples
36605 • Convert equirectangular video to cubemap with 3x2 layout and 1%
36607 padding using bicubic interpolation:
36608 ffmpeg -i input.mkv -vf v360=e:c3x2:cubic:out_pad=0.01 output.mkv
36610 • Extract back view of Equi-Angular Cubemap:
36612 ffmpeg -i input.mkv -vf v360=eac:flat:yaw=180 output.mkv
36614 • Convert transposed and horizontally flipped Equi-Angular Cubemap in
36616 side-by-side stereo format to equirectangular top-bottom stereo
36617 format:
36618 v360=eac:equirect:in_stereo=sbs:in_trans=1:ih_flip=1:out_stereo=tb
36620 Commands
36622 This filter supports subset of above options as commands.
36624 vaguedenoiser
36626 Apply a wavelet based denoiser.
36627 It transforms each frame from the video input into the wavelet domain,
36629 using Cohen-Daubechies-Feauveau 9/7. Then it applies some filtering to
36630 the obtained coefficients. It does an inverse wavelet transform after.
36631 Due to wavelet properties, it should give a nice smoothed result, and
36632 reduced noise, without blurring picture features.
36633 This filter accepts the following options:
36635 threshold
36637 The filtering strength. The higher, the more filtered the video
36638 will be. Hard thresholding can use a higher threshold than soft
36639 thresholding before the video looks overfiltered. Default value is
36640 2.
36641 method
36643 The filtering method the filter will use.
36644 It accepts the following values:
36646 hard
36648 All values under the threshold will be zeroed.
36649 soft
36651 All values under the threshold will be zeroed. All values above
36652 will be reduced by the threshold.
36653 garrote
36655 Scales or nullifies coefficients - intermediary between (more)
36656 soft and (less) hard thresholding.
36657 Default is garrote.
36659 nsteps
36661 Number of times, the wavelet will decompose the picture. Picture
36662 can't be decomposed beyond a particular point (typically, 8 for a
36663 640x480 frame - as 2^9 = 512 > 480). Valid values are integers
36664 between 1 and 32. Default value is 6.
36665 percent
36667 Partial of full denoising (limited coefficients shrinking), from 0
36668 to 100. Default value is 85.
36669 planes
36671 A list of the planes to process. By default all planes are
36672 processed.
36673 type
36675 The threshold type the filter will use.
36676 It accepts the following values:
36678 universal
36680 Threshold used is same for all decompositions.
36681 bayes
36683 Threshold used depends also on each decomposition coefficients.
36684 Default is universal.
36686 varblur
36688 Apply variable blur filter by using 2nd video stream to set blur
36689 radius. The 2nd stream must have the same dimensions.
36690 This filter accepts the following options:
36692 min_r
36694 Set min allowed radius. Allowed range is from 0 to 254. Default is
36695 0.
36696 max_r
36698 Set max allowed radius. Allowed range is from 1 to 255. Default is
36699 8.
36700 planes
36702 Set which planes to process. By default, all are used.
36703 The "varblur" filter also supports the framesync options.
36705 Commands
36707 This filter supports all the above options as commands.
36709 vectorscope
36711 Display 2 color component values in the two dimensional graph (which is
36712 called a vectorscope).
36713 This filter accepts the following options:
36715 mode, m
36717 Set vectorscope mode.
36718 It accepts the following values:
36720 gray
36722 tint
36723 Gray values are displayed on graph, higher brightness means
36724 more pixels have same component color value on location in
36725 graph. This is the default mode.
36726 color
36728 Gray values are displayed on graph. Surrounding pixels values
36729 which are not present in video frame are drawn in gradient of 2
36730 color components which are set by option "x" and "y". The 3rd
36731 color component is static.
36732 color2
36734 Actual color components values present in video frame are
36735 displayed on graph.
36736 color3
36738 Similar as color2 but higher frequency of same values "x" and
36739 "y" on graph increases value of another color component, which
36740 is luminance by default values of "x" and "y".
36741 color4
36743 Actual colors present in video frame are displayed on graph. If
36744 two different colors map to same position on graph then color
36745 with higher value of component not present in graph is picked.
36746 color5
36748 Gray values are displayed on graph. Similar to "color" but with
36749 3rd color component picked from radial gradient.
36750 x Set which color component will be represented on X-axis. Default is
36752 1.
36753 y Set which color component will be represented on Y-axis. Default is
36755 2.
36756 intensity, i
36758 Set intensity, used by modes: gray, color, color3 and color5 for
36759 increasing brightness of color component which represents frequency
36760 of (X, Y) location in graph.
36761 envelope, e
36763 none
36764 No envelope, this is default.
36765 instant
36767 Instant envelope, even darkest single pixel will be clearly
36768 highlighted.
36769 peak
36771 Hold maximum and minimum values presented in graph over time.
36772 This way you can still spot out of range values without
36773 constantly looking at vectorscope.
36774 peak+instant
36776 Peak and instant envelope combined together.
36777 graticule, g
36779 Set what kind of graticule to draw.
36780 none
36782 green
36783 color
36784 invert
36785 opacity, o
36786 Set graticule opacity.
36787 flags, f
36789 Set graticule flags.
36790 white
36792 Draw graticule for white point.
36793 black
36795 Draw graticule for black point.
36796 name
36798 Draw color points short names.
36799 bgopacity, b
36801 Set background opacity.
36802 lthreshold, l
36804 Set low threshold for color component not represented on X or Y
36805 axis. Values lower than this value will be ignored. Default is 0.
36806 Note this value is multiplied with actual max possible value one
36807 pixel component can have. So for 8-bit input and low threshold
36808 value of 0.1 actual threshold is 0.1 * 255 = 25.
36809 hthreshold, h
36811 Set high threshold for color component not represented on X or Y
36812 axis. Values higher than this value will be ignored. Default is 1.
36813 Note this value is multiplied with actual max possible value one
36814 pixel component can have. So for 8-bit input and high threshold
36815 value of 0.9 actual threshold is 0.9 * 255 = 230.
36816 colorspace, c
36818 Set what kind of colorspace to use when drawing graticule.
36819 auto
36821 601
36822 709
36823 Default is auto.
36825 tint0, t0
36827 tint1, t1
36828 Set color tint for gray/tint vectorscope mode. By default both
36829 options are zero. This means no tint, and output will remain gray.
36830 vidstabdetect
36832 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
36833 vidstabtransform for pass 2.
36834 This filter generates a file with relative translation and rotation
36836 transform information about subsequent frames, which is then used by
36837 the vidstabtransform filter.
36838 To enable compilation of this filter you need to configure FFmpeg with
36840 "--enable-libvidstab".
36841 This filter accepts the following options:
36843 result
36845 Set the path to the file used to write the transforms information.
36846 Default value is transforms.trf.
36847 shakiness
36849 Set how shaky the video is and how quick the camera is. It accepts
36850 an integer in the range 1-10, a value of 1 means little shakiness,
36851 a value of 10 means strong shakiness. Default value is 5.
36852 accuracy
36854 Set the accuracy of the detection process. It must be a value in
36855 the range 1-15. A value of 1 means low accuracy, a value of 15
36856 means high accuracy. Default value is 15.
36857 stepsize
36859 Set stepsize of the search process. The region around minimum is
36860 scanned with 1 pixel resolution. Default value is 6.
36861 mincontrast
36863 Set minimum contrast. Below this value a local measurement field is
36864 discarded. Must be a floating point value in the range 0-1. Default
36865 value is 0.3.
36866 tripod
36868 Set reference frame number for tripod mode.
36869 If enabled, the motion of the frames is compared to a reference
36871 frame in the filtered stream, identified by the specified number.
36872 The idea is to compensate all movements in a more-or-less static
36873 scene and keep the camera view absolutely still.
36874 If set to 0, it is disabled. The frames are counted starting from
36876 1.
36877 show
36879 Show fields and transforms in the resulting frames. It accepts an
36880 integer in the range 0-2. Default value is 0, which disables any
36881 visualization.
36882 Examples
36884 • Use default values:
36886 vidstabdetect
36888 • Analyze strongly shaky movie and put the results in file
36890 mytransforms.trf:
36891 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
36893 • Visualize the result of internal transformations in the resulting
36895 video:
36896 vidstabdetect=show=1
36898 • Analyze a video with medium shakiness using ffmpeg:
36900 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
36902 vidstabtransform
36904 Video stabilization/deshaking: pass 2 of 2, see vidstabdetect for pass
36905 1.
36906 Read a file with transform information for each frame and
36908 apply/compensate them. Together with the vidstabdetect filter this can
36909 be used to deshake videos. See also
36910 <http://public.hronopik.de/vid.stab>. It is important to also use the
36911 unsharp filter, see below.
36912 To enable compilation of this filter you need to configure FFmpeg with
36914 "--enable-libvidstab".
36915 Options
36917 input
36919 Set path to the file used to read the transforms. Default value is
36920 transforms.trf.
36921 smoothing
36923 Set the number of frames (value*2 + 1) used for lowpass filtering
36924 the camera movements. Default value is 10.
36925 For example a number of 10 means that 21 frames are used (10 in the
36927 past and 10 in the future) to smoothen the motion in the video. A
36928 larger value leads to a smoother video, but limits the acceleration
36929 of the camera (pan/tilt movements). 0 is a special case where a
36930 static camera is simulated.
36931 optalgo
36933 Set the camera path optimization algorithm.
36934 Accepted values are:
36936 gauss
36938 gaussian kernel low-pass filter on camera motion (default)
36939 avg averaging on transformations
36941 maxshift
36943 Set maximal number of pixels to translate frames. Default value is
36944 -1, meaning no limit.
36945 maxangle
36947 Set maximal angle in radians (degree*PI/180) to rotate frames.
36948 Default value is -1, meaning no limit.
36949 crop
36951 Specify how to deal with borders that may be visible due to
36952 movement compensation.
36953 Available values are:
36955 keep
36957 keep image information from previous frame (default)
36958 black
36960 fill the border black
36961 invert
36963 Invert transforms if set to 1. Default value is 0.
36964 relative
36966 Consider transforms as relative to previous frame if set to 1,
36967 absolute if set to 0. Default value is 0.
36968 zoom
36970 Set percentage to zoom. A positive value will result in a zoom-in
36971 effect, a negative value in a zoom-out effect. Default value is 0
36972 (no zoom).
36973 optzoom
36975 Set optimal zooming to avoid borders.
36976 Accepted values are:
36978 0 disabled
36980 1 optimal static zoom value is determined (only very strong
36982 movements will lead to visible borders) (default)
36983 2 optimal adaptive zoom value is determined (no borders will be
36985 visible), see zoomspeed
36986 Note that the value given at zoom is added to the one calculated
36988 here.
36989 zoomspeed
36991 Set percent to zoom maximally each frame (enabled when optzoom is
36992 set to 2). Range is from 0 to 5, default value is 0.25.
36993 interpol
36995 Specify type of interpolation.
36996 Available values are:
36998 no no interpolation
37000 linear
37002 linear only horizontal
37003 bilinear
37005 linear in both directions (default)
37006 bicubic
37008 cubic in both directions (slow)
37009 tripod
37011 Enable virtual tripod mode if set to 1, which is equivalent to
37012 "relative=0:smoothing=0". Default value is 0.
37013 Use also "tripod" option of vidstabdetect.
37015 debug
37017 Increase log verbosity if set to 1. Also the detected global
37018 motions are written to the temporary file global_motions.trf.
37019 Default value is 0.
37020 Examples
37022 • Use ffmpeg for a typical stabilization with default values:
37024 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
37026 Note the use of the unsharp filter which is always recommended.
37028 • Zoom in a bit more and load transform data from a given file:
37030 vidstabtransform=zoom=5:input="mytransforms.trf"
37032 • Smoothen the video even more:
37034 vidstabtransform=smoothing=30
37036 vflip
37038 Flip the input video vertically.
37039 For example, to vertically flip a video with ffmpeg:
37041 ffmpeg -i in.avi -vf "vflip" out.avi
37043 vfrdet
37045 Detect variable frame rate video.
37046 This filter tries to detect if the input is variable or constant frame
37048 rate.
37049 At end it will output number of frames detected as having variable
37051 delta pts, and ones with constant delta pts. If there was frames with
37052 variable delta, than it will also show min, max and average delta
37053 encountered.
37054 vibrance
37056 Boost or alter saturation.
37057 The filter accepts the following options:
37059 intensity
37061 Set strength of boost if positive value or strength of alter if
37062 negative value. Default is 0. Allowed range is from -2 to 2.
37063 rbal
37065 Set the red balance. Default is 1. Allowed range is from -10 to 10.
37066 gbal
37068 Set the green balance. Default is 1. Allowed range is from -10 to
37069 10.
37070 bbal
37072 Set the blue balance. Default is 1. Allowed range is from -10 to
37073 10.
37074 rlum
37076 Set the red luma coefficient.
37077 glum
37079 Set the green luma coefficient.
37080 blum
37082 Set the blue luma coefficient.
37083 alternate
37085 If "intensity" is negative and this is set to 1, colors will
37086 change, otherwise colors will be less saturated, more towards gray.
37087 Commands
37089 This filter supports the all above options as commands.
37091 vif
37093 Obtain the average VIF (Visual Information Fidelity) between two input
37094 videos.
37095 This filter takes two input videos.
37097 Both input videos must have the same resolution and pixel format for
37099 this filter to work correctly. Also it assumes that both inputs have
37100 the same number of frames, which are compared one by one.
37101 The obtained average VIF score is printed through the logging system.
37103 The filter stores the calculated VIF score of each frame.
37105 In the below example the input file main.mpg being processed is
37107 compared with the reference file ref.mpg.
37108 ffmpeg -i main.mpg -i ref.mpg -lavfi vif -f null -
37110 vignette
37112 Make or reverse a natural vignetting effect.
37113 The filter accepts the following options:
37115 angle, a
37117 Set lens angle expression as a number of radians.
37118 The value is clipped in the "[0,PI/2]" range.
37120 Default value: "PI/5"
37122 x0
37124 y0 Set center coordinates expressions. Respectively "w/2" and "h/2" by
37125 default.
37126 mode
37128 Set forward/backward mode.
37129 Available modes are:
37131 forward
37133 The larger the distance from the central point, the darker the
37134 image becomes.
37135 backward
37137 The larger the distance from the central point, the brighter
37138 the image becomes. This can be used to reverse a vignette
37139 effect, though there is no automatic detection to extract the
37140 lens angle and other settings (yet). It can also be used to
37141 create a burning effect.
37142 Default value is forward.
37144 eval
37146 Set evaluation mode for the expressions (angle, x0, y0).
37147 It accepts the following values:
37149 init
37151 Evaluate expressions only once during the filter
37152 initialization.
37153 frame
37155 Evaluate expressions for each incoming frame. This is way
37156 slower than the init mode since it requires all the scalers to
37157 be re-computed, but it allows advanced dynamic expressions.
37158 Default value is init.
37160 dither
37162 Set dithering to reduce the circular banding effects. Default is 1
37163 (enabled).
37164 aspect
37166 Set vignette aspect. This setting allows one to adjust the shape of
37167 the vignette. Setting this value to the SAR of the input will make
37168 a rectangular vignetting following the dimensions of the video.
37169 Default is "1/1".
37171 Expressions
37173 The alpha, x0 and y0 expressions can contain the following parameters.
37175 w
37177 h input width and height
37178 n the number of input frame, starting from 0
37180 pts the PTS (Presentation TimeStamp) time of the filtered video frame,
37182 expressed in TB units, NAN if undefined
37183 r frame rate of the input video, NAN if the input frame rate is
37185 unknown
37186 t the PTS (Presentation TimeStamp) of the filtered video frame,
37188 expressed in seconds, NAN if undefined
37189 tb time base of the input video
37191 Examples
37193 • Apply simple strong vignetting effect:
37195 vignette=PI/4
37197 • Make a flickering vignetting:
37199 vignette='PI/4+random(1)*PI/50':eval=frame
37201 vmafmotion
37203 Obtain the average VMAF motion score of a video. It is one of the
37204 component metrics of VMAF.
37205 The obtained average motion score is printed through the logging
37207 system.
37208 The filter accepts the following options:
37210 stats_file
37212 If specified, the filter will use the named file to save the motion
37213 score of each frame with respect to the previous frame. When
37214 filename equals "-" the data is sent to standard output.
37215 Example:
37217 ffmpeg -i ref.mpg -vf vmafmotion -f null -
37219 vstack
37221 Stack input videos vertically.
37222 All streams must be of same pixel format and of same width.
37224 Note that this filter is faster than using overlay and pad filter to
37226 create same output.
37227 The filter accepts the following options:
37229 inputs
37231 Set number of input streams. Default is 2.
37232 shortest
37234 If set to 1, force the output to terminate when the shortest input
37235 terminates. Default value is 0.
37236 w3fdif
37238 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
37239 Deinterlacing Filter").
37240 Based on the process described by Martin Weston for BBC R&D, and
37242 implemented based on the de-interlace algorithm written by Jim
37243 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter uses
37244 filter coefficients calculated by BBC R&D.
37245 This filter uses field-dominance information in frame to decide which
37247 of each pair of fields to place first in the output. If it gets it
37248 wrong use setfield filter before "w3fdif" filter.
37249 There are two sets of filter coefficients, so called "simple" and
37251 "complex". Which set of filter coefficients is used can be set by
37252 passing an optional parameter:
37253 filter
37255 Set the interlacing filter coefficients. Accepts one of the
37256 following values:
37257 simple
37259 Simple filter coefficient set.
37260 complex
37262 More-complex filter coefficient set.
37263 Default value is complex.
37265 mode
37267 The interlacing mode to adopt. It accepts one of the following
37268 values:
37269 frame
37271 Output one frame for each frame.
37272 field
37274 Output one frame for each field.
37275 The default value is "field".
37277 parity
37279 The picture field parity assumed for the input interlaced video. It
37280 accepts one of the following values:
37281 tff Assume the top field is first.
37283 bff Assume the bottom field is first.
37285 auto
37287 Enable automatic detection of field parity.
37288 The default value is "auto". If the interlacing is unknown or the
37290 decoder does not export this information, top field first will be
37291 assumed.
37292 deint
37294 Specify which frames to deinterlace. Accepts one of the following
37295 values:
37296 all Deinterlace all frames,
37298 interlaced
37300 Only deinterlace frames marked as interlaced.
37301 Default value is all.
37303 Commands
37305 This filter supports same commands as options.
37307 waveform
37309 Video waveform monitor.
37310 The waveform monitor plots color component intensity. By default
37312 luminance only. Each column of the waveform corresponds to a column of
37313 pixels in the source video.
37314 It accepts the following options:
37316 mode, m
37318 Can be either "row", or "column". Default is "column". In row
37319 mode, the graph on the left side represents color component value 0
37320 and the right side represents value = 255. In column mode, the top
37321 side represents color component value = 0 and bottom side
37322 represents value = 255.
37323 intensity, i
37325 Set intensity. Smaller values are useful to find out how many
37326 values of the same luminance are distributed across input
37327 rows/columns. Default value is 0.04. Allowed range is [0, 1].
37328 mirror, r
37330 Set mirroring mode. 0 means unmirrored, 1 means mirrored. In
37331 mirrored mode, higher values will be represented on the left side
37332 for "row" mode and at the top for "column" mode. Default is 1
37333 (mirrored).
37334 display, d
37336 Set display mode. It accepts the following values:
37337 overlay
37339 Presents information identical to that in the "parade", except
37340 that the graphs representing color components are superimposed
37341 directly over one another.
37342 This display mode makes it easier to spot relative differences
37344 or similarities in overlapping areas of the color components
37345 that are supposed to be identical, such as neutral whites,
37346 grays, or blacks.
37347 stack
37349 Display separate graph for the color components side by side in
37350 "row" mode or one below the other in "column" mode.
37351 parade
37353 Display separate graph for the color components side by side in
37354 "column" mode or one below the other in "row" mode.
37355 Using this display mode makes it easy to spot color casts in
37357 the highlights and shadows of an image, by comparing the
37358 contours of the top and the bottom graphs of each waveform.
37359 Since whites, grays, and blacks are characterized by exactly
37360 equal amounts of red, green, and blue, neutral areas of the
37361 picture should display three waveforms of roughly equal
37362 width/height. If not, the correction is easy to perform by
37363 making level adjustments the three waveforms.
37364 Default is "stack".
37366 components, c
37368 Set which color components to display. Default is 1, which means
37369 only luminance or red color component if input is in RGB
37370 colorspace. If is set for example to 7 it will display all 3 (if)
37371 available color components.
37372 envelope, e
37374 none
37375 No envelope, this is default.
37376 instant
37378 Instant envelope, minimum and maximum values presented in graph
37379 will be easily visible even with small "step" value.
37380 peak
37382 Hold minimum and maximum values presented in graph across time.
37383 This way you can still spot out of range values without
37384 constantly looking at waveforms.
37385 peak+instant
37387 Peak and instant envelope combined together.
37388 filter, f
37390 lowpass
37391 No filtering, this is default.
37392 flat
37394 Luma and chroma combined together.
37395 aflat
37397 Similar as above, but shows difference between blue and red
37398 chroma.
37399 xflat
37401 Similar as above, but use different colors.
37402 yflat
37404 Similar as above, but again with different colors.
37405 chroma
37407 Displays only chroma.
37408 color
37410 Displays actual color value on waveform.
37411 acolor
37413 Similar as above, but with luma showing frequency of chroma
37414 values.
37415 graticule, g
37417 Set which graticule to display.
37418 none
37420 Do not display graticule.
37421 green
37423 Display green graticule showing legal broadcast ranges.
37424 orange
37426 Display orange graticule showing legal broadcast ranges.
37427 invert
37429 Display invert graticule showing legal broadcast ranges.
37430 opacity, o
37432 Set graticule opacity.
37433 flags, fl
37435 Set graticule flags.
37436 numbers
37438 Draw numbers above lines. By default enabled.
37439 dots
37441 Draw dots instead of lines.
37442 scale, s
37444 Set scale used for displaying graticule.
37445 digital
37447 millivolts
37448 ire
37449 Default is digital.
37451 bgopacity, b
37453 Set background opacity.
37454 tint0, t0
37456 tint1, t1
37457 Set tint for output. Only used with lowpass filter and when
37458 display is not overlay and input pixel formats are not RGB.
37459 fitmode, fm
37461 Set sample aspect ratio of video output frames. Can be used to
37462 configure waveform so it is not streched too much in one of
37463 directions.
37464 none
37466 Set sample aspect ration to 1/1.
37467 size
37469 Set sample aspect ratio to match input size of video
37470 Default is none.
37472 weave, doubleweave
37474 The "weave" takes a field-based video input and join each two
37475 sequential fields into single frame, producing a new double height clip
37476 with half the frame rate and half the frame count.
37477 The "doubleweave" works same as "weave" but without halving frame rate
37479 and frame count.
37480 It accepts the following option:
37482 first_field
37484 Set first field. Available values are:
37485 top, t
37487 Set the frame as top-field-first.
37488 bottom, b
37490 Set the frame as bottom-field-first.
37491 Examples
37493 • Interlace video using select and separatefields filter:
37495 separatefields,select=eq(mod(n,4),0)+eq(mod(n,4),3),weave
37497 xbr
37499 Apply the xBR high-quality magnification filter which is designed for
37500 pixel art. It follows a set of edge-detection rules, see
37501 <https://forums.libretro.com/t/xbr-algorithm-tutorial/123>.
37502 It accepts the following option:
37504 n Set the scaling dimension: 2 for "2xBR", 3 for "3xBR" and 4 for
37506 "4xBR". Default is 3.
37507 xcorrelate
37509 Apply normalized cross-correlation between first and second input video
37510 stream.
37511 Second input video stream dimensions must be lower than first input
37513 video stream.
37514 The filter accepts the following options:
37516 planes
37518 Set which planes to process.
37519 secondary
37521 Set which secondary video frames will be processed from second
37522 input video stream, can be first or all. Default is all.
37523 The "xcorrelate" filter also supports the framesync options.
37525 xfade
37527 Apply cross fade from one input video stream to another input video
37528 stream. The cross fade is applied for specified duration.
37529 Both inputs must be constant frame-rate and have the same resolution,
37531 pixel format, frame rate and timebase.
37532 The filter accepts the following options:
37534 transition
37536 Set one of available transition effects:
37537 custom
37539 fade
37540 wipeleft
37541 wiperight
37542 wipeup
37543 wipedown
37544 slideleft
37545 slideright
37546 slideup
37547 slidedown
37548 circlecrop
37549 rectcrop
37550 distance
37551 fadeblack
37552 fadewhite
37553 radial
37554 smoothleft
37555 smoothright
37556 smoothup
37557 smoothdown
37558 circleopen
37559 circleclose
37560 vertopen
37561 vertclose
37562 horzopen
37563 horzclose
37564 dissolve
37565 pixelize
37566 diagtl
37567 diagtr
37568 diagbl
37569 diagbr
37570 hlslice
37571 hrslice
37572 vuslice
37573 vdslice
37574 hblur
37575 fadegrays
37576 wipetl
37577 wipetr
37578 wipebl
37579 wipebr
37580 squeezeh
37581 squeezev
37582 zoomin
37583 Default transition effect is fade.
37585 duration
37587 Set cross fade duration in seconds. Range is 0 to 60 seconds.
37588 Default duration is 1 second.
37589 offset
37591 Set cross fade start relative to first input stream in seconds.
37592 Default offset is 0.
37593 expr
37595 Set expression for custom transition effect.
37596 The expressions can use the following variables and functions:
37598 X
37600 Y The coordinates of the current sample.
37601 W
37603 H The width and height of the image.
37604 P Progress of transition effect.
37606 PLANE
37608 Currently processed plane.
37609 A Return value of first input at current location and plane.
37611 B Return value of second input at current location and plane.
37613 a0(x, y)
37615 a1(x, y)
37616 a2(x, y)
37617 a3(x, y)
37618 Return the value of the pixel at location (x,y) of the
37619 first/second/third/fourth component of first input.
37620 b0(x, y)
37622 b1(x, y)
37623 b2(x, y)
37624 b3(x, y)
37625 Return the value of the pixel at location (x,y) of the
37626 first/second/third/fourth component of second input.
37627 Examples
37629 • Cross fade from one input video to another input video, with fade
37631 transition and duration of transition of 2 seconds starting at
37632 offset of 5 seconds:
37633 ffmpeg -i first.mp4 -i second.mp4 -filter_complex xfade=transition=fade:duration=2:offset=5 output.mp4
37635 xmedian
37637 Pick median pixels from several input videos.
37638 The filter accepts the following options:
37640 inputs
37642 Set number of inputs. Default is 3. Allowed range is from 3 to
37643 255. If number of inputs is even number, than result will be mean
37644 value between two median values.
37645 planes
37647 Set which planes to filter. Default value is 15, by which all
37648 planes are processed.
37649 percentile
37651 Set median percentile. Default value is 0.5. Default value of 0.5
37652 will pick always median values, while 0 will pick minimum values,
37653 and 1 maximum values.
37654 Commands
37656 This filter supports all above options as commands, excluding option
37658 "inputs".
37659 xstack
37661 Stack video inputs into custom layout.
37662 All streams must be of same pixel format.
37664 The filter accepts the following options:
37666 inputs
37668 Set number of input streams. Default is 2.
37669 layout
37671 Specify layout of inputs. This option requires the desired layout
37672 configuration to be explicitly set by the user. This sets position
37673 of each video input in output. Each input is separated by '|'. The
37674 first number represents the column, and the second number
37675 represents the row. Numbers start at 0 and are separated by '_'.
37676 Optionally one can use wX and hX, where X is video input from which
37677 to take width or height. Multiple values can be used when
37678 separated by '+'. In such case values are summed together.
37679 Note that if inputs are of different sizes gaps may appear, as not
37681 all of the output video frame will be filled. Similarly, videos can
37682 overlap each other if their position doesn't leave enough space for
37683 the full frame of adjoining videos.
37684 For 2 inputs, a default layout of "0_0|w0_0" is set. In all other
37686 cases, a layout must be set by the user.
37687 shortest
37689 If set to 1, force the output to terminate when the shortest input
37690 terminates. Default value is 0.
37691 fill
37693 If set to valid color, all unused pixels will be filled with that
37694 color. By default fill is set to none, so it is disabled.
37695 Examples
37697 • Display 4 inputs into 2x2 grid.
37699 Layout:
37701 input1(0, 0) | input3(w0, 0)
37703 input2(0, h0) | input4(w0, h0)
37704 xstack=inputs=4:layout=0_0|0_h0|w0_0|w0_h0
37708 Note that if inputs are of different sizes, gaps or overlaps may
37710 occur.
37711 • Display 4 inputs into 1x4 grid.
37713 Layout:
37715 input1(0, 0)
37717 input2(0, h0)
37718 input3(0, h0+h1)
37719 input4(0, h0+h1+h2)
37720 xstack=inputs=4:layout=0_0|0_h0|0_h0+h1|0_h0+h1+h2
37724 Note that if inputs are of different widths, unused space will
37726 appear.
37727 • Display 9 inputs into 3x3 grid.
37729 Layout:
37731 input1(0, 0) | input4(w0, 0) | input7(w0+w3, 0)
37733 input2(0, h0) | input5(w0, h0) | input8(w0+w3, h0)
37734 input3(0, h0+h1) | input6(w0, h0+h1) | input9(w0+w3, h0+h1)
37735 xstack=inputs=9:layout=0_0|0_h0|0_h0+h1|w0_0|w0_h0|w0_h0+h1|w0+w3_0|w0+w3_h0|w0+w3_h0+h1
37739 Note that if inputs are of different sizes, gaps or overlaps may
37741 occur.
37742 • Display 16 inputs into 4x4 grid.
37744 Layout:
37746 input1(0, 0) | input5(w0, 0) | input9 (w0+w4, 0) | input13(w0+w4+w8, 0)
37748 input2(0, h0) | input6(w0, h0) | input10(w0+w4, h0) | input14(w0+w4+w8, h0)
37749 input3(0, h0+h1) | input7(w0, h0+h1) | input11(w0+w4, h0+h1) | input15(w0+w4+w8, h0+h1)
37750 input4(0, h0+h1+h2)| input8(w0, h0+h1+h2)| input12(w0+w4, h0+h1+h2)| input16(w0+w4+w8, h0+h1+h2)
37751 xstack=inputs=16:layout=0_0|0_h0|0_h0+h1|0_h0+h1+h2|w0_0|w0_h0|w0_h0+h1|w0_h0+h1+h2|w0+w4_0|
37755 w0+w4_h0|w0+w4_h0+h1|w0+w4_h0+h1+h2|w0+w4+w8_0|w0+w4+w8_h0|w0+w4+w8_h0+h1|w0+w4+w8_h0+h1+h2
37756 Note that if inputs are of different sizes, gaps or overlaps may
37758 occur.
37759 yadif
37761 Deinterlace the input video ("yadif" means "yet another deinterlacing
37762 filter").
37763 It accepts the following parameters:
37765 mode
37767 The interlacing mode to adopt. It accepts one of the following
37768 values:
37769 0, send_frame
37771 Output one frame for each frame.
37772 1, send_field
37774 Output one frame for each field.
37775 2, send_frame_nospatial
37777 Like "send_frame", but it skips the spatial interlacing check.
37778 3, send_field_nospatial
37780 Like "send_field", but it skips the spatial interlacing check.
37781 The default value is "send_frame".
37783 parity
37785 The picture field parity assumed for the input interlaced video. It
37786 accepts one of the following values:
37787 0, tff
37789 Assume the top field is first.
37790 1, bff
37792 Assume the bottom field is first.
37793 -1, auto
37795 Enable automatic detection of field parity.
37796 The default value is "auto". If the interlacing is unknown or the
37798 decoder does not export this information, top field first will be
37799 assumed.
37800 deint
37802 Specify which frames to deinterlace. Accepts one of the following
37803 values:
37804 0, all
37806 Deinterlace all frames.
37807 1, interlaced
37809 Only deinterlace frames marked as interlaced.
37810 The default value is "all".
37812 yadif_cuda
37814 Deinterlace the input video using the yadif algorithm, but implemented
37815 in CUDA so that it can work as part of a GPU accelerated pipeline with
37816 nvdec and/or nvenc.
37817 It accepts the following parameters:
37819 mode
37821 The interlacing mode to adopt. It accepts one of the following
37822 values:
37823 0, send_frame
37825 Output one frame for each frame.
37826 1, send_field
37828 Output one frame for each field.
37829 2, send_frame_nospatial
37831 Like "send_frame", but it skips the spatial interlacing check.
37832 3, send_field_nospatial
37834 Like "send_field", but it skips the spatial interlacing check.
37835 The default value is "send_frame".
37837 parity
37839 The picture field parity assumed for the input interlaced video. It
37840 accepts one of the following values:
37841 0, tff
37843 Assume the top field is first.
37844 1, bff
37846 Assume the bottom field is first.
37847 -1, auto
37849 Enable automatic detection of field parity.
37850 The default value is "auto". If the interlacing is unknown or the
37852 decoder does not export this information, top field first will be
37853 assumed.
37854 deint
37856 Specify which frames to deinterlace. Accepts one of the following
37857 values:
37858 0, all
37860 Deinterlace all frames.
37861 1, interlaced
37863 Only deinterlace frames marked as interlaced.
37864 The default value is "all".
37866 yaepblur
37868 Apply blur filter while preserving edges ("yaepblur" means "yet another
37869 edge preserving blur filter"). The algorithm is described in "J. S.
37870 Lee, Digital image enhancement and noise filtering by use of local
37871 statistics, IEEE Trans. Pattern Anal. Mach. Intell. PAMI-2, 1980."
37872 It accepts the following parameters:
37874 radius, r
37876 Set the window radius. Default value is 3.
37877 planes, p
37879 Set which planes to filter. Default is only the first plane.
37880 sigma, s
37882 Set blur strength. Default value is 128.
37883 Commands
37885 This filter supports same commands as options.
37887 zoompan
37889 Apply Zoom & Pan effect.
37890 This filter accepts the following options:
37892 zoom, z
37894 Set the zoom expression. Range is 1-10. Default is 1.
37895 x
37897 y Set the x and y expression. Default is 0.
37898 d Set the duration expression in number of frames. This sets for how
37900 many number of frames effect will last for single input image.
37901 Default is 90.
37902 s Set the output image size, default is 'hd720'.
37904 fps Set the output frame rate, default is '25'.
37906 Each expression can contain the following constants:
37908 in_w, iw
37910 Input width.
37911 in_h, ih
37913 Input height.
37914 out_w, ow
37916 Output width.
37917 out_h, oh
37919 Output height.
37920 in Input frame count.
37922 on Output frame count.
37924 in_time, it
37926 The input timestamp expressed in seconds. It's NAN if the input
37927 timestamp is unknown.
37928 out_time, time, ot
37930 The output timestamp expressed in seconds.
37931 x
37933 y Last calculated 'x' and 'y' position from 'x' and 'y' expression
37934 for current input frame.
37935 px
37937 py 'x' and 'y' of last output frame of previous input frame or 0 when
37938 there was not yet such frame (first input frame).
37939 zoom
37941 Last calculated zoom from 'z' expression for current input frame.
37942 pzoom
37944 Last calculated zoom of last output frame of previous input frame.
37945 duration
37947 Number of output frames for current input frame. Calculated from
37948 'd' expression for each input frame.
37949 pduration
37951 number of output frames created for previous input frame
37952 a Rational number: input width / input height
37954 sar sample aspect ratio
37956 dar display aspect ratio
37958 Examples
37960 • Zoom in up to 1.5x and pan at same time to some spot near center of
37962 picture:
37963 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='if(gte(zoom,1.5),x,x+1/a)':y='if(gte(zoom,1.5),y,y+1)':s=640x360
37965 • Zoom in up to 1.5x and pan always at center of picture:
37967 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
37969 • Same as above but without pausing:
37971 zoompan=z='min(max(zoom,pzoom)+0.0015,1.5)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
37973 • Zoom in 2x into center of picture only for the first second of the
37975 input video:
37976 zoompan=z='if(between(in_time,0,1),2,1)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
37978 zscale
37980 Scale (resize) the input video, using the z.lib library:
37981 <https://github.com/sekrit-twc/zimg>. To enable compilation of this
37982 filter, you need to configure FFmpeg with "--enable-libzimg".
37983 The zscale filter forces the output display aspect ratio to be the same
37985 as the input, by changing the output sample aspect ratio.
37986 If the input image format is different from the format requested by the
37988 next filter, the zscale filter will convert the input to the requested
37989 format.
37990 Options
37992 The filter accepts the following options.
37994 width, w
37996 height, h
37997 Set the output video dimension expression. Default value is the
37998 input dimension.
37999 If the width or w value is 0, the input width is used for the
38001 output. If the height or h value is 0, the input height is used for
38002 the output.
38003 If one and only one of the values is -n with n >= 1, the zscale
38005 filter will use a value that maintains the aspect ratio of the
38006 input image, calculated from the other specified dimension. After
38007 that it will, however, make sure that the calculated dimension is
38008 divisible by n and adjust the value if necessary.
38009 If both values are -n with n >= 1, the behavior will be identical
38011 to both values being set to 0 as previously detailed.
38012 See below for the list of accepted constants for use in the
38014 dimension expression.
38015 size, s
38017 Set the video size. For the syntax of this option, check the "Video
38018 size" section in the ffmpeg-utils manual.
38019 dither, d
38021 Set the dither type.
38022 Possible values are:
38024 none
38026 ordered
38027 random
38028 error_diffusion
38029 Default is none.
38031 filter, f
38033 Set the resize filter type.
38034 Possible values are:
38036 point
38038 bilinear
38039 bicubic
38040 spline16
38041 spline36
38042 lanczos
38043 Default is bilinear.
38045 range, r
38047 Set the color range.
38048 Possible values are:
38050 input
38052 limited
38053 full
38054 Default is same as input.
38056 primaries, p
38058 Set the color primaries.
38059 Possible values are:
38061 input
38063 709
38064 unspecified
38065 170m
38066 240m
38067 2020
38068 Default is same as input.
38070 transfer, t
38072 Set the transfer characteristics.
38073 Possible values are:
38075 input
38077 709
38078 unspecified
38079 601
38080 linear
38081 2020_10
38082 2020_12
38083 smpte2084
38084 iec61966-2-1
38085 arib-std-b67
38086 Default is same as input.
38088 matrix, m
38090 Set the colorspace matrix.
38091 Possible value are:
38093 input
38095 709
38096 unspecified
38097 470bg
38098 170m
38099 2020_ncl
38100 2020_cl
38101 Default is same as input.
38103 rangein, rin
38105 Set the input color range.
38106 Possible values are:
38108 input
38110 limited
38111 full
38112 Default is same as input.
38114 primariesin, pin
38116 Set the input color primaries.
38117 Possible values are:
38119 input
38121 709
38122 unspecified
38123 170m
38124 240m
38125 2020
38126 Default is same as input.
38128 transferin, tin
38130 Set the input transfer characteristics.
38131 Possible values are:
38133 input
38135 709
38136 unspecified
38137 601
38138 linear
38139 2020_10
38140 2020_12
38141 Default is same as input.
38143 matrixin, min
38145 Set the input colorspace matrix.
38146 Possible value are:
38148 input
38150 709
38151 unspecified
38152 470bg
38153 170m
38154 2020_ncl
38155 2020_cl
38156 chromal, c
38157 Set the output chroma location.
38158 Possible values are:
38160 input
38162 left
38163 center
38164 topleft
38165 top
38166 bottomleft
38167 bottom
38168 chromalin, cin
38169 Set the input chroma location.
38170 Possible values are:
38172 input
38174 left
38175 center
38176 topleft
38177 top
38178 bottomleft
38179 bottom
38180 npl Set the nominal peak luminance.
38181 param_a
38183 Parameter A for scaling filters. Parameter "b" for bicubic, and the
38184 number of filter taps for lanczos.
38185 param_b
38187 Parameter B for scaling filters. Parameter "c" for bicubic.
38188 The values of the w and h options are expressions containing the
38190 following constants:
38191 in_w
38193 in_h
38194 The input width and height
38195 iw
38197 ih These are the same as in_w and in_h.
38198 out_w
38200 out_h
38201 The output (scaled) width and height
38202 ow
38204 oh These are the same as out_w and out_h
38205 a The same as iw / ih
38207 sar input sample aspect ratio
38209 dar The input display aspect ratio. Calculated from "(iw / ih) * sar".
38211 hsub
38213 vsub
38214 horizontal and vertical input chroma subsample values. For example
38215 for the pixel format "yuv422p" hsub is 2 and vsub is 1.
38216 ohsub
38218 ovsub
38219 horizontal and vertical output chroma subsample values. For example
38220 for the pixel format "yuv422p" hsub is 2 and vsub is 1.
38221 Commands
38223 This filter supports the following commands:
38225 width, w
38227 height, h
38228 Set the output video dimension expression. The command accepts the
38229 same syntax of the corresponding option.
38230 If the specified expression is not valid, it is kept at its current
38232 value.
38233
38235 Below is a description of the currently available OpenCL video filters.
38236 To enable compilation of these filters you need to configure FFmpeg
38238 with "--enable-opencl".
38239 Running OpenCL filters requires you to initialize a hardware device and
38241 to pass that device to all filters in any filter graph.
38242 -init_hw_device opencl[=name][:device[,key=value...]]
38244 Initialise a new hardware device of type opencl called name, using
38245 the given device parameters.
38246 -filter_hw_device name
38248 Pass the hardware device called name to all filters in any filter
38249 graph.
38250 For more detailed information see
38252 <https://www.ffmpeg.org/ffmpeg.html#Advanced-Video-options>
38253 • Example of choosing the first device on the second platform and
38255 running avgblur_opencl filter with default parameters on it.
38256 -init_hw_device opencl=gpu:1.0 -filter_hw_device gpu -i INPUT -vf "hwupload, avgblur_opencl, hwdownload" OUTPUT
38258 Since OpenCL filters are not able to access frame data in normal
38260 memory, all frame data needs to be uploaded(hwupload) to hardware
38261 surfaces connected to the appropriate device before being used and then
38262 downloaded(hwdownload) back to normal memory. Note that hwupload will
38263 upload to a surface with the same layout as the software frame, so it
38264 may be necessary to add a format filter immediately before to get the
38265 input into the right format and hwdownload does not support all formats
38266 on the output - it may be necessary to insert an additional format
38267 filter immediately following in the graph to get the output in a
38268 supported format.
38269 avgblur_opencl
38271 Apply average blur filter.
38272 The filter accepts the following options:
38274 sizeX
38276 Set horizontal radius size. Range is "[1, 1024]" and default value
38277 is 1.
38278 planes
38280 Set which planes to filter. Default value is 0xf, by which all
38281 planes are processed.
38282 sizeY
38284 Set vertical radius size. Range is "[1, 1024]" and default value is
38285 0. If zero, "sizeX" value will be used.
38286 Example
38288 • Apply average blur filter with horizontal and vertical size of 3,
38290 setting each pixel of the output to the average value of the 7x7
38291 region centered on it in the input. For pixels on the edges of the
38292 image, the region does not extend beyond the image boundaries, and
38293 so out-of-range coordinates are not used in the calculations.
38294 -i INPUT -vf "hwupload, avgblur_opencl=3, hwdownload" OUTPUT
38296 boxblur_opencl
38298 Apply a boxblur algorithm to the input video.
38299 It accepts the following parameters:
38301 luma_radius, lr
38303 luma_power, lp
38304 chroma_radius, cr
38305 chroma_power, cp
38306 alpha_radius, ar
38307 alpha_power, ap
38308 A description of the accepted options follows.
38310 luma_radius, lr
38312 chroma_radius, cr
38313 alpha_radius, ar
38314 Set an expression for the box radius in pixels used for blurring
38315 the corresponding input plane.
38316 The radius value must be a non-negative number, and must not be
38318 greater than the value of the expression "min(w,h)/2" for the luma
38319 and alpha planes, and of "min(cw,ch)/2" for the chroma planes.
38320 Default value for luma_radius is "2". If not specified,
38322 chroma_radius and alpha_radius default to the corresponding value
38323 set for luma_radius.
38324 The expressions can contain the following constants:
38326 w
38328 h The input width and height in pixels.
38329 cw
38331 ch The input chroma image width and height in pixels.
38332 hsub
38334 vsub
38335 The horizontal and vertical chroma subsample values. For
38336 example, for the pixel format "yuv422p", hsub is 2 and vsub is
38337 1.
38338 luma_power, lp
38340 chroma_power, cp
38341 alpha_power, ap
38342 Specify how many times the boxblur filter is applied to the
38343 corresponding plane.
38344 Default value for luma_power is 2. If not specified, chroma_power
38346 and alpha_power default to the corresponding value set for
38347 luma_power.
38348 A value of 0 will disable the effect.
38350 Examples
38352 Apply boxblur filter, setting each pixel of the output to the average
38354 value of box-radiuses luma_radius, chroma_radius, alpha_radius for each
38355 plane respectively. The filter will apply luma_power, chroma_power,
38356 alpha_power times onto the corresponding plane. For pixels on the edges
38357 of the image, the radius does not extend beyond the image boundaries,
38358 and so out-of-range coordinates are not used in the calculations.
38359 • Apply a boxblur filter with the luma, chroma, and alpha radius set
38361 to 2 and luma, chroma, and alpha power set to 3. The filter will
38362 run 3 times with box-radius set to 2 for every plane of the image.
38363 -i INPUT -vf "hwupload, boxblur_opencl=luma_radius=2:luma_power=3, hwdownload" OUTPUT
38365 -i INPUT -vf "hwupload, boxblur_opencl=2:3, hwdownload" OUTPUT
38366 • Apply a boxblur filter with luma radius set to 2, luma_power to 1,
38368 chroma_radius to 4, chroma_power to 5, alpha_radius to 3 and
38369 alpha_power to 7.
38370 For the luma plane, a 2x2 box radius will be run once.
38372 For the chroma plane, a 4x4 box radius will be run 5 times.
38374 For the alpha plane, a 3x3 box radius will be run 7 times.
38376 -i INPUT -vf "hwupload, boxblur_opencl=2:1:4:5:3:7, hwdownload" OUTPUT
38378 colorkey_opencl
38380 RGB colorspace color keying.
38381 The filter accepts the following options:
38383 color
38385 The color which will be replaced with transparency.
38386 similarity
38388 Similarity percentage with the key color.
38389 0.01 matches only the exact key color, while 1.0 matches
38391 everything.
38392 blend
38394 Blend percentage.
38395 0.0 makes pixels either fully transparent, or not transparent at
38397 all.
38398 Higher values result in semi-transparent pixels, with a higher
38400 transparency the more similar the pixels color is to the key color.
38401 Examples
38403 • Make every semi-green pixel in the input transparent with some
38405 slight blending:
38406 -i INPUT -vf "hwupload, colorkey_opencl=green:0.3:0.1, hwdownload" OUTPUT
38408 convolution_opencl
38410 Apply convolution of 3x3, 5x5, 7x7 matrix.
38411 The filter accepts the following options:
38413 0m
38415 1m
38416 2m
38417 3m Set matrix for each plane. Matrix is sequence of 9, 25 or 49
38418 signed numbers. Default value for each plane is "0 0 0 0 1 0 0 0
38419 0".
38420 0rdiv
38422 1rdiv
38423 2rdiv
38424 3rdiv
38425 Set multiplier for calculated value for each plane. If unset or 0,
38426 it will be sum of all matrix elements. The option value must be a
38427 float number greater or equal to 0.0. Default value is 1.0.
38428 0bias
38430 1bias
38431 2bias
38432 3bias
38433 Set bias for each plane. This value is added to the result of the
38434 multiplication. Useful for making the overall image brighter or
38435 darker. The option value must be a float number greater or equal
38436 to 0.0. Default value is 0.0.
38437 Examples
38439 • Apply sharpen:
38441 -i INPUT -vf "hwupload, convolution_opencl=0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0, hwdownload" OUTPUT
38443 • Apply blur:
38445 -i INPUT -vf "hwupload, convolution_opencl=1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1/9:1/9:1/9:1/9, hwdownload" OUTPUT
38447 • Apply edge enhance:
38449 -i INPUT -vf "hwupload, convolution_opencl=0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:5:1:1:1:0:128:128:128, hwdownload" OUTPUT
38451 • Apply edge detect:
38453 -i INPUT -vf "hwupload, convolution_opencl=0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:5:5:5:1:0:128:128:128, hwdownload" OUTPUT
38455 • Apply laplacian edge detector which includes diagonals:
38457 -i INPUT -vf "hwupload, convolution_opencl=1 1 1 1 -8 1 1 1 1:1 1 1 1 -8 1 1 1 1:1 1 1 1 -8 1 1 1 1:1 1 1 1 -8 1 1 1 1:5:5:5:1:0:128:128:0, hwdownload" OUTPUT
38459 • Apply emboss:
38461 -i INPUT -vf "hwupload, convolution_opencl=-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2, hwdownload" OUTPUT
38463 erosion_opencl
38465 Apply erosion effect to the video.
38466 This filter replaces the pixel by the local(3x3) minimum.
38468 It accepts the following options:
38470 threshold0
38472 threshold1
38473 threshold2
38474 threshold3
38475 Limit the maximum change for each plane. Range is "[0, 65535]" and
38476 default value is 65535. If 0, plane will remain unchanged.
38477 coordinates
38479 Flag which specifies the pixel to refer to. Range is "[0, 255]"
38480 and default value is 255, i.e. all eight pixels are used.
38481 Flags to local 3x3 coordinates region centered on "x":
38483 1 2 3
38485 4 x 5
38487 6 7 8
38489 Example
38491 • Apply erosion filter with threshold0 set to 30, threshold1 set 40,
38493 threshold2 set to 50 and coordinates set to 231, setting each pixel
38494 of the output to the local minimum between pixels: 1, 2, 3, 6, 7, 8
38495 of the 3x3 region centered on it in the input. If the difference
38496 between input pixel and local minimum is more then threshold of the
38497 corresponding plane, output pixel will be set to input pixel -
38498 threshold of corresponding plane.
38499 -i INPUT -vf "hwupload, erosion_opencl=30:40:50:coordinates=231, hwdownload" OUTPUT
38501 deshake_opencl
38503 Feature-point based video stabilization filter.
38504 The filter accepts the following options:
38506 tripod
38508 Simulates a tripod by preventing any camera movement whatsoever
38509 from the original frame. Defaults to 0.
38510 debug
38512 Whether or not additional debug info should be displayed, both in
38513 the processed output and in the console.
38514 Note that in order to see console debug output you will also need
38516 to pass "-v verbose" to ffmpeg.
38517 Viewing point matches in the output video is only supported for RGB
38519 input.
38520 Defaults to 0.
38522 adaptive_crop
38524 Whether or not to do a tiny bit of cropping at the borders to cut
38525 down on the amount of mirrored pixels.
38526 Defaults to 1.
38528 refine_features
38530 Whether or not feature points should be refined at a sub-pixel
38531 level.
38532 This can be turned off for a slight performance gain at the cost of
38534 precision.
38535 Defaults to 1.
38537 smooth_strength
38539 The strength of the smoothing applied to the camera path from 0.0
38540 to 1.0.
38541 1.0 is the maximum smoothing strength while values less than that
38543 result in less smoothing.
38544 0.0 causes the filter to adaptively choose a smoothing strength on
38546 a per-frame basis.
38547 Defaults to 0.0.
38549 smooth_window_multiplier
38551 Controls the size of the smoothing window (the number of frames
38552 buffered to determine motion information from).
38553 The size of the smoothing window is determined by multiplying the
38555 framerate of the video by this number.
38556 Acceptable values range from 0.1 to 10.0.
38558 Larger values increase the amount of motion data available for
38560 determining how to smooth the camera path, potentially improving
38561 smoothness, but also increase latency and memory usage.
38562 Defaults to 2.0.
38564 Examples
38566 • Stabilize a video with a fixed, medium smoothing strength:
38568 -i INPUT -vf "hwupload, deshake_opencl=smooth_strength=0.5, hwdownload" OUTPUT
38570 • Stabilize a video with debugging (both in console and in rendered
38572 video):
38573 -i INPUT -filter_complex "[0:v]format=rgba, hwupload, deshake_opencl=debug=1, hwdownload, format=rgba, format=yuv420p" -v verbose OUTPUT
38575 dilation_opencl
38577 Apply dilation effect to the video.
38578 This filter replaces the pixel by the local(3x3) maximum.
38580 It accepts the following options:
38582 threshold0
38584 threshold1
38585 threshold2
38586 threshold3
38587 Limit the maximum change for each plane. Range is "[0, 65535]" and
38588 default value is 65535. If 0, plane will remain unchanged.
38589 coordinates
38591 Flag which specifies the pixel to refer to. Range is "[0, 255]"
38592 and default value is 255, i.e. all eight pixels are used.
38593 Flags to local 3x3 coordinates region centered on "x":
38595 1 2 3
38597 4 x 5
38599 6 7 8
38601 Example
38603 • Apply dilation filter with threshold0 set to 30, threshold1 set 40,
38605 threshold2 set to 50 and coordinates set to 231, setting each pixel
38606 of the output to the local maximum between pixels: 1, 2, 3, 6, 7, 8
38607 of the 3x3 region centered on it in the input. If the difference
38608 between input pixel and local maximum is more then threshold of the
38609 corresponding plane, output pixel will be set to input pixel +
38610 threshold of corresponding plane.
38611 -i INPUT -vf "hwupload, dilation_opencl=30:40:50:coordinates=231, hwdownload" OUTPUT
38613 nlmeans_opencl
38615 Non-local Means denoise filter through OpenCL, this filter accepts same
38616 options as nlmeans.
38617 overlay_opencl
38619 Overlay one video on top of another.
38620 It takes two inputs and has one output. The first input is the "main"
38622 video on which the second input is overlaid. This filter requires same
38623 memory layout for all the inputs. So, format conversion may be needed.
38624 The filter accepts the following options:
38626 x Set the x coordinate of the overlaid video on the main video.
38628 Default value is 0.
38629 y Set the y coordinate of the overlaid video on the main video.
38631 Default value is 0.
38632 Examples
38634 • Overlay an image LOGO at the top-left corner of the INPUT video.
38636 Both inputs are yuv420p format.
38637 -i INPUT -i LOGO -filter_complex "[0:v]hwupload[a], [1:v]format=yuv420p, hwupload[b], [a][b]overlay_opencl, hwdownload" OUTPUT
38639 • The inputs have same memory layout for color channels , the overlay
38641 has additional alpha plane, like INPUT is yuv420p, and the LOGO is
38642 yuva420p.
38643 -i INPUT -i LOGO -filter_complex "[0:v]hwupload[a], [1:v]format=yuva420p, hwupload[b], [a][b]overlay_opencl, hwdownload" OUTPUT
38645 pad_opencl
38647 Add paddings to the input image, and place the original input at the
38648 provided x, y coordinates.
38649 It accepts the following options:
38651 width, w
38653 height, h
38654 Specify an expression for the size of the output image with the
38655 paddings added. If the value for width or height is 0, the
38656 corresponding input size is used for the output.
38657 The width expression can reference the value set by the height
38659 expression, and vice versa.
38660 The default value of width and height is 0.
38662 x
38664 y Specify the offsets to place the input image at within the padded
38665 area, with respect to the top/left border of the output image.
38666 The x expression can reference the value set by the y expression,
38668 and vice versa.
38669 The default value of x and y is 0.
38671 If x or y evaluate to a negative number, they'll be changed so the
38673 input image is centered on the padded area.
38674 color
38676 Specify the color of the padded area. For the syntax of this
38677 option, check the "Color" section in the ffmpeg-utils manual.
38678 aspect
38680 Pad to an aspect instead to a resolution.
38681 The value for the width, height, x, and y options are expressions
38683 containing the following constants:
38684 in_w
38686 in_h
38687 The input video width and height.
38688 iw
38690 ih These are the same as in_w and in_h.
38691 out_w
38693 out_h
38694 The output width and height (the size of the padded area), as
38695 specified by the width and height expressions.
38696 ow
38698 oh These are the same as out_w and out_h.
38699 x
38701 y The x and y offsets as specified by the x and y expressions, or NAN
38702 if not yet specified.
38703 a same as iw / ih
38705 sar input sample aspect ratio
38707 dar input display aspect ratio, it is the same as (iw / ih) * sar
38709 prewitt_opencl
38711 Apply the Prewitt operator
38712 (<https://en.wikipedia.org/wiki/Prewitt_operator>) to input video
38713 stream.
38714 The filter accepts the following option:
38716 planes
38718 Set which planes to filter. Default value is 0xf, by which all
38719 planes are processed.
38720 scale
38722 Set value which will be multiplied with filtered result. Range is
38723 "[0.0, 65535]" and default value is 1.0.
38724 delta
38726 Set value which will be added to filtered result. Range is
38727 "[-65535, 65535]" and default value is 0.0.
38728 Example
38730 • Apply the Prewitt operator with scale set to 2 and delta set to 10.
38732 -i INPUT -vf "hwupload, prewitt_opencl=scale=2:delta=10, hwdownload" OUTPUT
38734 program_opencl
38736 Filter video using an OpenCL program.
38737 source
38739 OpenCL program source file.
38740 kernel
38742 Kernel name in program.
38743 inputs
38745 Number of inputs to the filter. Defaults to 1.
38746 size, s
38748 Size of output frames. Defaults to the same as the first input.
38749 The "program_opencl" filter also supports the framesync options.
38751 The program source file must contain a kernel function with the given
38753 name, which will be run once for each plane of the output. Each run on
38754 a plane gets enqueued as a separate 2D global NDRange with one work-
38755 item for each pixel to be generated. The global ID offset for each
38756 work-item is therefore the coordinates of a pixel in the destination
38757 image.
38758 The kernel function needs to take the following arguments:
38760 • Destination image, __write_only image2d_t.
38762 This image will become the output; the kernel should write all of
38764 it.
38765 • Frame index, unsigned int.
38767 This is a counter starting from zero and increasing by one for each
38769 frame.
38770 • Source images, __read_only image2d_t.
38772 These are the most recent images on each input. The kernel may
38774 read from them to generate the output, but they can't be written
38775 to.
38776 Example programs:
38778 • Copy the input to the output (output must be the same size as the
38780 input).
38781 __kernel void copy(__write_only image2d_t destination,
38783 unsigned int index,
38784 __read_only image2d_t source)
38785 {
38786 const sampler_t sampler = CLK_NORMALIZED_COORDS_FALSE;
38787 int2 location = (int2)(get_global_id(0), get_global_id(1));
38789 float4 value = read_imagef(source, sampler, location);
38791 write_imagef(destination, location, value);
38793 }
38794 • Apply a simple transformation, rotating the input by an amount
38796 increasing with the index counter. Pixel values are linearly
38797 interpolated by the sampler, and the output need not have the same
38798 dimensions as the input.
38799 __kernel void rotate_image(__write_only image2d_t dst,
38801 unsigned int index,
38802 __read_only image2d_t src)
38803 {
38804 const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE |
38805 CLK_FILTER_LINEAR);
38806 float angle = (float)index / 100.0f;
38808 float2 dst_dim = convert_float2(get_image_dim(dst));
38810 float2 src_dim = convert_float2(get_image_dim(src));
38811 float2 dst_cen = dst_dim / 2.0f;
38813 float2 src_cen = src_dim / 2.0f;
38814 int2 dst_loc = (int2)(get_global_id(0), get_global_id(1));
38816 float2 dst_pos = convert_float2(dst_loc) - dst_cen;
38818 float2 src_pos = {
38819 cos(angle) * dst_pos.x - sin(angle) * dst_pos.y,
38820 sin(angle) * dst_pos.x + cos(angle) * dst_pos.y
38821 };
38822 src_pos = src_pos * src_dim / dst_dim;
38823 float2 src_loc = src_pos + src_cen;
38825 if (src_loc.x < 0.0f || src_loc.y < 0.0f ||
38827 src_loc.x > src_dim.x || src_loc.y > src_dim.y)
38828 write_imagef(dst, dst_loc, 0.5f);
38829 else
38830 write_imagef(dst, dst_loc, read_imagef(src, sampler, src_loc));
38831 }
38832 • Blend two inputs together, with the amount of each input used
38834 varying with the index counter.
38835 __kernel void blend_images(__write_only image2d_t dst,
38837 unsigned int index,
38838 __read_only image2d_t src1,
38839 __read_only image2d_t src2)
38840 {
38841 const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE |
38842 CLK_FILTER_LINEAR);
38843 float blend = (cos((float)index / 50.0f) + 1.0f) / 2.0f;
38845 int2 dst_loc = (int2)(get_global_id(0), get_global_id(1));
38847 int2 src1_loc = dst_loc * get_image_dim(src1) / get_image_dim(dst);
38848 int2 src2_loc = dst_loc * get_image_dim(src2) / get_image_dim(dst);
38849 float4 val1 = read_imagef(src1, sampler, src1_loc);
38851 float4 val2 = read_imagef(src2, sampler, src2_loc);
38852 write_imagef(dst, dst_loc, val1 * blend + val2 * (1.0f - blend));
38854 }
38855 roberts_opencl
38857 Apply the Roberts cross operator
38858 (<https://en.wikipedia.org/wiki/Roberts_cross>) to input video stream.
38859 The filter accepts the following option:
38861 planes
38863 Set which planes to filter. Default value is 0xf, by which all
38864 planes are processed.
38865 scale
38867 Set value which will be multiplied with filtered result. Range is
38868 "[0.0, 65535]" and default value is 1.0.
38869 delta
38871 Set value which will be added to filtered result. Range is
38872 "[-65535, 65535]" and default value is 0.0.
38873 Example
38875 • Apply the Roberts cross operator with scale set to 2 and delta set
38877 to 10
38878 -i INPUT -vf "hwupload, roberts_opencl=scale=2:delta=10, hwdownload" OUTPUT
38880 sobel_opencl
38882 Apply the Sobel operator
38883 (<https://en.wikipedia.org/wiki/Sobel_operator>) to input video stream.
38884 The filter accepts the following option:
38886 planes
38888 Set which planes to filter. Default value is 0xf, by which all
38889 planes are processed.
38890 scale
38892 Set value which will be multiplied with filtered result. Range is
38893 "[0.0, 65535]" and default value is 1.0.
38894 delta
38896 Set value which will be added to filtered result. Range is
38897 "[-65535, 65535]" and default value is 0.0.
38898 Example
38900 • Apply sobel operator with scale set to 2 and delta set to 10
38902 -i INPUT -vf "hwupload, sobel_opencl=scale=2:delta=10, hwdownload" OUTPUT
38904 tonemap_opencl
38906 Perform HDR(PQ/HLG) to SDR conversion with tone-mapping.
38907 It accepts the following parameters:
38909 tonemap
38911 Specify the tone-mapping operator to be used. Same as tonemap
38912 option in tonemap.
38913 param
38915 Tune the tone mapping algorithm. same as param option in tonemap.
38916 desat
38918 Apply desaturation for highlights that exceed this level of
38919 brightness. The higher the parameter, the more color information
38920 will be preserved. This setting helps prevent unnaturally blown-out
38921 colors for super-highlights, by (smoothly) turning into white
38922 instead. This makes images feel more natural, at the cost of
38923 reducing information about out-of-range colors.
38924 The default value is 0.5, and the algorithm here is a little
38926 different from the cpu version tonemap currently. A setting of 0.0
38927 disables this option.
38928 threshold
38930 The tonemapping algorithm parameters is fine-tuned per each scene.
38931 And a threshold is used to detect whether the scene has changed or
38932 not. If the distance between the current frame average brightness
38933 and the current running average exceeds a threshold value, we would
38934 re-calculate scene average and peak brightness. The default value
38935 is 0.2.
38936 format
38938 Specify the output pixel format.
38939 Currently supported formats are:
38941 p010
38943 nv12
38944 range, r
38945 Set the output color range.
38946 Possible values are:
38948 tv/mpeg
38950 pc/jpeg
38951 Default is same as input.
38953 primaries, p
38955 Set the output color primaries.
38956 Possible values are:
38958 bt709
38960 bt2020
38961 Default is same as input.
38963 transfer, t
38965 Set the output transfer characteristics.
38966 Possible values are:
38968 bt709
38970 bt2020
38971 Default is bt709.
38973 matrix, m
38975 Set the output colorspace matrix.
38976 Possible value are:
38978 bt709
38980 bt2020
38981 Default is same as input.
38983 Example
38985 • Convert HDR(PQ/HLG) video to bt2020-transfer-characteristic p010
38987 format using linear operator.
38988 -i INPUT -vf "format=p010,hwupload,tonemap_opencl=t=bt2020:tonemap=linear:format=p010,hwdownload,format=p010" OUTPUT
38990 unsharp_opencl
38992 Sharpen or blur the input video.
38993 It accepts the following parameters:
38995 luma_msize_x, lx
38997 Set the luma matrix horizontal size. Range is "[1, 23]" and
38998 default value is 5.
38999 luma_msize_y, ly
39001 Set the luma matrix vertical size. Range is "[1, 23]" and default
39002 value is 5.
39003 luma_amount, la
39005 Set the luma effect strength. Range is "[-10, 10]" and default
39006 value is 1.0.
39007 Negative values will blur the input video, while positive values
39009 will sharpen it, a value of zero will disable the effect.
39010 chroma_msize_x, cx
39012 Set the chroma matrix horizontal size. Range is "[1, 23]" and
39013 default value is 5.
39014 chroma_msize_y, cy
39016 Set the chroma matrix vertical size. Range is "[1, 23]" and
39017 default value is 5.
39018 chroma_amount, ca
39020 Set the chroma effect strength. Range is "[-10, 10]" and default
39021 value is 0.0.
39022 Negative values will blur the input video, while positive values
39024 will sharpen it, a value of zero will disable the effect.
39025 All parameters are optional and default to the equivalent of the string
39027 '5:5:1.0:5:5:0.0'.
39028 Examples
39030 • Apply strong luma sharpen effect:
39032 -i INPUT -vf "hwupload, unsharp_opencl=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5, hwdownload" OUTPUT
39034 • Apply a strong blur of both luma and chroma parameters:
39036 -i INPUT -vf "hwupload, unsharp_opencl=7:7:-2:7:7:-2, hwdownload" OUTPUT
39038 xfade_opencl
39040 Cross fade two videos with custom transition effect by using OpenCL.
39041 It accepts the following options:
39043 transition
39045 Set one of possible transition effects.
39046 custom
39048 Select custom transition effect, the actual transition
39049 description will be picked from source and kernel options.
39050 fade
39052 wipeleft
39053 wiperight
39054 wipeup
39055 wipedown
39056 slideleft
39057 slideright
39058 slideup
39059 slidedown
39060 Default transition is fade.
39061 source
39063 OpenCL program source file for custom transition.
39064 kernel
39066 Set name of kernel to use for custom transition from program source
39067 file.
39068 duration
39070 Set duration of video transition.
39071 offset
39073 Set time of start of transition relative to first video.
39074 The program source file must contain a kernel function with the given
39076 name, which will be run once for each plane of the output. Each run on
39077 a plane gets enqueued as a separate 2D global NDRange with one work-
39078 item for each pixel to be generated. The global ID offset for each
39079 work-item is therefore the coordinates of a pixel in the destination
39080 image.
39081 The kernel function needs to take the following arguments:
39083 • Destination image, __write_only image2d_t.
39085 This image will become the output; the kernel should write all of
39087 it.
39088 • First Source image, __read_only image2d_t. Second Source image,
39090 __read_only image2d_t.
39091 These are the most recent images on each input. The kernel may
39093 read from them to generate the output, but they can't be written
39094 to.
39095 • Transition progress, float. This value is always between 0 and 1
39097 inclusive.
39098 Example programs:
39100 • Apply dots curtain transition effect:
39102 __kernel void blend_images(__write_only image2d_t dst,
39104 __read_only image2d_t src1,
39105 __read_only image2d_t src2,
39106 float progress)
39107 {
39108 const sampler_t sampler = (CLK_NORMALIZED_COORDS_FALSE |
39109 CLK_FILTER_LINEAR);
39110 int2 p = (int2)(get_global_id(0), get_global_id(1));
39111 float2 rp = (float2)(get_global_id(0), get_global_id(1));
39112 float2 dim = (float2)(get_image_dim(src1).x, get_image_dim(src1).y);
39113 rp = rp / dim;
39114 float2 dots = (float2)(20.0, 20.0);
39116 float2 center = (float2)(0,0);
39117 float2 unused;
39118 float4 val1 = read_imagef(src1, sampler, p);
39120 float4 val2 = read_imagef(src2, sampler, p);
39121 bool next = distance(fract(rp * dots, &unused), (float2)(0.5, 0.5)) < (progress / distance(rp, center));
39122 write_imagef(dst, p, next ? val1 : val2);
39124 }
39125
39127 VAAPI Video filters are usually used with VAAPI decoder and VAAPI
39128 encoder. Below is a description of VAAPI video filters.
39129 To enable compilation of these filters you need to configure FFmpeg
39131 with "--enable-vaapi".
39132 To use vaapi filters, you need to setup the vaapi device correctly. For
39134 more information, please read
39135 <https://trac.ffmpeg.org/wiki/Hardware/VAAPI>
39136 tonemap_vaapi
39138 Perform HDR(High Dynamic Range) to SDR(Standard Dynamic Range)
39139 conversion with tone-mapping. It maps the dynamic range of HDR10
39140 content to the SDR content. It currently only accepts HDR10 as input.
39141 It accepts the following parameters:
39143 format
39145 Specify the output pixel format.
39146 Currently supported formats are:
39148 p010
39150 nv12
39151 Default is nv12.
39153 primaries, p
39155 Set the output color primaries.
39156 Default is same as input.
39158 transfer, t
39160 Set the output transfer characteristics.
39161 Default is bt709.
39163 matrix, m
39165 Set the output colorspace matrix.
39166 Default is same as input.
39168 Example
39170 • Convert HDR(HDR10) video to bt2020-transfer-characteristic p010
39172 format
39173 tonemap_vaapi=format=p010:t=bt2020-10
39175
39177 Below is a description of the currently available video sources.
39178 buffer
39180 Buffer video frames, and make them available to the filter chain.
39181 This source is mainly intended for a programmatic use, in particular
39183 through the interface defined in libavfilter/buffersrc.h.
39184 It accepts the following parameters:
39186 video_size
39188 Specify the size (width and height) of the buffered video frames.
39189 For the syntax of this option, check the "Video size" section in
39190 the ffmpeg-utils manual.
39191 width
39193 The input video width.
39194 height
39196 The input video height.
39197 pix_fmt
39199 A string representing the pixel format of the buffered video
39200 frames. It may be a number corresponding to a pixel format, or a
39201 pixel format name.
39202 time_base
39204 Specify the timebase assumed by the timestamps of the buffered
39205 frames.
39206 frame_rate
39208 Specify the frame rate expected for the video stream.
39209 pixel_aspect, sar
39211 The sample (pixel) aspect ratio of the input video.
39212 sws_param
39214 This option is deprecated and ignored. Prepend "sws_flags=flags;"
39215 to the filtergraph description to specify swscale flags for
39216 automatically inserted scalers. See Filtergraph syntax.
39217 hw_frames_ctx
39219 When using a hardware pixel format, this should be a reference to
39220 an AVHWFramesContext describing input frames.
39221 For example:
39223 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
39225 will instruct the source to accept video frames with size 320x240 and
39227 with format "yuv410p", assuming 1/24 as the timestamps timebase and
39228 square pixels (1:1 sample aspect ratio). Since the pixel format with
39229 name "yuv410p" corresponds to the number 6 (check the enum
39230 AVPixelFormat definition in libavutil/pixfmt.h), this example
39231 corresponds to:
39232 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
39234 Alternatively, the options can be specified as a flat string, but this
39236 syntax is deprecated:
39237 width:height:pix_fmt:time_base.num:time_base.den:pixel_aspect.num:pixel_aspect.den
39239 cellauto
39241 Create a pattern generated by an elementary cellular automaton.
39242 The initial state of the cellular automaton can be defined through the
39244 filename and pattern options. If such options are not specified an
39245 initial state is created randomly.
39246 At each new frame a new row in the video is filled with the result of
39248 the cellular automaton next generation. The behavior when the whole
39249 frame is filled is defined by the scroll option.
39250 This source accepts the following options:
39252 filename, f
39254 Read the initial cellular automaton state, i.e. the starting row,
39255 from the specified file. In the file, each non-whitespace
39256 character is considered an alive cell, a newline will terminate the
39257 row, and further characters in the file will be ignored.
39258 pattern, p
39260 Read the initial cellular automaton state, i.e. the starting row,
39261 from the specified string.
39262 Each non-whitespace character in the string is considered an alive
39264 cell, a newline will terminate the row, and further characters in
39265 the string will be ignored.
39266 rate, r
39268 Set the video rate, that is the number of frames generated per
39269 second. Default is 25.
39270 random_fill_ratio, ratio
39272 Set the random fill ratio for the initial cellular automaton row.
39273 It is a floating point number value ranging from 0 to 1, defaults
39274 to 1/PHI.
39275 This option is ignored when a file or a pattern is specified.
39277 random_seed, seed
39279 Set the seed for filling randomly the initial row, must be an
39280 integer included between 0 and UINT32_MAX. If not specified, or if
39281 explicitly set to -1, the filter will try to use a good random seed
39282 on a best effort basis.
39283 rule
39285 Set the cellular automaton rule, it is a number ranging from 0 to
39286 255. Default value is 110.
39287 size, s
39289 Set the size of the output video. For the syntax of this option,
39290 check the "Video size" section in the ffmpeg-utils manual.
39291 If filename or pattern is specified, the size is set by default to
39293 the width of the specified initial state row, and the height is set
39294 to width * PHI.
39295 If size is set, it must contain the width of the specified pattern
39297 string, and the specified pattern will be centered in the larger
39298 row.
39299 If a filename or a pattern string is not specified, the size value
39301 defaults to "320x518" (used for a randomly generated initial
39302 state).
39303 scroll
39305 If set to 1, scroll the output upward when all the rows in the
39306 output have been already filled. If set to 0, the new generated row
39307 will be written over the top row just after the bottom row is
39308 filled. Defaults to 1.
39309 start_full, full
39311 If set to 1, completely fill the output with generated rows before
39312 outputting the first frame. This is the default behavior, for
39313 disabling set the value to 0.
39314 stitch
39316 If set to 1, stitch the left and right row edges together. This is
39317 the default behavior, for disabling set the value to 0.
39318 Examples
39320 • Read the initial state from pattern, and specify an output of size
39322 200x400.
39323 cellauto=f=pattern:s=200x400
39325 • Generate a random initial row with a width of 200 cells, with a
39327 fill ratio of 2/3:
39328 cellauto=ratio=2/3:s=200x200
39330 • Create a pattern generated by rule 18 starting by a single alive
39332 cell centered on an initial row with width 100:
39333 cellauto=p=@s=100x400:full=0:rule=18
39335 • Specify a more elaborated initial pattern:
39337 cellauto=p='@@ @ @@':s=100x400:full=0:rule=18
39339 coreimagesrc
39341 Video source generated on GPU using Apple's CoreImage API on OSX.
39342 This video source is a specialized version of the coreimage video
39344 filter. Use a core image generator at the beginning of the applied
39345 filterchain to generate the content.
39346 The coreimagesrc video source accepts the following options:
39348 list_generators
39350 List all available generators along with all their respective
39351 options as well as possible minimum and maximum values along with
39352 the default values.
39353 list_generators=true
39355 size, s
39357 Specify the size of the sourced video. For the syntax of this
39358 option, check the "Video size" section in the ffmpeg-utils manual.
39359 The default value is "320x240".
39360 rate, r
39362 Specify the frame rate of the sourced video, as the number of
39363 frames generated per second. It has to be a string in the format
39364 frame_rate_num/frame_rate_den, an integer number, a floating point
39365 number or a valid video frame rate abbreviation. The default value
39366 is "25".
39367 sar Set the sample aspect ratio of the sourced video.
39369 duration, d
39371 Set the duration of the sourced video. See the Time duration
39372 section in the ffmpeg-utils(1) manual for the accepted syntax.
39373 If not specified, or the expressed duration is negative, the video
39375 is supposed to be generated forever.
39376 Additionally, all options of the coreimage video filter are accepted.
39378 A complete filterchain can be used for further processing of the
39379 generated input without CPU-HOST transfer. See coreimage documentation
39380 and examples for details.
39381 Examples
39383 • Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
39385 given as complete and escaped command-line for Apple's standard
39386 bash shell:
39387 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@inputMessage=https\\\\\://FFmpeg.org/@inputCorrectionLevel=H -frames:v 1 QRCode.png
39389 This example is equivalent to the QRCode example of coreimage
39391 without the need for a nullsrc video source.
39392 gradients
39394 Generate several gradients.
39395 size, s
39397 Set frame size. For the syntax of this option, check the "Video
39398 size" section in the ffmpeg-utils manual. Default value is
39399 "640x480".
39400 rate, r
39402 Set frame rate, expressed as number of frames per second. Default
39403 value is "25".
39404 c0, c1, c2, c3, c4, c5, c6, c7
39406 Set 8 colors. Default values for colors is to pick random one.
39407 x0, y0, y0, y1
39409 Set gradient line source and destination points. If negative or out
39410 of range, random ones are picked.
39411 nb_colors, n
39413 Set number of colors to use at once. Allowed range is from 2 to 8.
39414 Default value is 2.
39415 seed
39417 Set seed for picking gradient line points.
39418 duration, d
39420 Set the duration of the sourced video. See the Time duration
39421 section in the ffmpeg-utils(1) manual for the accepted syntax.
39422 If not specified, or the expressed duration is negative, the video
39424 is supposed to be generated forever.
39425 speed
39427 Set speed of gradients rotation.
39428 mandelbrot
39430 Generate a Mandelbrot set fractal, and progressively zoom towards the
39431 point specified with start_x and start_y.
39432 This source accepts the following options:
39434 end_pts
39436 Set the terminal pts value. Default value is 400.
39437 end_scale
39439 Set the terminal scale value. Must be a floating point value.
39440 Default value is 0.3.
39441 inner
39443 Set the inner coloring mode, that is the algorithm used to draw the
39444 Mandelbrot fractal internal region.
39445 It shall assume one of the following values:
39447 black
39449 Set black mode.
39450 convergence
39452 Show time until convergence.
39453 mincol
39455 Set color based on point closest to the origin of the
39456 iterations.
39457 period
39459 Set period mode.
39460 Default value is mincol.
39462 bailout
39464 Set the bailout value. Default value is 10.0.
39465 maxiter
39467 Set the maximum of iterations performed by the rendering algorithm.
39468 Default value is 7189.
39469 outer
39471 Set outer coloring mode. It shall assume one of following values:
39472 iteration_count
39474 Set iteration count mode.
39475 normalized_iteration_count
39477 set normalized iteration count mode.
39478 Default value is normalized_iteration_count.
39480 rate, r
39482 Set frame rate, expressed as number of frames per second. Default
39483 value is "25".
39484 size, s
39486 Set frame size. For the syntax of this option, check the "Video
39487 size" section in the ffmpeg-utils manual. Default value is
39488 "640x480".
39489 start_scale
39491 Set the initial scale value. Default value is 3.0.
39492 start_x
39494 Set the initial x position. Must be a floating point value between
39495 -100 and 100. Default value is
39496 -0.743643887037158704752191506114774.
39497 start_y
39499 Set the initial y position. Must be a floating point value between
39500 -100 and 100. Default value is
39501 -0.131825904205311970493132056385139.
39502 mptestsrc
39504 Generate various test patterns, as generated by the MPlayer test
39505 filter.
39506 The size of the generated video is fixed, and is 256x256. This source
39508 is useful in particular for testing encoding features.
39509 This source accepts the following options:
39511 rate, r
39513 Specify the frame rate of the sourced video, as the number of
39514 frames generated per second. It has to be a string in the format
39515 frame_rate_num/frame_rate_den, an integer number, a floating point
39516 number or a valid video frame rate abbreviation. The default value
39517 is "25".
39518 duration, d
39520 Set the duration of the sourced video. See the Time duration
39521 section in the ffmpeg-utils(1) manual for the accepted syntax.
39522 If not specified, or the expressed duration is negative, the video
39524 is supposed to be generated forever.
39525 test, t
39527 Set the number or the name of the test to perform. Supported tests
39528 are:
39529 dc_luma
39531 dc_chroma
39532 freq_luma
39533 freq_chroma
39534 amp_luma
39535 amp_chroma
39536 cbp
39537 mv
39538 ring1
39539 ring2
39540 all
39541 max_frames, m
39542 Set the maximum number of frames generated for each test,
39543 default value is 30.
39544 Default value is "all", which will cycle through the list of all
39546 tests.
39547 Some examples:
39549 mptestsrc=t=dc_luma
39551 will generate a "dc_luma" test pattern.
39553 frei0r_src
39555 Provide a frei0r source.
39556 To enable compilation of this filter you need to install the frei0r
39558 header and configure FFmpeg with "--enable-frei0r".
39559 This source accepts the following parameters:
39561 size
39563 The size of the video to generate. For the syntax of this option,
39564 check the "Video size" section in the ffmpeg-utils manual.
39565 framerate
39567 The framerate of the generated video. It may be a string of the
39568 form num/den or a frame rate abbreviation.
39569 filter_name
39571 The name to the frei0r source to load. For more information
39572 regarding frei0r and how to set the parameters, read the frei0r
39573 section in the video filters documentation.
39574 filter_params
39576 A '|'-separated list of parameters to pass to the frei0r source.
39577 For example, to generate a frei0r partik0l source with size 200x200 and
39579 frame rate 10 which is overlaid on the overlay filter main input:
39580 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
39582 life
39584 Generate a life pattern.
39585 This source is based on a generalization of John Conway's life game.
39587 The sourced input represents a life grid, each pixel represents a cell
39589 which can be in one of two possible states, alive or dead. Every cell
39590 interacts with its eight neighbours, which are the cells that are
39591 horizontally, vertically, or diagonally adjacent.
39592 At each interaction the grid evolves according to the adopted rule,
39594 which specifies the number of neighbor alive cells which will make a
39595 cell stay alive or born. The rule option allows one to specify the rule
39596 to adopt.
39597 This source accepts the following options:
39599 filename, f
39601 Set the file from which to read the initial grid state. In the
39602 file, each non-whitespace character is considered an alive cell,
39603 and newline is used to delimit the end of each row.
39604 If this option is not specified, the initial grid is generated
39606 randomly.
39607 rate, r
39609 Set the video rate, that is the number of frames generated per
39610 second. Default is 25.
39611 random_fill_ratio, ratio
39613 Set the random fill ratio for the initial random grid. It is a
39614 floating point number value ranging from 0 to 1, defaults to 1/PHI.
39615 It is ignored when a file is specified.
39616 random_seed, seed
39618 Set the seed for filling the initial random grid, must be an
39619 integer included between 0 and UINT32_MAX. If not specified, or if
39620 explicitly set to -1, the filter will try to use a good random seed
39621 on a best effort basis.
39622 rule
39624 Set the life rule.
39625 A rule can be specified with a code of the kind "SNS/BNB", where NS
39627 and NB are sequences of numbers in the range 0-8, NS specifies the
39628 number of alive neighbor cells which make a live cell stay alive,
39629 and NB the number of alive neighbor cells which make a dead cell to
39630 become alive (i.e. to "born"). "s" and "b" can be used in place of
39631 "S" and "B", respectively.
39632 Alternatively a rule can be specified by an 18-bits integer. The 9
39634 high order bits are used to encode the next cell state if it is
39635 alive for each number of neighbor alive cells, the low order bits
39636 specify the rule for "borning" new cells. Higher order bits encode
39637 for an higher number of neighbor cells. For example the number
39638 6153 = "(12<<9)+9" specifies a stay alive rule of 12 and a born
39639 rule of 9, which corresponds to "S23/B03".
39640 Default value is "S23/B3", which is the original Conway's game of
39642 life rule, and will keep a cell alive if it has 2 or 3 neighbor
39643 alive cells, and will born a new cell if there are three alive
39644 cells around a dead cell.
39645 size, s
39647 Set the size of the output video. For the syntax of this option,
39648 check the "Video size" section in the ffmpeg-utils manual.
39649 If filename is specified, the size is set by default to the same
39651 size of the input file. If size is set, it must contain the size
39652 specified in the input file, and the initial grid defined in that
39653 file is centered in the larger resulting area.
39654 If a filename is not specified, the size value defaults to
39656 "320x240" (used for a randomly generated initial grid).
39657 stitch
39659 If set to 1, stitch the left and right grid edges together, and the
39660 top and bottom edges also. Defaults to 1.
39661 mold
39663 Set cell mold speed. If set, a dead cell will go from death_color
39664 to mold_color with a step of mold. mold can have a value from 0 to
39665 255.
39666 life_color
39668 Set the color of living (or new born) cells.
39669 death_color
39671 Set the color of dead cells. If mold is set, this is the first
39672 color used to represent a dead cell.
39673 mold_color
39675 Set mold color, for definitely dead and moldy cells.
39676 For the syntax of these 3 color options, check the "Color" section
39678 in the ffmpeg-utils manual.
39679 Examples
39681 • Read a grid from pattern, and center it on a grid of size 300x300
39683 pixels:
39684 life=f=pattern:s=300x300
39686 • Generate a random grid of size 200x200, with a fill ratio of 2/3:
39688 life=ratio=2/3:s=200x200
39690 • Specify a custom rule for evolving a randomly generated grid:
39692 life=rule=S14/B34
39694 • Full example with slow death effect (mold) using ffplay:
39696 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
39698 allrgb, allyuv, color, colorspectrum, haldclutsrc, nullsrc, pal75bars,
39700 pal100bars, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2,
39701 yuvtestsrc
39702 The "allrgb" source returns frames of size 4096x4096 of all rgb colors.
39703 The "allyuv" source returns frames of size 4096x4096 of all yuv colors.
39705 The "color" source provides an uniformly colored input.
39707 The "colorspectrum" source provides a color spectrum input.
39709 The "haldclutsrc" source provides an identity Hald CLUT. See also
39711 haldclut filter.
39712 The "nullsrc" source returns unprocessed video frames. It is mainly
39714 useful to be employed in analysis / debugging tools, or as the source
39715 for filters which ignore the input data.
39716 The "pal75bars" source generates a color bars pattern, based on EBU PAL
39718 recommendations with 75% color levels.
39719 The "pal100bars" source generates a color bars pattern, based on EBU
39721 PAL recommendations with 100% color levels.
39722 The "rgbtestsrc" source generates an RGB test pattern useful for
39724 detecting RGB vs BGR issues. You should see a red, green and blue
39725 stripe from top to bottom.
39726 The "smptebars" source generates a color bars pattern, based on the
39728 SMPTE Engineering Guideline EG 1-1990.
39729 The "smptehdbars" source generates a color bars pattern, based on the
39731 SMPTE RP 219-2002.
39732 The "testsrc" source generates a test video pattern, showing a color
39734 pattern, a scrolling gradient and a timestamp. This is mainly intended
39735 for testing purposes.
39736 The "testsrc2" source is similar to testsrc, but supports more pixel
39738 formats instead of just "rgb24". This allows using it as an input for
39739 other tests without requiring a format conversion.
39740 The "yuvtestsrc" source generates an YUV test pattern. You should see a
39742 y, cb and cr stripe from top to bottom.
39743 The sources accept the following parameters:
39745 level
39747 Specify the level of the Hald CLUT, only available in the
39748 "haldclutsrc" source. A level of "N" generates a picture of "N*N*N"
39749 by "N*N*N" pixels to be used as identity matrix for 3D lookup
39750 tables. Each component is coded on a "1/(N*N)" scale.
39751 color, c
39753 Specify the color of the source, only available in the "color"
39754 source. For the syntax of this option, check the "Color" section in
39755 the ffmpeg-utils manual.
39756 size, s
39758 Specify the size of the sourced video. For the syntax of this
39759 option, check the "Video size" section in the ffmpeg-utils manual.
39760 The default value is "320x240".
39761 This option is not available with the "allrgb", "allyuv", and
39763 "haldclutsrc" filters.
39764 rate, r
39766 Specify the frame rate of the sourced video, as the number of
39767 frames generated per second. It has to be a string in the format
39768 frame_rate_num/frame_rate_den, an integer number, a floating point
39769 number or a valid video frame rate abbreviation. The default value
39770 is "25".
39771 duration, d
39773 Set the duration of the sourced video. See the Time duration
39774 section in the ffmpeg-utils(1) manual for the accepted syntax.
39775 If not specified, or the expressed duration is negative, the video
39777 is supposed to be generated forever.
39778 Since the frame rate is used as time base, all frames including the
39780 last one will have their full duration. If the specified duration
39781 is not a multiple of the frame duration, it will be rounded up.
39782 sar Set the sample aspect ratio of the sourced video.
39784 alpha
39786 Specify the alpha (opacity) of the background, only available in
39787 the "testsrc2" source. The value must be between 0 (fully
39788 transparent) and 255 (fully opaque, the default).
39789 decimals, n
39791 Set the number of decimals to show in the timestamp, only available
39792 in the "testsrc" source.
39793 The displayed timestamp value will correspond to the original
39795 timestamp value multiplied by the power of 10 of the specified
39796 value. Default value is 0.
39797 type
39799 Set the type of the color spectrum, only available in the
39800 "colorspectrum" source. Can be one of the following:
39801 black
39803 white
39804 all
39805 Examples
39807 • Generate a video with a duration of 5.3 seconds, with size 176x144
39809 and a frame rate of 10 frames per second:
39810 testsrc=duration=5.3:size=qcif:rate=10
39812 • The following graph description will generate a red source with an
39814 opacity of 0.2, with size "qcif" and a frame rate of 10 frames per
39815 second:
39816 color=c=red@0.2:s=qcif:r=10
39818 • If the input content is to be ignored, "nullsrc" can be used. The
39820 following command generates noise in the luminance plane by
39821 employing the "geq" filter:
39822 nullsrc=s=256x256, geq=random(1)*255:128:128
39824 Commands
39826 The "color" source supports the following commands:
39828 c, color
39830 Set the color of the created image. Accepts the same syntax of the
39831 corresponding color option.
39832 openclsrc
39834 Generate video using an OpenCL program.
39835 source
39837 OpenCL program source file.
39838 kernel
39840 Kernel name in program.
39841 size, s
39843 Size of frames to generate. This must be set.
39844 format
39846 Pixel format to use for the generated frames. This must be set.
39847 rate, r
39849 Number of frames generated every second. Default value is '25'.
39850 For details of how the program loading works, see the program_opencl
39852 filter.
39853 Example programs:
39855 • Generate a colour ramp by setting pixel values from the position of
39857 the pixel in the output image. (Note that this will work with all
39858 pixel formats, but the generated output will not be the same.)
39859 __kernel void ramp(__write_only image2d_t dst,
39861 unsigned int index)
39862 {
39863 int2 loc = (int2)(get_global_id(0), get_global_id(1));
39864 float4 val;
39866 val.xy = val.zw = convert_float2(loc) / convert_float2(get_image_dim(dst));
39867 write_imagef(dst, loc, val);
39869 }
39870 • Generate a Sierpinski carpet pattern, panning by a single pixel
39872 each frame.
39873 __kernel void sierpinski_carpet(__write_only image2d_t dst,
39875 unsigned int index)
39876 {
39877 int2 loc = (int2)(get_global_id(0), get_global_id(1));
39878 float4 value = 0.0f;
39880 int x = loc.x + index;
39881 int y = loc.y + index;
39882 while (x > 0 || y > 0) {
39883 if (x % 3 == 1 && y % 3 == 1) {
39884 value = 1.0f;
39885 break;
39886 }
39887 x /= 3;
39888 y /= 3;
39889 }
39890 write_imagef(dst, loc, value);
39892 }
39893 sierpinski
39895 Generate a Sierpinski carpet/triangle fractal, and randomly pan around.
39896 This source accepts the following options:
39898 size, s
39900 Set frame size. For the syntax of this option, check the "Video
39901 size" section in the ffmpeg-utils manual. Default value is
39902 "640x480".
39903 rate, r
39905 Set frame rate, expressed as number of frames per second. Default
39906 value is "25".
39907 seed
39909 Set seed which is used for random panning.
39910 jump
39912 Set max jump for single pan destination. Allowed range is from 1 to
39913 10000.
39914 type
39916 Set fractal type, can be default "carpet" or "triangle".
39917
39919 Below is a description of the currently available video sinks.
39920 buffersink
39922 Buffer video frames, and make them available to the end of the filter
39923 graph.
39924 This sink is mainly intended for programmatic use, in particular
39926 through the interface defined in libavfilter/buffersink.h or the
39927 options system.
39928 It accepts a pointer to an AVBufferSinkContext structure, which defines
39930 the incoming buffers' formats, to be passed as the opaque parameter to
39931 "avfilter_init_filter" for initialization.
39932 nullsink
39934 Null video sink: do absolutely nothing with the input video. It is
39935 mainly useful as a template and for use in analysis / debugging tools.
39936
39938 Below is a description of the currently available multimedia filters.
39939 abitscope
39941 Convert input audio to a video output, displaying the audio bit scope.
39942 The filter accepts the following options:
39944 rate, r
39946 Set frame rate, expressed as number of frames per second. Default
39947 value is "25".
39948 size, s
39950 Specify the video size for the output. For the syntax of this
39951 option, check the "Video size" section in the ffmpeg-utils manual.
39952 Default value is "1024x256".
39953 colors
39955 Specify list of colors separated by space or by '|' which will be
39956 used to draw channels. Unrecognized or missing colors will be
39957 replaced by white color.
39958 adrawgraph
39960 Draw a graph using input audio metadata.
39961 See drawgraph
39963 agraphmonitor
39965 See graphmonitor.
39966 ahistogram
39968 Convert input audio to a video output, displaying the volume histogram.
39969 The filter accepts the following options:
39971 dmode
39973 Specify how histogram is calculated.
39974 It accepts the following values:
39976 single
39978 Use single histogram for all channels.
39979 separate
39981 Use separate histogram for each channel.
39982 Default is "single".
39984 rate, r
39986 Set frame rate, expressed as number of frames per second. Default
39987 value is "25".
39988 size, s
39990 Specify the video size for the output. For the syntax of this
39991 option, check the "Video size" section in the ffmpeg-utils manual.
39992 Default value is "hd720".
39993 scale
39995 Set display scale.
39996 It accepts the following values:
39998 log logarithmic
40000 sqrt
40002 square root
40003 cbrt
40005 cubic root
40006 lin linear
40008 rlog
40010 reverse logarithmic
40011 Default is "log".
40013 ascale
40015 Set amplitude scale.
40016 It accepts the following values:
40018 log logarithmic
40020 lin linear
40022 Default is "log".
40024 acount
40026 Set how much frames to accumulate in histogram. Default is 1.
40027 Setting this to -1 accumulates all frames.
40028 rheight
40030 Set histogram ratio of window height.
40031 slide
40033 Set sonogram sliding.
40034 It accepts the following values:
40036 replace
40038 replace old rows with new ones.
40039 scroll
40041 scroll from top to bottom.
40042 Default is "replace".
40044 aphasemeter
40046 Measures phase of input audio, which is exported as metadata
40047 "lavfi.aphasemeter.phase", representing mean phase of current audio
40048 frame. A video output can also be produced and is enabled by default.
40049 The audio is passed through as first output.
40050 Audio will be rematrixed to stereo if it has a different channel
40052 layout. Phase value is in range "[-1, 1]" where "-1" means left and
40053 right channels are completely out of phase and 1 means channels are in
40054 phase.
40055 The filter accepts the following options, all related to its video
40057 output:
40058 rate, r
40060 Set the output frame rate. Default value is 25.
40061 size, s
40063 Set the video size for the output. For the syntax of this option,
40064 check the "Video size" section in the ffmpeg-utils manual. Default
40065 value is "800x400".
40066 rc
40068 gc
40069 bc Specify the red, green, blue contrast. Default values are 2, 7 and
40070 1. Allowed range is "[0, 255]".
40071 mpc Set color which will be used for drawing median phase. If color is
40073 "none" which is default, no median phase value will be drawn.
40074 video
40076 Enable video output. Default is enabled.
40077 phasing detection
40079 The filter also detects out of phase and mono sequences in stereo
40081 streams. It logs the sequence start, end and duration when it lasts
40082 longer or as long as the minimum set.
40083 The filter accepts the following options for this detection:
40085 phasing
40087 Enable mono and out of phase detection. Default is disabled.
40088 tolerance, t
40090 Set phase tolerance for mono detection, in amplitude ratio. Default
40091 is 0. Allowed range is "[0, 1]".
40092 angle, a
40094 Set angle threshold for out of phase detection, in degree. Default
40095 is 170. Allowed range is "[90, 180]".
40096 duration, d
40098 Set mono or out of phase duration until notification, expressed in
40099 seconds. Default is 2.
40100 Examples
40102 • Complete example with ffmpeg to detect 1 second of mono with 0.001
40104 phase tolerance:
40105 ffmpeg -i stereo.wav -af aphasemeter=video=0:phasing=1:duration=1:tolerance=0.001 -f null -
40107 avectorscope
40109 Convert input audio to a video output, representing the audio vector
40110 scope.
40111 The filter is used to measure the difference between channels of stereo
40113 audio stream. A monaural signal, consisting of identical left and right
40114 signal, results in straight vertical line. Any stereo separation is
40115 visible as a deviation from this line, creating a Lissajous figure. If
40116 the straight (or deviation from it) but horizontal line appears this
40117 indicates that the left and right channels are out of phase.
40118 The filter accepts the following options:
40120 mode, m
40122 Set the vectorscope mode.
40123 Available values are:
40125 lissajous
40127 Lissajous rotated by 45 degrees.
40128 lissajous_xy
40130 Same as above but not rotated.
40131 polar
40133 Shape resembling half of circle.
40134 Default value is lissajous.
40136 size, s
40138 Set the video size for the output. For the syntax of this option,
40139 check the "Video size" section in the ffmpeg-utils manual. Default
40140 value is "400x400".
40141 rate, r
40143 Set the output frame rate. Default value is 25.
40144 rc
40146 gc
40147 bc
40148 ac Specify the red, green, blue and alpha contrast. Default values are
40149 40, 160, 80 and 255. Allowed range is "[0, 255]".
40150 rf
40152 gf
40153 bf
40154 af Specify the red, green, blue and alpha fade. Default values are 15,
40155 10, 5 and 5. Allowed range is "[0, 255]".
40156 zoom
40158 Set the zoom factor. Default value is 1. Allowed range is "[0,
40159 10]". Values lower than 1 will auto adjust zoom factor to maximal
40160 possible value.
40161 draw
40163 Set the vectorscope drawing mode.
40164 Available values are:
40166 dot Draw dot for each sample.
40168 line
40170 Draw line between previous and current sample.
40171 Default value is dot.
40173 scale
40175 Specify amplitude scale of audio samples.
40176 Available values are:
40178 lin Linear.
40180 sqrt
40182 Square root.
40183 cbrt
40185 Cubic root.
40186 log Logarithmic.
40188 swap
40190 Swap left channel axis with right channel axis.
40191 mirror
40193 Mirror axis.
40194 none
40196 No mirror.
40197 x Mirror only x axis.
40199 y Mirror only y axis.
40201 xy Mirror both axis.
40203 Examples
40205 • Complete example using ffplay:
40207 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
40209 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
40210 Commands
40212 This filter supports the all above options as commands except options
40214 "size" and "rate".
40215 bench, abench
40217 Benchmark part of a filtergraph.
40218 The filter accepts the following options:
40220 action
40222 Start or stop a timer.
40223 Available values are:
40225 start
40227 Get the current time, set it as frame metadata (using the key
40228 "lavfi.bench.start_time"), and forward the frame to the next
40229 filter.
40230 stop
40232 Get the current time and fetch the "lavfi.bench.start_time"
40233 metadata from the input frame metadata to get the time
40234 difference. Time difference, average, maximum and minimum time
40235 (respectively "t", "avg", "max" and "min") are then printed.
40236 The timestamps are expressed in seconds.
40237 Examples
40239 • Benchmark selectivecolor filter:
40241 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
40243 concat
40245 Concatenate audio and video streams, joining them together one after
40246 the other.
40247 The filter works on segments of synchronized video and audio streams.
40249 All segments must have the same number of streams of each type, and
40250 that will also be the number of streams at output.
40251 The filter accepts the following options:
40253 n Set the number of segments. Default is 2.
40255 v Set the number of output video streams, that is also the number of
40257 video streams in each segment. Default is 1.
40258 a Set the number of output audio streams, that is also the number of
40260 audio streams in each segment. Default is 0.
40261 unsafe
40263 Activate unsafe mode: do not fail if segments have a different
40264 format.
40265 The filter has v+a outputs: first v video outputs, then a audio
40267 outputs.
40268 There are nx(v+a) inputs: first the inputs for the first segment, in
40270 the same order as the outputs, then the inputs for the second segment,
40271 etc.
40272 Related streams do not always have exactly the same duration, for
40274 various reasons including codec frame size or sloppy authoring. For
40275 that reason, related synchronized streams (e.g. a video and its audio
40276 track) should be concatenated at once. The concat filter will use the
40277 duration of the longest stream in each segment (except the last one),
40278 and if necessary pad shorter audio streams with silence.
40279 For this filter to work correctly, all segments must start at timestamp
40281 0.
40282 All corresponding streams must have the same parameters in all
40284 segments; the filtering system will automatically select a common pixel
40285 format for video streams, and a common sample format, sample rate and
40286 channel layout for audio streams, but other settings, such as
40287 resolution, must be converted explicitly by the user.
40288 Different frame rates are acceptable but will result in variable frame
40290 rate at output; be sure to configure the output file to handle it.
40291 Examples
40293 • Concatenate an opening, an episode and an ending, all in bilingual
40295 version (video in stream 0, audio in streams 1 and 2):
40296 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
40298 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
40299 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
40300 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
40301 • Concatenate two parts, handling audio and video separately, using
40303 the (a)movie sources, and adjusting the resolution:
40304 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
40306 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
40307 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
40308 Note that a desync will happen at the stitch if the audio and video
40310 streams do not have exactly the same duration in the first file.
40311 Commands
40313 This filter supports the following commands:
40315 next
40317 Close the current segment and step to the next one
40318 ebur128
40320 EBU R128 scanner filter. This filter takes an audio stream and analyzes
40321 its loudness level. By default, it logs a message at a frequency of
40322 10Hz with the Momentary loudness (identified by "M"), Short-term
40323 loudness ("S"), Integrated loudness ("I") and Loudness Range ("LRA").
40324 The filter can only analyze streams which have sample format is double-
40326 precision floating point. The input stream will be converted to this
40327 specification, if needed. Users may need to insert aformat and/or
40328 aresample filters after this filter to obtain the original parameters.
40329 The filter also has a video output (see the video option) with a real
40331 time graph to observe the loudness evolution. The graphic contains the
40332 logged message mentioned above, so it is not printed anymore when this
40333 option is set, unless the verbose logging is set. The main graphing
40334 area contains the short-term loudness (3 seconds of analysis), and the
40335 gauge on the right is for the momentary loudness (400 milliseconds),
40336 but can optionally be configured to instead display short-term loudness
40337 (see gauge).
40338 The green area marks a +/- 1LU target range around the target loudness
40340 (-23LUFS by default, unless modified through target).
40341 More information about the Loudness Recommendation EBU R128 on
40343 <http://tech.ebu.ch/loudness>.
40344 The filter accepts the following options:
40346 video
40348 Activate the video output. The audio stream is passed unchanged
40349 whether this option is set or no. The video stream will be the
40350 first output stream if activated. Default is 0.
40351 size
40353 Set the video size. This option is for video only. For the syntax
40354 of this option, check the "Video size" section in the ffmpeg-utils
40355 manual. Default and minimum resolution is "640x480".
40356 meter
40358 Set the EBU scale meter. Default is 9. Common values are 9 and 18,
40359 respectively for EBU scale meter +9 and EBU scale meter +18. Any
40360 other integer value between this range is allowed.
40361 metadata
40363 Set metadata injection. If set to 1, the audio input will be
40364 segmented into 100ms output frames, each of them containing various
40365 loudness information in metadata. All the metadata keys are
40366 prefixed with "lavfi.r128.".
40367 Default is 0.
40369 framelog
40371 Force the frame logging level.
40372 Available values are:
40374 info
40376 information logging level
40377 verbose
40379 verbose logging level
40380 By default, the logging level is set to info. If the video or the
40382 metadata options are set, it switches to verbose.
40383 peak
40385 Set peak mode(s).
40386 Available modes can be cumulated (the option is a "flag" type).
40388 Possible values are:
40389 none
40391 Disable any peak mode (default).
40392 sample
40394 Enable sample-peak mode.
40395 Simple peak mode looking for the higher sample value. It logs a
40397 message for sample-peak (identified by "SPK").
40398 true
40400 Enable true-peak mode.
40401 If enabled, the peak lookup is done on an over-sampled version
40403 of the input stream for better peak accuracy. It logs a message
40404 for true-peak. (identified by "TPK") and true-peak per frame
40405 (identified by "FTPK"). This mode requires a build with
40406 "libswresample".
40407 dualmono
40409 Treat mono input files as "dual mono". If a mono file is intended
40410 for playback on a stereo system, its EBU R128 measurement will be
40411 perceptually incorrect. If set to "true", this option will
40412 compensate for this effect. Multi-channel input files are not
40413 affected by this option.
40414 panlaw
40416 Set a specific pan law to be used for the measurement of dual mono
40417 files. This parameter is optional, and has a default value of
40418 -3.01dB.
40419 target
40421 Set a specific target level (in LUFS) used as relative zero in the
40422 visualization. This parameter is optional and has a default value
40423 of -23LUFS as specified by EBU R128. However, material published
40424 online may prefer a level of -16LUFS (e.g. for use with podcasts or
40425 video platforms).
40426 gauge
40428 Set the value displayed by the gauge. Valid values are "momentary"
40429 and s "shortterm". By default the momentary value will be used, but
40430 in certain scenarios it may be more useful to observe the short
40431 term value instead (e.g. live mixing).
40432 scale
40434 Sets the display scale for the loudness. Valid parameters are
40435 "absolute" (in LUFS) or "relative" (LU) relative to the target.
40436 This only affects the video output, not the summary or continuous
40437 log output.
40438 Examples
40440 • Real-time graph using ffplay, with a EBU scale meter +18:
40442 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
40444 • Run an analysis with ffmpeg:
40446 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
40448 interleave, ainterleave
40450 Temporally interleave frames from several inputs.
40451 "interleave" works with video inputs, "ainterleave" with audio.
40453 These filters read frames from several inputs and send the oldest
40455 queued frame to the output.
40456 Input streams must have well defined, monotonically increasing frame
40458 timestamp values.
40459 In order to submit one frame to output, these filters need to enqueue
40461 at least one frame for each input, so they cannot work in case one
40462 input is not yet terminated and will not receive incoming frames.
40463 For example consider the case when one input is a "select" filter which
40465 always drops input frames. The "interleave" filter will keep reading
40466 from that input, but it will never be able to send new frames to output
40467 until the input sends an end-of-stream signal.
40468 Also, depending on inputs synchronization, the filters will drop frames
40470 in case one input receives more frames than the other ones, and the
40471 queue is already filled.
40472 These filters accept the following options:
40474 nb_inputs, n
40476 Set the number of different inputs, it is 2 by default.
40477 duration
40479 How to determine the end-of-stream.
40480 longest
40482 The duration of the longest input. (default)
40483 shortest
40485 The duration of the shortest input.
40486 first
40488 The duration of the first input.
40489 Examples
40491 • Interleave frames belonging to different streams using ffmpeg:
40493 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
40495 • Add flickering blur effect:
40497 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
40499 latency, alatency
40501 Measure filtering latency.
40502 Report previous filter filtering latency, delay in number of audio
40504 samples for audio filters or number of video frames for video filters.
40505 On end of input stream, filter will report min and max measured latency
40507 for previous running filter in filtergraph.
40508 metadata, ametadata
40510 Manipulate frame metadata.
40511 This filter accepts the following options:
40513 mode
40515 Set mode of operation of the filter.
40516 Can be one of the following:
40518 select
40520 If both "value" and "key" is set, select frames which have such
40521 metadata. If only "key" is set, select every frame that has
40522 such key in metadata.
40523 add Add new metadata "key" and "value". If key is already available
40525 do nothing.
40526 modify
40528 Modify value of already present key.
40529 delete
40531 If "value" is set, delete only keys that have such value.
40532 Otherwise, delete key. If "key" is not set, delete all metadata
40533 values in the frame.
40534 print
40536 Print key and its value if metadata was found. If "key" is not
40537 set print all metadata values available in frame.
40538 key Set key used with all modes. Must be set for all modes except
40540 "print" and "delete".
40541 value
40543 Set metadata value which will be used. This option is mandatory for
40544 "modify" and "add" mode.
40545 function
40547 Which function to use when comparing metadata value and "value".
40548 Can be one of following:
40550 same_str
40552 Values are interpreted as strings, returns true if metadata
40553 value is same as "value".
40554 starts_with
40556 Values are interpreted as strings, returns true if metadata
40557 value starts with the "value" option string.
40558 less
40560 Values are interpreted as floats, returns true if metadata
40561 value is less than "value".
40562 equal
40564 Values are interpreted as floats, returns true if "value" is
40565 equal with metadata value.
40566 greater
40568 Values are interpreted as floats, returns true if metadata
40569 value is greater than "value".
40570 expr
40572 Values are interpreted as floats, returns true if expression
40573 from option "expr" evaluates to true.
40574 ends_with
40576 Values are interpreted as strings, returns true if metadata
40577 value ends with the "value" option string.
40578 expr
40580 Set expression which is used when "function" is set to "expr". The
40581 expression is evaluated through the eval API and can contain the
40582 following constants:
40583 VALUE1, FRAMEVAL
40585 Float representation of "value" from metadata key.
40586 VALUE2, USERVAL
40588 Float representation of "value" as supplied by user in "value"
40589 option.
40590 file
40592 If specified in "print" mode, output is written to the named file.
40593 Instead of plain filename any writable url can be specified.
40594 Filename ``-'' is a shorthand for standard output. If "file" option
40595 is not set, output is written to the log with AV_LOG_INFO loglevel.
40596 direct
40598 Reduces buffering in print mode when output is written to a URL set
40599 using file.
40600 Examples
40602 • Print all metadata values for frames with key
40604 "lavfi.signalstats.YDIF" with values between 0 and 1.
40605 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
40607 • Print silencedetect output to file metadata.txt.
40609 silencedetect,ametadata=mode=print:file=metadata.txt
40611 • Direct all metadata to a pipe with file descriptor 4.
40613 metadata=mode=print:file='pipe\:4'
40615 perms, aperms
40617 Set read/write permissions for the output frames.
40618 These filters are mainly aimed at developers to test direct path in the
40620 following filter in the filtergraph.
40621 The filters accept the following options:
40623 mode
40625 Select the permissions mode.
40626 It accepts the following values:
40628 none
40630 Do nothing. This is the default.
40631 ro Set all the output frames read-only.
40633 rw Set all the output frames directly writable.
40635 toggle
40637 Make the frame read-only if writable, and writable if read-
40638 only.
40639 random
40641 Set each output frame read-only or writable randomly.
40642 seed
40644 Set the seed for the random mode, must be an integer included
40645 between 0 and "UINT32_MAX". If not specified, or if explicitly set
40646 to "-1", the filter will try to use a good random seed on a best
40647 effort basis.
40648 Note: in case of auto-inserted filter between the permission filter and
40650 the following one, the permission might not be received as expected in
40651 that following filter. Inserting a format or aformat filter before the
40652 perms/aperms filter can avoid this problem.
40653 realtime, arealtime
40655 Slow down filtering to match real time approximately.
40656 These filters will pause the filtering for a variable amount of time to
40658 match the output rate with the input timestamps. They are similar to
40659 the re option to "ffmpeg".
40660 They accept the following options:
40662 limit
40664 Time limit for the pauses. Any pause longer than that will be
40665 considered a timestamp discontinuity and reset the timer. Default
40666 is 2 seconds.
40667 speed
40669 Speed factor for processing. The value must be a float larger than
40670 zero. Values larger than 1.0 will result in faster than realtime
40671 processing, smaller will slow processing down. The limit is
40672 automatically adapted accordingly. Default is 1.0.
40673 A processing speed faster than what is possible without these
40675 filters cannot be achieved.
40676 segment, asegment
40678 Split single input stream into multiple streams.
40679 This filter does opposite of concat filters.
40681 "segment" works on video frames, "asegment" on audio samples.
40683 This filter accepts the following options:
40685 timestamps
40687 Timestamps of output segments separated by '|'. The first segment
40688 will run from the beginning of the input stream. The last segment
40689 will run until the end of the input stream
40690 frames, samples
40692 Exact frame/sample count to split the segments.
40693 In all cases, prefixing an each segment with '+' will make it relative
40695 to the previous segment.
40696 Examples
40698 • Split input audio stream into three output audio streams, starting
40700 at start of input audio stream and storing that in 1st output audio
40701 stream, then following at 60th second and storing than in 2nd
40702 output audio stream, and last after 150th second of input audio
40703 stream store in 3rd output audio stream:
40704 asegment=timestamps="60|150"
40706 select, aselect
40708 Select frames to pass in output.
40709 This filter accepts the following options:
40711 expr, e
40713 Set expression, which is evaluated for each input frame.
40714 If the expression is evaluated to zero, the frame is discarded.
40716 If the evaluation result is negative or NaN, the frame is sent to
40718 the first output; otherwise it is sent to the output with index
40719 "ceil(val)-1", assuming that the input index starts from 0.
40720 For example a value of 1.2 corresponds to the output with index
40722 "ceil(1.2)-1 = 2-1 = 1", that is the second output.
40723 outputs, n
40725 Set the number of outputs. The output to which to send the selected
40726 frame is based on the result of the evaluation. Default value is 1.
40727 The expression can contain the following constants:
40729 n The (sequential) number of the filtered frame, starting from 0.
40731 selected_n
40733 The (sequential) number of the selected frame, starting from 0.
40734 prev_selected_n
40736 The sequential number of the last selected frame. It's NAN if
40737 undefined.
40738 TB The timebase of the input timestamps.
40740 pts The PTS (Presentation TimeStamp) of the filtered frame, expressed
40742 in TB units. It's NAN if undefined.
40743 t The PTS of the filtered frame, expressed in seconds. It's NAN if
40745 undefined.
40746 prev_pts
40748 The PTS of the previously filtered frame. It's NAN if undefined.
40749 prev_selected_pts
40751 The PTS of the last previously filtered frame. It's NAN if
40752 undefined.
40753 prev_selected_t
40755 The PTS of the last previously selected frame, expressed in
40756 seconds. It's NAN if undefined.
40757 start_pts
40759 The first PTS in the stream which is not NAN. It remains NAN if not
40760 found.
40761 start_t
40763 The first PTS, in seconds, in the stream which is not NAN. It
40764 remains NAN if not found.
40765 pict_type (video only)
40767 The type of the filtered frame. It can assume one of the following
40768 values:
40769 I
40771 P
40772 B
40773 S
40774 SI
40775 SP
40776 BI
40777 interlace_type (video only)
40778 The frame interlace type. It can assume one of the following
40779 values:
40780 PROGRESSIVE
40782 The frame is progressive (not interlaced).
40783 TOPFIRST
40785 The frame is top-field-first.
40786 BOTTOMFIRST
40788 The frame is bottom-field-first.
40789 consumed_sample_n (audio only)
40791 the number of selected samples before the current frame
40792 samples_n (audio only)
40794 the number of samples in the current frame
40795 sample_rate (audio only)
40797 the input sample rate
40798 key This is 1 if the filtered frame is a key-frame, 0 otherwise.
40800 pos the position in the file of the filtered frame, -1 if the
40802 information is not available (e.g. for synthetic video)
40803 scene (video only)
40805 value between 0 and 1 to indicate a new scene; a low value reflects
40806 a low probability for the current frame to introduce a new scene,
40807 while a higher value means the current frame is more likely to be
40808 one (see the example below)
40809 concatdec_select
40811 The concat demuxer can select only part of a concat input file by
40812 setting an inpoint and an outpoint, but the output packets may not
40813 be entirely contained in the selected interval. By using this
40814 variable, it is possible to skip frames generated by the concat
40815 demuxer which are not exactly contained in the selected interval.
40816 This works by comparing the frame pts against the
40818 lavf.concat.start_time and the lavf.concat.duration packet metadata
40819 values which are also present in the decoded frames.
40820 The concatdec_select variable is -1 if the frame pts is at least
40822 start_time and either the duration metadata is missing or the frame
40823 pts is less than start_time + duration, 0 otherwise, and NaN if the
40824 start_time metadata is missing.
40825 That basically means that an input frame is selected if its pts is
40827 within the interval set by the concat demuxer.
40828 The default value of the select expression is "1".
40830 Examples
40832 • Select all frames in input:
40834 select
40836 The example above is the same as:
40838 select=1
40840 • Skip all frames:
40842 select=0
40844 • Select only I-frames:
40846 select='eq(pict_type\,I)'
40848 • Select one frame every 100:
40850 select='not(mod(n\,100))'
40852 • Select only frames contained in the 10-20 time interval:
40854 select=between(t\,10\,20)
40856 • Select only I-frames contained in the 10-20 time interval:
40858 select=between(t\,10\,20)*eq(pict_type\,I)
40860 • Select frames with a minimum distance of 10 seconds:
40862 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
40864 • Use aselect to select only audio frames with samples number > 100:
40866 aselect='gt(samples_n\,100)'
40868 • Create a mosaic of the first scenes:
40870 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
40872 Comparing scene against a value between 0.3 and 0.5 is generally a
40874 sane choice.
40875 • Send even and odd frames to separate outputs, and compose them:
40877 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
40879 • Select useful frames from an ffconcat file which is using inpoints
40881 and outpoints but where the source files are not intra frame only.
40882 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
40884 sendcmd, asendcmd
40886 Send commands to filters in the filtergraph.
40887 These filters read commands to be sent to other filters in the
40889 filtergraph.
40890 "sendcmd" must be inserted between two video filters, "asendcmd" must
40892 be inserted between two audio filters, but apart from that they act the
40893 same way.
40894 The specification of commands can be provided in the filter arguments
40896 with the commands option, or in a file specified by the filename
40897 option.
40898 These filters accept the following options:
40900 commands, c
40902 Set the commands to be read and sent to the other filters.
40903 filename, f
40905 Set the filename of the commands to be read and sent to the other
40906 filters.
40907 Commands syntax
40909 A commands description consists of a sequence of interval
40911 specifications, comprising a list of commands to be executed when a
40912 particular event related to that interval occurs. The occurring event
40913 is typically the current frame time entering or leaving a given time
40914 interval.
40915 An interval is specified by the following syntax:
40917 <START>[-<END>] <COMMANDS>;
40919 The time interval is specified by the START and END times. END is
40921 optional and defaults to the maximum time.
40922 The current frame time is considered within the specified interval if
40924 it is included in the interval [START, END), that is when the time is
40925 greater or equal to START and is lesser than END.
40926 COMMANDS consists of a sequence of one or more command specifications,
40928 separated by ",", relating to that interval. The syntax of a command
40929 specification is given by:
40930 [<FLAGS>] <TARGET> <COMMAND> <ARG>
40932 FLAGS is optional and specifies the type of events relating to the time
40934 interval which enable sending the specified command, and must be a non-
40935 null sequence of identifier flags separated by "+" or "|" and enclosed
40936 between "[" and "]".
40937 The following flags are recognized:
40939 enter
40941 The command is sent when the current frame timestamp enters the
40942 specified interval. In other words, the command is sent when the
40943 previous frame timestamp was not in the given interval, and the
40944 current is.
40945 leave
40947 The command is sent when the current frame timestamp leaves the
40948 specified interval. In other words, the command is sent when the
40949 previous frame timestamp was in the given interval, and the current
40950 is not.
40951 expr
40953 The command ARG is interpreted as expression and result of
40954 expression is passed as ARG.
40955 The expression is evaluated through the eval API and can contain
40957 the following constants:
40958 POS Original position in the file of the frame, or undefined if
40960 undefined for the current frame.
40961 PTS The presentation timestamp in input.
40963 N The count of the input frame for video or audio, starting from
40965 0.
40966 T The time in seconds of the current frame.
40968 TS The start time in seconds of the current command interval.
40970 TE The end time in seconds of the current command interval.
40972 TI The interpolated time of the current command interval, TI = (T
40974 - TS) / (TE - TS).
40975 If FLAGS is not specified, a default value of "[enter]" is assumed.
40977 TARGET specifies the target of the command, usually the name of the
40979 filter class or a specific filter instance name.
40980 COMMAND specifies the name of the command for the target filter.
40982 ARG is optional and specifies the optional list of argument for the
40984 given COMMAND.
40985 Between one interval specification and another, whitespaces, or
40987 sequences of characters starting with "#" until the end of line, are
40988 ignored and can be used to annotate comments.
40989 A simplified BNF description of the commands specification syntax
40991 follows:
40992 <COMMAND_FLAG> ::= "enter" | "leave"
40994 <COMMAND_FLAGS> ::= <COMMAND_FLAG> [(+|"|")<COMMAND_FLAG>]
40995 <COMMAND> ::= ["[" <COMMAND_FLAGS> "]"] <TARGET> <COMMAND> [<ARG>]
40996 <COMMANDS> ::= <COMMAND> [,<COMMANDS>]
40997 <INTERVAL> ::= <START>[-<END>] <COMMANDS>
40998 <INTERVALS> ::= <INTERVAL>[;<INTERVALS>]
40999 Examples
41001 • Specify audio tempo change at second 4:
41003 asendcmd=c='4.0 atempo tempo 1.5',atempo
41005 • Target a specific filter instance:
41007 asendcmd=c='4.0 atempo@my tempo 1.5',atempo@my
41009 • Specify a list of drawtext and hue commands in a file.
41011 # show text in the interval 5-10
41013 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
41014 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
41015 # desaturate the image in the interval 15-20
41017 15.0-20.0 [enter] hue s 0,
41018 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
41019 [leave] hue s 1,
41020 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
41021 # apply an exponential saturation fade-out effect, starting from time 25
41023 25 [enter] hue s exp(25-t)
41024 A filtergraph allowing to read and process the above command list
41026 stored in a file test.cmd, can be specified with:
41027 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
41029 setpts, asetpts
41031 Change the PTS (presentation timestamp) of the input frames.
41032 "setpts" works on video frames, "asetpts" on audio frames.
41034 This filter accepts the following options:
41036 expr
41038 The expression which is evaluated for each frame to construct its
41039 timestamp.
41040 The expression is evaluated through the eval API and can contain the
41042 following constants:
41043 FRAME_RATE, FR
41045 frame rate, only defined for constant frame-rate video
41046 PTS The presentation timestamp in input
41048 N The count of the input frame for video or the number of consumed
41050 samples, not including the current frame for audio, starting from
41051 0.
41052 NB_CONSUMED_SAMPLES
41054 The number of consumed samples, not including the current frame
41055 (only audio)
41056 NB_SAMPLES, S
41058 The number of samples in the current frame (only audio)
41059 SAMPLE_RATE, SR
41061 The audio sample rate.
41062 STARTPTS
41064 The PTS of the first frame.
41065 STARTT
41067 the time in seconds of the first frame
41068 INTERLACED
41070 State whether the current frame is interlaced.
41071 T the time in seconds of the current frame
41073 POS original position in the file of the frame, or undefined if
41075 undefined for the current frame
41076 PREV_INPTS
41078 The previous input PTS.
41079 PREV_INT
41081 previous input time in seconds
41082 PREV_OUTPTS
41084 The previous output PTS.
41085 PREV_OUTT
41087 previous output time in seconds
41088 RTCTIME
41090 The wallclock (RTC) time in microseconds. This is deprecated, use
41091 time(0) instead.
41092 RTCSTART
41094 The wallclock (RTC) time at the start of the movie in microseconds.
41095 TB The timebase of the input timestamps.
41097 Examples
41099 • Start counting PTS from zero
41101 setpts=PTS-STARTPTS
41103 • Apply fast motion effect:
41105 setpts=0.5*PTS
41107 • Apply slow motion effect:
41109 setpts=2.0*PTS
41111 • Set fixed rate of 25 frames per second:
41113 setpts=N/(25*TB)
41115 • Set fixed rate 25 fps with some jitter:
41117 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
41119 • Apply an offset of 10 seconds to the input PTS:
41121 setpts=PTS+10/TB
41123 • Generate timestamps from a "live source" and rebase onto the
41125 current timebase:
41126 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
41128 • Generate timestamps by counting samples:
41130 asetpts=N/SR/TB
41132 setrange
41134 Force color range for the output video frame.
41135 The "setrange" filter marks the color range property for the output
41137 frames. It does not change the input frame, but only sets the
41138 corresponding property, which affects how the frame is treated by
41139 following filters.
41140 The filter accepts the following options:
41142 range
41144 Available values are:
41145 auto
41147 Keep the same color range property.
41148 unspecified, unknown
41150 Set the color range as unspecified.
41151 limited, tv, mpeg
41153 Set the color range as limited.
41154 full, pc, jpeg
41156 Set the color range as full.
41157 settb, asettb
41159 Set the timebase to use for the output frames timestamps. It is mainly
41160 useful for testing timebase configuration.
41161 It accepts the following parameters:
41163 expr, tb
41165 The expression which is evaluated into the output timebase.
41166 The value for tb is an arithmetic expression representing a rational.
41168 The expression can contain the constants "AVTB" (the default timebase),
41169 "intb" (the input timebase) and "sr" (the sample rate, audio only).
41170 Default value is "intb".
41171 Examples
41173 • Set the timebase to 1/25:
41175 settb=expr=1/25
41177 • Set the timebase to 1/10:
41179 settb=expr=0.1
41181 • Set the timebase to 1001/1000:
41183 settb=1+0.001
41185 • Set the timebase to 2*intb:
41187 settb=2*intb
41189 • Set the default timebase value:
41191 settb=AVTB
41193 showcqt
41195 Convert input audio to a video output representing frequency spectrum
41196 logarithmically using Brown-Puckette constant Q transform algorithm
41197 with direct frequency domain coefficient calculation (but the transform
41198 itself is not really constant Q, instead the Q factor is actually
41199 variable/clamped), with musical tone scale, from E0 to D#10.
41200 The filter accepts the following options:
41202 size, s
41204 Specify the video size for the output. It must be even. For the
41205 syntax of this option, check the "Video size" section in the
41206 ffmpeg-utils manual. Default value is "1920x1080".
41207 fps, rate, r
41209 Set the output frame rate. Default value is 25.
41210 bar_h
41212 Set the bargraph height. It must be even. Default value is "-1"
41213 which computes the bargraph height automatically.
41214 axis_h
41216 Set the axis height. It must be even. Default value is "-1" which
41217 computes the axis height automatically.
41218 sono_h
41220 Set the sonogram height. It must be even. Default value is "-1"
41221 which computes the sonogram height automatically.
41222 fullhd
41224 Set the fullhd resolution. This option is deprecated, use size, s
41225 instead. Default value is 1.
41226 sono_v, volume
41228 Specify the sonogram volume expression. It can contain variables:
41229 bar_v
41231 the bar_v evaluated expression
41232 frequency, freq, f
41234 the frequency where it is evaluated
41235 timeclamp, tc
41237 the value of timeclamp option
41238 and functions:
41240 a_weighting(f)
41242 A-weighting of equal loudness
41243 b_weighting(f)
41245 B-weighting of equal loudness
41246 c_weighting(f)
41248 C-weighting of equal loudness.
41249 Default value is 16.
41251 bar_v, volume2
41253 Specify the bargraph volume expression. It can contain variables:
41254 sono_v
41256 the sono_v evaluated expression
41257 frequency, freq, f
41259 the frequency where it is evaluated
41260 timeclamp, tc
41262 the value of timeclamp option
41263 and functions:
41265 a_weighting(f)
41267 A-weighting of equal loudness
41268 b_weighting(f)
41270 B-weighting of equal loudness
41271 c_weighting(f)
41273 C-weighting of equal loudness.
41274 Default value is "sono_v".
41276 sono_g, gamma
41278 Specify the sonogram gamma. Lower gamma makes the spectrum more
41279 contrast, higher gamma makes the spectrum having more range.
41280 Default value is 3. Acceptable range is "[1, 7]".
41281 bar_g, gamma2
41283 Specify the bargraph gamma. Default value is 1. Acceptable range is
41284 "[1, 7]".
41285 bar_t
41287 Specify the bargraph transparency level. Lower value makes the
41288 bargraph sharper. Default value is 1. Acceptable range is "[0,
41289 1]".
41290 timeclamp, tc
41292 Specify the transform timeclamp. At low frequency, there is trade-
41293 off between accuracy in time domain and frequency domain. If
41294 timeclamp is lower, event in time domain is represented more
41295 accurately (such as fast bass drum), otherwise event in frequency
41296 domain is represented more accurately (such as bass guitar).
41297 Acceptable range is "[0.002, 1]". Default value is 0.17.
41298 attack
41300 Set attack time in seconds. The default is 0 (disabled). Otherwise,
41301 it limits future samples by applying asymmetric windowing in time
41302 domain, useful when low latency is required. Accepted range is "[0,
41303 1]".
41304 basefreq
41306 Specify the transform base frequency. Default value is
41307 20.01523126408007475, which is frequency 50 cents below E0.
41308 Acceptable range is "[10, 100000]".
41309 endfreq
41311 Specify the transform end frequency. Default value is
41312 20495.59681441799654, which is frequency 50 cents above D#10.
41313 Acceptable range is "[10, 100000]".
41314 coeffclamp
41316 This option is deprecated and ignored.
41317 tlength
41319 Specify the transform length in time domain. Use this option to
41320 control accuracy trade-off between time domain and frequency domain
41321 at every frequency sample. It can contain variables:
41322 frequency, freq, f
41324 the frequency where it is evaluated
41325 timeclamp, tc
41327 the value of timeclamp option.
41328 Default value is "384*tc/(384+tc*f)".
41330 count
41332 Specify the transform count for every video frame. Default value is
41333 6. Acceptable range is "[1, 30]".
41334 fcount
41336 Specify the transform count for every single pixel. Default value
41337 is 0, which makes it computed automatically. Acceptable range is
41338 "[0, 10]".
41339 fontfile
41341 Specify font file for use with freetype to draw the axis. If not
41342 specified, use embedded font. Note that drawing with font file or
41343 embedded font is not implemented with custom basefreq and endfreq,
41344 use axisfile option instead.
41345 font
41347 Specify fontconfig pattern. This has lower priority than fontfile.
41348 The ":" in the pattern may be replaced by "|" to avoid unnecessary
41349 escaping.
41350 fontcolor
41352 Specify font color expression. This is arithmetic expression that
41353 should return integer value 0xRRGGBB. It can contain variables:
41354 frequency, freq, f
41356 the frequency where it is evaluated
41357 timeclamp, tc
41359 the value of timeclamp option
41360 and functions:
41362 midi(f)
41364 midi number of frequency f, some midi numbers: E0(16), C1(24),
41365 C2(36), A4(69)
41366 r(x), g(x), b(x)
41368 red, green, and blue value of intensity x.
41369 Default value is "st(0, (midi(f)-59.5)/12); st(1,
41371 if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0)); r(1-ld(1)) +
41372 b(ld(1))".
41373 axisfile
41375 Specify image file to draw the axis. This option override fontfile
41376 and fontcolor option.
41377 axis, text
41379 Enable/disable drawing text to the axis. If it is set to 0, drawing
41380 to the axis is disabled, ignoring fontfile and axisfile option.
41381 Default value is 1.
41382 csp Set colorspace. The accepted values are:
41384 unspecified
41386 Unspecified (default)
41387 bt709
41389 BT.709
41390 fcc FCC
41392 bt470bg
41394 BT.470BG or BT.601-6 625
41395 smpte170m
41397 SMPTE-170M or BT.601-6 525
41398 smpte240m
41400 SMPTE-240M
41401 bt2020ncl
41403 BT.2020 with non-constant luminance
41404 cscheme
41406 Set spectrogram color scheme. This is list of floating point values
41407 with format "left_r|left_g|left_b|right_r|right_g|right_b". The
41408 default is "1|0.5|0|0|0.5|1".
41409 Examples
41411 • Playing audio while showing the spectrum:
41413 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
41415 • Same as above, but with frame rate 30 fps:
41417 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
41419 • Playing at 1280x720:
41421 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
41423 • Disable sonogram display:
41425 sono_h=0
41427 • A1 and its harmonics: A1, A2, (near)E3, A3:
41429 ffplay -f lavfi 'aevalsrc=0.1*sin(2*PI*55*t)+0.1*sin(4*PI*55*t)+0.1*sin(6*PI*55*t)+0.1*sin(8*PI*55*t),
41431 asplit[a][out1]; [a] showcqt [out0]'
41432 • Same as above, but with more accuracy in frequency domain:
41434 ffplay -f lavfi 'aevalsrc=0.1*sin(2*PI*55*t)+0.1*sin(4*PI*55*t)+0.1*sin(6*PI*55*t)+0.1*sin(8*PI*55*t),
41436 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
41437 • Custom volume:
41439 bar_v=10:sono_v=bar_v*a_weighting(f)
41441 • Custom gamma, now spectrum is linear to the amplitude.
41443 bar_g=2:sono_g=2
41445 • Custom tlength equation:
41447 tc=0.33:tlength='st(0,0.17); 384*tc / (384 / ld(0) + tc*f /(1-ld(0))) + 384*tc / (tc*f / ld(0) + 384 /(1-ld(0)))'
41449 • Custom fontcolor and fontfile, C-note is colored green, others are
41451 colored blue:
41452 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
41454 • Custom font using fontconfig:
41456 font='Courier New,Monospace,mono|bold'
41458 • Custom frequency range with custom axis using image file:
41460 axisfile=myaxis.png:basefreq=40:endfreq=10000
41462 showfreqs
41464 Convert input audio to video output representing the audio power
41465 spectrum. Audio amplitude is on Y-axis while frequency is on X-axis.
41466 The filter accepts the following options:
41468 size, s
41470 Specify size of video. For the syntax of this option, check the
41471 "Video size" section in the ffmpeg-utils manual. Default is
41472 "1024x512".
41473 mode
41475 Set display mode. This set how each frequency bin will be
41476 represented.
41477 It accepts the following values:
41479 line
41481 bar
41482 dot
41483 Default is "bar".
41485 ascale
41487 Set amplitude scale.
41488 It accepts the following values:
41490 lin Linear scale.
41492 sqrt
41494 Square root scale.
41495 cbrt
41497 Cubic root scale.
41498 log Logarithmic scale.
41500 Default is "log".
41502 fscale
41504 Set frequency scale.
41505 It accepts the following values:
41507 lin Linear scale.
41509 log Logarithmic scale.
41511 rlog
41513 Reverse logarithmic scale.
41514 Default is "lin".
41516 win_size
41518 Set window size. Allowed range is from 16 to 65536.
41519 Default is 2048
41521 win_func
41523 Set windowing function.
41524 It accepts the following values:
41526 rect
41528 bartlett
41529 hanning
41530 hamming
41531 blackman
41532 welch
41533 flattop
41534 bharris
41535 bnuttall
41536 bhann
41537 sine
41538 nuttall
41539 lanczos
41540 gauss
41541 tukey
41542 dolph
41543 cauchy
41544 parzen
41545 poisson
41546 bohman
41547 Default is "hanning".
41549 overlap
41551 Set window overlap. In range "[0, 1]". Default is 1, which means
41552 optimal overlap for selected window function will be picked.
41553 averaging
41555 Set time averaging. Setting this to 0 will display current maximal
41556 peaks. Default is 1, which means time averaging is disabled.
41557 colors
41559 Specify list of colors separated by space or by '|' which will be
41560 used to draw channel frequencies. Unrecognized or missing colors
41561 will be replaced by white color.
41562 cmode
41564 Set channel display mode.
41565 It accepts the following values:
41567 combined
41569 separate
41570 Default is "combined".
41572 minamp
41574 Set minimum amplitude used in "log" amplitude scaler.
41575 data
41577 Set data display mode.
41578 It accepts the following values:
41580 magnitude
41582 phase
41583 delay
41584 Default is "magnitude".
41586 showspatial
41588 Convert stereo input audio to a video output, representing the spatial
41589 relationship between two channels.
41590 The filter accepts the following options:
41592 size, s
41594 Specify the video size for the output. For the syntax of this
41595 option, check the "Video size" section in the ffmpeg-utils manual.
41596 Default value is "512x512".
41597 win_size
41599 Set window size. Allowed range is from 1024 to 65536. Default size
41600 is 4096.
41601 win_func
41603 Set window function.
41604 It accepts the following values:
41606 rect
41608 bartlett
41609 hann
41610 hanning
41611 hamming
41612 blackman
41613 welch
41614 flattop
41615 bharris
41616 bnuttall
41617 bhann
41618 sine
41619 nuttall
41620 lanczos
41621 gauss
41622 tukey
41623 dolph
41624 cauchy
41625 parzen
41626 poisson
41627 bohman
41628 Default value is "hann".
41630 overlap
41632 Set ratio of overlap window. Default value is 0.5. When value is 1
41633 overlap is set to recommended size for specific window function
41634 currently used.
41635 showspectrum
41637 Convert input audio to a video output, representing the audio frequency
41638 spectrum.
41639 The filter accepts the following options:
41641 size, s
41643 Specify the video size for the output. For the syntax of this
41644 option, check the "Video size" section in the ffmpeg-utils manual.
41645 Default value is "640x512".
41646 slide
41648 Specify how the spectrum should slide along the window.
41649 It accepts the following values:
41651 replace
41653 the samples start again on the left when they reach the right
41654 scroll
41656 the samples scroll from right to left
41657 fullframe
41659 frames are only produced when the samples reach the right
41660 rscroll
41662 the samples scroll from left to right
41663 lreplace
41665 the samples start again on the right when they reach the left
41666 Default value is "replace".
41668 mode
41670 Specify display mode.
41671 It accepts the following values:
41673 combined
41675 all channels are displayed in the same row
41676 separate
41678 all channels are displayed in separate rows
41679 Default value is combined.
41681 color
41683 Specify display color mode.
41684 It accepts the following values:
41686 channel
41688 each channel is displayed in a separate color
41689 intensity
41691 each channel is displayed using the same color scheme
41692 rainbow
41694 each channel is displayed using the rainbow color scheme
41695 moreland
41697 each channel is displayed using the moreland color scheme
41698 nebulae
41700 each channel is displayed using the nebulae color scheme
41701 fire
41703 each channel is displayed using the fire color scheme
41704 fiery
41706 each channel is displayed using the fiery color scheme
41707 fruit
41709 each channel is displayed using the fruit color scheme
41710 cool
41712 each channel is displayed using the cool color scheme
41713 magma
41715 each channel is displayed using the magma color scheme
41716 green
41718 each channel is displayed using the green color scheme
41719 viridis
41721 each channel is displayed using the viridis color scheme
41722 plasma
41724 each channel is displayed using the plasma color scheme
41725 cividis
41727 each channel is displayed using the cividis color scheme
41728 terrain
41730 each channel is displayed using the terrain color scheme
41731 Default value is channel.
41733 scale
41735 Specify scale used for calculating intensity color values.
41736 It accepts the following values:
41738 lin linear
41740 sqrt
41742 square root, default
41743 cbrt
41745 cubic root
41746 log logarithmic
41748 4thrt
41750 4th root
41751 5thrt
41753 5th root
41754 Default value is sqrt.
41756 fscale
41758 Specify frequency scale.
41759 It accepts the following values:
41761 lin linear
41763 log logarithmic
41765 Default value is lin.
41767 saturation
41769 Set saturation modifier for displayed colors. Negative values
41770 provide alternative color scheme. 0 is no saturation at all.
41771 Saturation must be in [-10.0, 10.0] range. Default value is 1.
41772 win_func
41774 Set window function.
41775 It accepts the following values:
41777 rect
41779 bartlett
41780 hann
41781 hanning
41782 hamming
41783 blackman
41784 welch
41785 flattop
41786 bharris
41787 bnuttall
41788 bhann
41789 sine
41790 nuttall
41791 lanczos
41792 gauss
41793 tukey
41794 dolph
41795 cauchy
41796 parzen
41797 poisson
41798 bohman
41799 Default value is "hann".
41801 orientation
41803 Set orientation of time vs frequency axis. Can be "vertical" or
41804 "horizontal". Default is "vertical".
41805 overlap
41807 Set ratio of overlap window. Default value is 0. When value is 1
41808 overlap is set to recommended size for specific window function
41809 currently used.
41810 gain
41812 Set scale gain for calculating intensity color values. Default
41813 value is 1.
41814 data
41816 Set which data to display. Can be "magnitude", default or "phase",
41817 or unwrapped phase: "uphase".
41818 rotation
41820 Set color rotation, must be in [-1.0, 1.0] range. Default value is
41821 0.
41822 start
41824 Set start frequency from which to display spectrogram. Default is
41825 0.
41826 stop
41828 Set stop frequency to which to display spectrogram. Default is 0.
41829 fps Set upper frame rate limit. Default is "auto", unlimited.
41831 legend
41833 Draw time and frequency axes and legends. Default is disabled.
41834 drange
41836 Set dynamic range used to calculate intensity color values. Default
41837 is 120 dBFS. Allowed range is from 10 to 200.
41838 limit
41840 Set upper limit of input audio samples volume in dBFS. Default is 0
41841 dBFS. Allowed range is from -100 to 100.
41842 The usage is very similar to the showwaves filter; see the examples in
41844 that section.
41845 Examples
41847 • Large window with logarithmic color scaling:
41849 showspectrum=s=1280x480:scale=log
41851 • Complete example for a colored and sliding spectrum per channel
41853 using ffplay:
41854 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
41856 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
41857 showspectrumpic
41859 Convert input audio to a single video frame, representing the audio
41860 frequency spectrum.
41861 The filter accepts the following options:
41863 size, s
41865 Specify the video size for the output. For the syntax of this
41866 option, check the "Video size" section in the ffmpeg-utils manual.
41867 Default value is "4096x2048".
41868 mode
41870 Specify display mode.
41871 It accepts the following values:
41873 combined
41875 all channels are displayed in the same row
41876 separate
41878 all channels are displayed in separate rows
41879 Default value is combined.
41881 color
41883 Specify display color mode.
41884 It accepts the following values:
41886 channel
41888 each channel is displayed in a separate color
41889 intensity
41891 each channel is displayed using the same color scheme
41892 rainbow
41894 each channel is displayed using the rainbow color scheme
41895 moreland
41897 each channel is displayed using the moreland color scheme
41898 nebulae
41900 each channel is displayed using the nebulae color scheme
41901 fire
41903 each channel is displayed using the fire color scheme
41904 fiery
41906 each channel is displayed using the fiery color scheme
41907 fruit
41909 each channel is displayed using the fruit color scheme
41910 cool
41912 each channel is displayed using the cool color scheme
41913 magma
41915 each channel is displayed using the magma color scheme
41916 green
41918 each channel is displayed using the green color scheme
41919 viridis
41921 each channel is displayed using the viridis color scheme
41922 plasma
41924 each channel is displayed using the plasma color scheme
41925 cividis
41927 each channel is displayed using the cividis color scheme
41928 terrain
41930 each channel is displayed using the terrain color scheme
41931 Default value is intensity.
41933 scale
41935 Specify scale used for calculating intensity color values.
41936 It accepts the following values:
41938 lin linear
41940 sqrt
41942 square root, default
41943 cbrt
41945 cubic root
41946 log logarithmic
41948 4thrt
41950 4th root
41951 5thrt
41953 5th root
41954 Default value is log.
41956 fscale
41958 Specify frequency scale.
41959 It accepts the following values:
41961 lin linear
41963 log logarithmic
41965 Default value is lin.
41967 saturation
41969 Set saturation modifier for displayed colors. Negative values
41970 provide alternative color scheme. 0 is no saturation at all.
41971 Saturation must be in [-10.0, 10.0] range. Default value is 1.
41972 win_func
41974 Set window function.
41975 It accepts the following values:
41977 rect
41979 bartlett
41980 hann
41981 hanning
41982 hamming
41983 blackman
41984 welch
41985 flattop
41986 bharris
41987 bnuttall
41988 bhann
41989 sine
41990 nuttall
41991 lanczos
41992 gauss
41993 tukey
41994 dolph
41995 cauchy
41996 parzen
41997 poisson
41998 bohman
41999 Default value is "hann".
42001 orientation
42003 Set orientation of time vs frequency axis. Can be "vertical" or
42004 "horizontal". Default is "vertical".
42005 gain
42007 Set scale gain for calculating intensity color values. Default
42008 value is 1.
42009 legend
42011 Draw time and frequency axes and legends. Default is enabled.
42012 rotation
42014 Set color rotation, must be in [-1.0, 1.0] range. Default value is
42015 0.
42016 start
42018 Set start frequency from which to display spectrogram. Default is
42019 0.
42020 stop
42022 Set stop frequency to which to display spectrogram. Default is 0.
42023 drange
42025 Set dynamic range used to calculate intensity color values. Default
42026 is 120 dBFS. Allowed range is from 10 to 200.
42027 limit
42029 Set upper limit of input audio samples volume in dBFS. Default is 0
42030 dBFS. Allowed range is from -100 to 100.
42031 Examples
42033 • Extract an audio spectrogram of a whole audio track in a 1024x1024
42035 picture using ffmpeg:
42036 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
42038 showvolume
42040 Convert input audio volume to a video output.
42041 The filter accepts the following options:
42043 rate, r
42045 Set video rate.
42046 b Set border width, allowed range is [0, 5]. Default is 1.
42048 w Set channel width, allowed range is [80, 8192]. Default is 400.
42050 h Set channel height, allowed range is [1, 900]. Default is 20.
42052 f Set fade, allowed range is [0, 1]. Default is 0.95.
42054 c Set volume color expression.
42056 The expression can use the following variables:
42058 VOLUME
42060 Current max volume of channel in dB.
42061 PEAK
42063 Current peak.
42064 CHANNEL
42066 Current channel number, starting from 0.
42067 t If set, displays channel names. Default is enabled.
42069 v If set, displays volume values. Default is enabled.
42071 o Set orientation, can be horizontal: "h" or vertical: "v", default
42073 is "h".
42074 s Set step size, allowed range is [0, 5]. Default is 0, which means
42076 step is disabled.
42077 p Set background opacity, allowed range is [0, 1]. Default is 0.
42079 m Set metering mode, can be peak: "p" or rms: "r", default is "p".
42081 ds Set display scale, can be linear: "lin" or log: "log", default is
42083 "lin".
42084 dm In second. If set to > 0., display a line for the max level in the
42086 previous seconds. default is disabled: 0.
42087 dmc The color of the max line. Use when "dm" option is set to > 0.
42089 default is: "orange"
42090 showwaves
42092 Convert input audio to a video output, representing the samples waves.
42093 The filter accepts the following options:
42095 size, s
42097 Specify the video size for the output. For the syntax of this
42098 option, check the "Video size" section in the ffmpeg-utils manual.
42099 Default value is "600x240".
42100 mode
42102 Set display mode.
42103 Available values are:
42105 point
42107 Draw a point for each sample.
42108 line
42110 Draw a vertical line for each sample.
42111 p2p Draw a point for each sample and a line between them.
42113 cline
42115 Draw a centered vertical line for each sample.
42116 Default value is "point".
42118 n Set the number of samples which are printed on the same column. A
42120 larger value will decrease the frame rate. Must be a positive
42121 integer. This option can be set only if the value for rate is not
42122 explicitly specified.
42123 rate, r
42125 Set the (approximate) output frame rate. This is done by setting
42126 the option n. Default value is "25".
42127 split_channels
42129 Set if channels should be drawn separately or overlap. Default
42130 value is 0.
42131 colors
42133 Set colors separated by '|' which are going to be used for drawing
42134 of each channel.
42135 scale
42137 Set amplitude scale.
42138 Available values are:
42140 lin Linear.
42142 log Logarithmic.
42144 sqrt
42146 Square root.
42147 cbrt
42149 Cubic root.
42150 Default is linear.
42152 draw
42154 Set the draw mode. This is mostly useful to set for high n.
42155 Available values are:
42157 scale
42159 Scale pixel values for each drawn sample.
42160 full
42162 Draw every sample directly.
42163 Default value is "scale".
42165 Examples
42167 • Output the input file audio and the corresponding video
42169 representation at the same time:
42170 amovie=a.mp3,asplit[out0],showwaves[out1]
42172 • Create a synthetic signal and show it with showwaves, forcing a
42174 frame rate of 30 frames per second:
42175 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
42177 showwavespic
42179 Convert input audio to a single video frame, representing the samples
42180 waves.
42181 The filter accepts the following options:
42183 size, s
42185 Specify the video size for the output. For the syntax of this
42186 option, check the "Video size" section in the ffmpeg-utils manual.
42187 Default value is "600x240".
42188 split_channels
42190 Set if channels should be drawn separately or overlap. Default
42191 value is 0.
42192 colors
42194 Set colors separated by '|' which are going to be used for drawing
42195 of each channel.
42196 scale
42198 Set amplitude scale.
42199 Available values are:
42201 lin Linear.
42203 log Logarithmic.
42205 sqrt
42207 Square root.
42208 cbrt
42210 Cubic root.
42211 Default is linear.
42213 draw
42215 Set the draw mode.
42216 Available values are:
42218 scale
42220 Scale pixel values for each drawn sample.
42221 full
42223 Draw every sample directly.
42224 Default value is "scale".
42226 filter
42228 Set the filter mode.
42229 Available values are:
42231 average
42233 Use average samples values for each drawn sample.
42234 peak
42236 Use peak samples values for each drawn sample.
42237 Default value is "average".
42239 Examples
42241 • Extract a channel split representation of the wave form of a whole
42243 audio track in a 1024x800 picture using ffmpeg:
42244 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
42246 sidedata, asidedata
42248 Delete frame side data, or select frames based on it.
42249 This filter accepts the following options:
42251 mode
42253 Set mode of operation of the filter.
42254 Can be one of the following:
42256 select
42258 Select every frame with side data of "type".
42259 delete
42261 Delete side data of "type". If "type" is not set, delete all
42262 side data in the frame.
42263 type
42265 Set side data type used with all modes. Must be set for "select"
42266 mode. For the list of frame side data types, refer to the
42267 "AVFrameSideDataType" enum in libavutil/frame.h. For example, to
42268 choose "AV_FRAME_DATA_PANSCAN" side data, you must specify
42269 "PANSCAN".
42270 spectrumsynth
42272 Synthesize audio from 2 input video spectrums, first input stream
42273 represents magnitude across time and second represents phase across
42274 time. The filter will transform from frequency domain as displayed in
42275 videos back to time domain as presented in audio output.
42276 This filter is primarily created for reversing processed showspectrum
42278 filter outputs, but can synthesize sound from other spectrograms too.
42279 But in such case results are going to be poor if the phase data is not
42280 available, because in such cases phase data need to be recreated,
42281 usually it's just recreated from random noise. For best results use
42282 gray only output ("channel" color mode in showspectrum filter) and
42283 "log" scale for magnitude video and "lin" scale for phase video. To
42284 produce phase, for 2nd video, use "data" option. Inputs videos should
42285 generally use "fullframe" slide mode as that saves resources needed for
42286 decoding video.
42287 The filter accepts the following options:
42289 sample_rate
42291 Specify sample rate of output audio, the sample rate of audio from
42292 which spectrum was generated may differ.
42293 channels
42295 Set number of channels represented in input video spectrums.
42296 scale
42298 Set scale which was used when generating magnitude input spectrum.
42299 Can be "lin" or "log". Default is "log".
42300 slide
42302 Set slide which was used when generating inputs spectrums. Can be
42303 "replace", "scroll", "fullframe" or "rscroll". Default is
42304 "fullframe".
42305 win_func
42307 Set window function used for resynthesis.
42308 overlap
42310 Set window overlap. In range "[0, 1]". Default is 1, which means
42311 optimal overlap for selected window function will be picked.
42312 orientation
42314 Set orientation of input videos. Can be "vertical" or "horizontal".
42315 Default is "vertical".
42316 Examples
42318 • First create magnitude and phase videos from audio, assuming audio
42320 is stereo with 44100 sample rate, then resynthesize videos back to
42321 audio with spectrumsynth:
42322 ffmpeg -i input.flac -lavfi showspectrum=mode=separate:scale=log:overlap=0.875:color=channel:slide=fullframe:data=magnitude -an -c:v rawvideo magnitude.nut
42324 ffmpeg -i input.flac -lavfi showspectrum=mode=separate:scale=lin:overlap=0.875:color=channel:slide=fullframe:data=phase -an -c:v rawvideo phase.nut
42325 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
42326 split, asplit
42328 Split input into several identical outputs.
42329 "asplit" works with audio input, "split" with video.
42331 The filter accepts a single parameter which specifies the number of
42333 outputs. If unspecified, it defaults to 2.
42334 Examples
42336 • Create two separate outputs from the same input:
42338 [in] split [out0][out1]
42340 • To create 3 or more outputs, you need to specify the number of
42342 outputs, like in:
42343 [in] asplit=3 [out0][out1][out2]
42345 • Create two separate outputs from the same input, one cropped and
42347 one padded:
42348 [in] split [splitout1][splitout2];
42350 [splitout1] crop=100:100:0:0 [cropout];
42351 [splitout2] pad=200:200:100:100 [padout];
42352 • Create 5 copies of the input audio with ffmpeg:
42354 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
42356 zmq, azmq
42358 Receive commands sent through a libzmq client, and forward them to
42359 filters in the filtergraph.
42360 "zmq" and "azmq" work as a pass-through filters. "zmq" must be inserted
42362 between two video filters, "azmq" between two audio filters. Both are
42363 capable to send messages to any filter type.
42364 To enable these filters you need to install the libzmq library and
42366 headers and configure FFmpeg with "--enable-libzmq".
42367 For more information about libzmq see: <http://www.zeromq.org/>
42369 The "zmq" and "azmq" filters work as a libzmq server, which receives
42371 messages sent through a network interface defined by the bind_address
42372 (or the abbreviation "b") option. Default value of this option is
42373 tcp://localhost:5555. You may want to alter this value to your needs,
42374 but do not forget to escape any ':' signs (see filtergraph escaping).
42375 The received message must be in the form:
42377 <TARGET> <COMMAND> [<ARG>]
42379 TARGET specifies the target of the command, usually the name of the
42381 filter class or a specific filter instance name. The default filter
42382 instance name uses the pattern Parsed_<filter_name>_<index>, but you
42383 can override this by using the filter_name@id syntax (see Filtergraph
42384 syntax).
42385 COMMAND specifies the name of the command for the target filter.
42387 ARG is optional and specifies the optional argument list for the given
42389 COMMAND.
42390 Upon reception, the message is processed and the corresponding command
42392 is injected into the filtergraph. Depending on the result, the filter
42393 will send a reply to the client, adopting the format:
42394 <ERROR_CODE> <ERROR_REASON>
42396 <MESSAGE>
42397 MESSAGE is optional.
42399 Examples
42401 Look at tools/zmqsend for an example of a zmq client which can be used
42403 to send commands processed by these filters.
42404 Consider the following filtergraph generated by ffplay. In this
42406 example the last overlay filter has an instance name. All other filters
42407 will have default instance names.
42408 ffplay -dumpgraph 1 -f lavfi "
42410 color=s=100x100:c=red [l];
42411 color=s=100x100:c=blue [r];
42412 nullsrc=s=200x100, zmq [bg];
42413 [bg][l] overlay [bg+l];
42414 [bg+l][r] overlay@my=x=100 "
42415 To change the color of the left side of the video, the following
42417 command can be used:
42418 echo Parsed_color_0 c yellow | tools/zmqsend
42420 To change the right side:
42422 echo Parsed_color_1 c pink | tools/zmqsend
42424 To change the position of the right side:
42426 echo overlay@my x 150 | tools/zmqsend
42428
42430 Below is a description of the currently available multimedia sources.
42431 amovie
42433 This is the same as movie source, except it selects an audio stream by
42434 default.
42435 movie
42437 Read audio and/or video stream(s) from a movie container.
42438 It accepts the following parameters:
42440 filename
42442 The name of the resource to read (not necessarily a file; it can
42443 also be a device or a stream accessed through some protocol).
42444 format_name, f
42446 Specifies the format assumed for the movie to read, and can be
42447 either the name of a container or an input device. If not
42448 specified, the format is guessed from movie_name or by probing.
42449 seek_point, sp
42451 Specifies the seek point in seconds. The frames will be output
42452 starting from this seek point. The parameter is evaluated with
42453 "av_strtod", so the numerical value may be suffixed by an IS
42454 postfix. The default value is "0".
42455 streams, s
42457 Specifies the streams to read. Several streams can be specified,
42458 separated by "+". The source will then have as many outputs, in the
42459 same order. The syntax is explained in the "Stream specifiers"
42460 section in the ffmpeg manual. Two special names, "dv" and "da"
42461 specify respectively the default (best suited) video and audio
42462 stream. Default is "dv", or "da" if the filter is called as
42463 "amovie".
42464 stream_index, si
42466 Specifies the index of the video stream to read. If the value is
42467 -1, the most suitable video stream will be automatically selected.
42468 The default value is "-1". Deprecated. If the filter is called
42469 "amovie", it will select audio instead of video.
42470 loop
42472 Specifies how many times to read the stream in sequence. If the
42473 value is 0, the stream will be looped infinitely. Default value is
42474 "1".
42475 Note that when the movie is looped the source timestamps are not
42477 changed, so it will generate non monotonically increasing
42478 timestamps.
42479 discontinuity
42481 Specifies the time difference between frames above which the point
42482 is considered a timestamp discontinuity which is removed by
42483 adjusting the later timestamps.
42484 dec_threads
42486 Specifies the number of threads for decoding
42487 format_opts
42489 Specify format options for the opened file. Format options can be
42490 specified as a list of key=value pairs separated by ':'. The
42491 following example shows how to add protocol_whitelist and
42492 protocol_blacklist options:
42493 ffplay -f lavfi
42495 "movie=filename='1.sdp':format_opts='protocol_whitelist=file,rtp,udp\:protocol_blacklist=http'"
42496 It allows overlaying a second video on top of the main input of a
42498 filtergraph, as shown in this graph:
42499 input -----------> deltapts0 --> overlay --> output
42501 ^
42502 |
42503 movie --> scale--> deltapts1 -------+
42504 Examples
42506 • Skip 3.2 seconds from the start of the AVI file in.avi, and overlay
42508 it on top of the input labelled "in":
42509 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
42511 [in] setpts=PTS-STARTPTS [main];
42512 [main][over] overlay=16:16 [out]
42513 • Read from a video4linux2 device, and overlay it on top of the input
42515 labelled "in":
42516 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
42518 [in] setpts=PTS-STARTPTS [main];
42519 [main][over] overlay=16:16 [out]
42520 • Read the first video stream and the audio stream with id 0x81 from
42522 dvd.vob; the video is connected to the pad named "video" and the
42523 audio is connected to the pad named "audio":
42524 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
42526 Commands
42528 Both movie and amovie support the following commands:
42530 seek
42532 Perform seek using "av_seek_frame". The syntax is: seek
42533 stream_index|timestamp|flags
42534 • stream_index: If stream_index is -1, a default stream is
42536 selected, and timestamp is automatically converted from
42537 AV_TIME_BASE units to the stream specific time_base.
42538 • timestamp: Timestamp in AVStream.time_base units or, if no
42540 stream is specified, in AV_TIME_BASE units.
42541 • flags: Flags which select direction and seeking mode.
42543 get_duration
42545 Get movie duration in AV_TIME_BASE units.
42546
42548 FFmpeg can be hooked up with a number of external libraries to add
42549 support for more formats. None of them are used by default, their use
42550 has to be explicitly requested by passing the appropriate flags to
42551 ./configure.
42552 Alliance for Open Media (AOM)
42554 FFmpeg can make use of the AOM library for AV1 decoding and encoding.
42555 Go to <http://aomedia.org/> and follow the instructions for installing
42557 the library. Then pass "--enable-libaom" to configure to enable it.
42558 AMD AMF/VCE
42560 FFmpeg can use the AMD Advanced Media Framework library for accelerated
42561 H.264 and HEVC(only windows) encoding on hardware with Video Coding
42562 Engine (VCE).
42563 To enable support you must obtain the AMF framework header
42565 files(version 1.4.9+) from
42566 <https://github.com/GPUOpen-LibrariesAndSDKs/AMF.git>.
42567 Create an "AMF/" directory in the system include path. Copy the
42569 contents of "AMF/amf/public/include/" into that directory. Then
42570 configure FFmpeg with "--enable-amf".
42571 Initialization of amf encoder occurs in this order: 1) trying to
42573 initialize through dx11(only windows) 2) trying to initialize through
42574 dx9(only windows) 3) trying to initialize through vulkan
42575 To use h.264(AMD VCE) encoder on linux amdgru-pro version 19.20+ and
42577 amf-amdgpu-pro package(amdgru-pro contains, but does not install
42578 automatically) are required.
42579 This driver can be installed using amdgpu-pro-install script in
42581 official amd driver archive.
42582 AviSynth
42584 FFmpeg can read AviSynth scripts as input. To enable support, pass
42585 "--enable-avisynth" to configure after installing the headers provided
42586 by <https://github.com/AviSynth/AviSynthPlus>. AviSynth+ can be
42587 configured to install only the headers by either passing
42588 "-DHEADERS_ONLY:bool=on" to the normal CMake-based build system, or by
42589 using the supplied "GNUmakefile".
42590 For Windows, supported AviSynth variants are <http://avisynth.nl> for
42592 32-bit builds and <http://avisynth.nl/index.php/AviSynth+> for 32-bit
42593 and 64-bit builds.
42594 For Linux, macOS, and BSD, the only supported AviSynth variant is
42596 <https://github.com/AviSynth/AviSynthPlus>, starting with version 3.5.
42597 In 2016, AviSynth+ added support for building with GCC. However,
42599 due to the eccentricities of Windows' calling conventions, 32-bit
42600 GCC builds of AviSynth+ are not compatible with typical 32-bit
42601 builds of FFmpeg.
42602 By default, FFmpeg assumes compatibility with 32-bit MSVC builds of
42604 AviSynth+ since that is the most widely-used and entrenched build
42605 configuration. Users can override this and enable support for
42606 32-bit GCC builds of AviSynth+ by passing "-DAVSC_WIN32_GCC32" to
42607 "--extra-cflags" when configuring FFmpeg.
42608 64-bit builds of FFmpeg are not affected, and can use either MSVC
42610 or GCC builds of AviSynth+ without any special flags.
42611 AviSynth(+) is loaded dynamically. Distributors can build FFmpeg
42613 with "--enable-avisynth", and the binaries will work regardless of
42614 the end user having AviSynth installed. If/when an end user would
42615 like to use AviSynth scripts, then they can install AviSynth(+) and
42616 FFmpeg will be able to find and use it to open scripts.
42617 Chromaprint
42619 FFmpeg can make use of the Chromaprint library for generating audio
42620 fingerprints. Pass "--enable-chromaprint" to configure to enable it.
42621 See <https://acoustid.org/chromaprint>.
42622 codec2
42624 FFmpeg can make use of the codec2 library for codec2 decoding and
42625 encoding. There is currently no native decoder, so libcodec2 must be
42626 used for decoding.
42627 Go to <http://freedv.org/>, download "Codec 2 source archive". Build
42629 and install using CMake. Debian users can install the libcodec2-dev
42630 package instead. Once libcodec2 is installed you can pass
42631 "--enable-libcodec2" to configure to enable it.
42632 The easiest way to use codec2 is with .c2 files, since they contain the
42634 mode information required for decoding. To encode such a file, use a
42635 .c2 file extension and give the libcodec2 encoder the -mode option:
42636 "ffmpeg -i input.wav -mode 700C output.c2". Playback is as simple as
42637 "ffplay output.c2". For a list of supported modes, run "ffmpeg -h
42638 encoder=libcodec2". Raw codec2 files are also supported. To make
42639 sense of them the mode in use needs to be specified as a format option:
42640 "ffmpeg -f codec2raw -mode 1300 -i input.raw output.wav".
42641 dav1d
42643 FFmpeg can make use of the dav1d library for AV1 video decoding.
42644 Go to <https://code.videolan.org/videolan/dav1d> and follow the
42646 instructions for installing the library. Then pass "--enable-libdav1d"
42647 to configure to enable it.
42648 davs2
42650 FFmpeg can make use of the davs2 library for AVS2-P2/IEEE1857.4 video
42651 decoding.
42652 Go to <https://github.com/pkuvcl/davs2> and follow the instructions for
42654 installing the library. Then pass "--enable-libdavs2" to configure to
42655 enable it.
42656 libdavs2 is under the GNU Public License Version 2 or later (see
42658 <http://www.gnu.org/licenses/old-licenses/gpl-2.0.html> for
42659 details), you must upgrade FFmpeg's license to GPL in order to use
42660 it.
42661 uavs3d
42663 FFmpeg can make use of the uavs3d library for AVS3-P2/IEEE1857.10 video
42664 decoding.
42665 Go to <https://github.com/uavs3/uavs3d> and follow the instructions for
42667 installing the library. Then pass "--enable-libuavs3d" to configure to
42668 enable it.
42669 Game Music Emu
42671 FFmpeg can make use of the Game Music Emu library to read audio from
42672 supported video game music file formats. Pass "--enable-libgme" to
42673 configure to enable it. See
42674 <https://bitbucket.org/mpyne/game-music-emu/overview>.
42675 Intel QuickSync Video
42677 FFmpeg can use Intel QuickSync Video (QSV) for accelerated decoding and
42678 encoding of multiple codecs. To use QSV, FFmpeg must be linked against
42679 the "libmfx" dispatcher, which loads the actual decoding libraries.
42680 The dispatcher is open source and can be downloaded from
42682 <https://github.com/lu-zero/mfx_dispatch.git>. FFmpeg needs to be
42683 configured with the "--enable-libmfx" option and "pkg-config" needs to
42684 be able to locate the dispatcher's ".pc" files.
42685 Kvazaar
42687 FFmpeg can make use of the Kvazaar library for HEVC encoding.
42688 Go to <https://github.com/ultravideo/kvazaar> and follow the
42690 instructions for installing the library. Then pass
42691 "--enable-libkvazaar" to configure to enable it.
42692 LAME
42694 FFmpeg can make use of the LAME library for MP3 encoding.
42695 Go to <http://lame.sourceforge.net/> and follow the instructions for
42697 installing the library. Then pass "--enable-libmp3lame" to configure
42698 to enable it.
42699 libilbc
42701 iLBC is a narrowband speech codec that has been made freely available
42702 by Google as part of the WebRTC project. libilbc is a packaging
42703 friendly copy of the iLBC codec. FFmpeg can make use of the libilbc
42704 library for iLBC decoding and encoding.
42705 Go to <https://github.com/TimothyGu/libilbc> and follow the
42707 instructions for installing the library. Then pass "--enable-libilbc"
42708 to configure to enable it.
42709 libvpx
42711 FFmpeg can make use of the libvpx library for VP8/VP9 decoding and
42712 encoding.
42713 Go to <http://www.webmproject.org/> and follow the instructions for
42715 installing the library. Then pass "--enable-libvpx" to configure to
42716 enable it.
42717 ModPlug
42719 FFmpeg can make use of this library, originating in Modplug-XMMS, to
42720 read from MOD-like music files. See
42721 <https://github.com/Konstanty/libmodplug>. Pass "--enable-libmodplug"
42722 to configure to enable it.
42723 OpenCORE, VisualOn, and Fraunhofer libraries
42725 Spun off Google Android sources, OpenCore, VisualOn and Fraunhofer
42726 libraries provide encoders for a number of audio codecs.
42727 OpenCORE and VisualOn libraries are under the Apache License 2.0
42729 (see <http://www.apache.org/licenses/LICENSE-2.0> for details),
42730 which is incompatible to the LGPL version 2.1 and GPL version 2.
42731 You have to upgrade FFmpeg's license to LGPL version 3 (or if you
42732 have enabled GPL components, GPL version 3) by passing
42733 "--enable-version3" to configure in order to use it.
42734 The license of the Fraunhofer AAC library is incompatible with the
42736 GPL. Therefore, for GPL builds, you have to pass
42737 "--enable-nonfree" to configure in order to use it. To the best of
42738 our knowledge, it is compatible with the LGPL.
42739 OpenCORE AMR
42741 FFmpeg can make use of the OpenCORE libraries for AMR-NB
42743 decoding/encoding and AMR-WB decoding.
42744 Go to <http://sourceforge.net/projects/opencore-amr/> and follow the
42746 instructions for installing the libraries. Then pass
42747 "--enable-libopencore-amrnb" and/or "--enable-libopencore-amrwb" to
42748 configure to enable them.
42749 VisualOn AMR-WB encoder library
42751 FFmpeg can make use of the VisualOn AMR-WBenc library for AMR-WB
42753 encoding.
42754 Go to <http://sourceforge.net/projects/opencore-amr/> and follow the
42756 instructions for installing the library. Then pass
42757 "--enable-libvo-amrwbenc" to configure to enable it.
42758 Fraunhofer AAC library
42760 FFmpeg can make use of the Fraunhofer AAC library for AAC decoding &
42762 encoding.
42763 Go to <http://sourceforge.net/projects/opencore-amr/> and follow the
42765 instructions for installing the library. Then pass
42766 "--enable-libfdk-aac" to configure to enable it.
42767 OpenH264
42769 FFmpeg can make use of the OpenH264 library for H.264 decoding and
42770 encoding.
42771 Go to <http://www.openh264.org/> and follow the instructions for
42773 installing the library. Then pass "--enable-libopenh264" to configure
42774 to enable it.
42775 For decoding, this library is much more limited than the built-in
42777 decoder in libavcodec; currently, this library lacks support for
42778 decoding B-frames and some other main/high profile features. (It
42779 currently only supports constrained baseline profile and CABAC.) Using
42780 it is mostly useful for testing and for taking advantage of Cisco's
42781 patent portfolio license
42782 (<http://www.openh264.org/BINARY_LICENSE.txt>).
42783 OpenJPEG
42785 FFmpeg can use the OpenJPEG libraries for decoding/encoding J2K videos.
42786 Go to <http://www.openjpeg.org/> to get the libraries and follow the
42787 installation instructions. To enable using OpenJPEG in FFmpeg, pass
42788 "--enable-libopenjpeg" to ./configure.
42789 rav1e
42791 FFmpeg can make use of rav1e (Rust AV1 Encoder) via its C bindings to
42792 encode videos. Go to <https://github.com/xiph/rav1e/> and follow the
42793 instructions to build the C library. To enable using rav1e in FFmpeg,
42794 pass "--enable-librav1e" to ./configure.
42795 SVT-AV1
42797 FFmpeg can make use of the Scalable Video Technology for AV1 library
42798 for AV1 encoding.
42799 Go to <https://gitlab.com/AOMediaCodec/SVT-AV1/> and follow the
42801 instructions for installing the library. Then pass "--enable-libsvtav1"
42802 to configure to enable it.
42803 TwoLAME
42805 FFmpeg can make use of the TwoLAME library for MP2 encoding.
42806 Go to <http://www.twolame.org/> and follow the instructions for
42808 installing the library. Then pass "--enable-libtwolame" to configure
42809 to enable it.
42810 VapourSynth
42812 FFmpeg can read VapourSynth scripts as input. To enable support, pass
42813 "--enable-vapoursynth" to configure. Vapoursynth is detected via
42814 "pkg-config". Versions 42 or greater supported. See
42815 <http://www.vapoursynth.com/>.
42816 Due to security concerns, Vapoursynth scripts will not be autodetected
42818 so the input format has to be forced. For ff* CLI tools, add "-f
42819 vapoursynth" before the input "-i yourscript.vpy".
42820 x264
42822 FFmpeg can make use of the x264 library for H.264 encoding.
42823 Go to <http://www.videolan.org/developers/x264.html> and follow the
42825 instructions for installing the library. Then pass "--enable-libx264"
42826 to configure to enable it.
42827 x264 is under the GNU Public License Version 2 or later (see
42829 <http://www.gnu.org/licenses/old-licenses/gpl-2.0.html> for
42830 details), you must upgrade FFmpeg's license to GPL in order to use
42831 it.
42832 x265
42834 FFmpeg can make use of the x265 library for HEVC encoding.
42835 Go to <http://x265.org/developers.html> and follow the instructions for
42837 installing the library. Then pass "--enable-libx265" to configure to
42838 enable it.
42839 x265 is under the GNU Public License Version 2 or later (see
42841 <http://www.gnu.org/licenses/old-licenses/gpl-2.0.html> for
42842 details), you must upgrade FFmpeg's license to GPL in order to use
42843 it.
42844 xavs
42846 FFmpeg can make use of the xavs library for AVS encoding.
42847 Go to <http://xavs.sf.net/> and follow the instructions for installing
42849 the library. Then pass "--enable-libxavs" to configure to enable it.
42850 xavs2
42852 FFmpeg can make use of the xavs2 library for AVS2-P2/IEEE1857.4 video
42853 encoding.
42854 Go to <https://github.com/pkuvcl/xavs2> and follow the instructions for
42856 installing the library. Then pass "--enable-libxavs2" to configure to
42857 enable it.
42858 libxavs2 is under the GNU Public License Version 2 or later (see
42860 <http://www.gnu.org/licenses/old-licenses/gpl-2.0.html> for
42861 details), you must upgrade FFmpeg's license to GPL in order to use
42862 it.
42863 ZVBI
42865 ZVBI is a VBI decoding library which can be used by FFmpeg to decode
42866 DVB teletext pages and DVB teletext subtitles.
42867 Go to <http://sourceforge.net/projects/zapping/> and follow the
42869 instructions for installing the library. Then pass "--enable-libzvbi"
42870 to configure to enable it.
42871
42873 You can use the "-formats" and "-codecs" options to have an exhaustive
42874 list.
42875 File Formats
42877 FFmpeg supports the following file formats through the "libavformat"
42878 library:
42879 Name : Encoding @tab Decoding @tab Comments
42881 3dostr : @tab X
42882 4xm : @tab X
42883 @tab 4X Technologies format, used in some games.
42884 8088flex TMV : @tab X
42886 AAX : @tab X
42887 @tab Audible Enhanced Audio format, used in audiobooks.
42888 AA : @tab X
42890 @tab Audible Format 2, 3, and 4, used in audiobooks.
42891 ACT Voice : @tab X
42893 @tab contains G.729 audio
42894 Adobe Filmstrip : X @tab X
42896 Audio IFF (AIFF) : X @tab X
42897 American Laser Games MM : @tab X
42898 @tab Multimedia format used in games like Mad Dog McCree.
42899 3GPP AMR : X @tab X
42901 Amazing Studio Packed Animation File : @tab X
42902 @tab Multimedia format used in game Heart Of Darkness.
42903 Apple HTTP Live Streaming : @tab X
42905 Artworx Data Format : @tab X
42906 Interplay ACM : @tab X
42907 @tab Audio only format used in some Interplay games.
42908 ADP : @tab X
42910 @tab Audio format used on the Nintendo Gamecube.
42911 AFC : @tab X
42913 @tab Audio format used on the Nintendo Gamecube.
42914 ADS/SS2 : @tab X
42916 @tab Audio format used on the PS2.
42917 APNG : X @tab X
42919 ASF : X @tab X
42920 @tab Advanced / Active Streaming Format.
42921 AST : X @tab X
42923 @tab Audio format used on the Nintendo Wii.
42924 AVI : X @tab X
42926 AviSynth : @tab X
42927 AVR : @tab X
42928 @tab Audio format used on Mac.
42929 AVS : @tab X
42931 @tab Multimedia format used by the Creature Shock game.
42932 Beam Software SIFF : @tab X
42934 @tab Audio and video format used in some games by Beam Software.
42935 Bethesda Softworks VID : @tab X
42937 @tab Used in some games from Bethesda Softworks.
42938 Binary text : @tab X
42940 Bink : @tab X
42941 @tab Multimedia format used by many games.
42942 Bink Audio : @tab X
42944 @tab Audio only multimedia format used by some games.
42945 Bitmap Brothers JV : @tab X
42947 @tab Used in Z and Z95 games.
42948 BRP : @tab X
42950 @tab Argonaut Games format.
42951 Brute Force & Ignorance : @tab X
42953 @tab Used in the game Flash Traffic: City of Angels.
42954 BFSTM : @tab X
42956 @tab Audio format used on the Nintendo WiiU (based on BRSTM).
42957 BRSTM : @tab X
42959 @tab Audio format used on the Nintendo Wii.
42960 BW64 : @tab X
42962 @tab Broadcast Wave 64bit.
42963 BWF : X @tab X
42965 codec2 (raw) : X @tab X
42966 @tab Must be given -mode format option to decode correctly.
42967 codec2 (.c2 files) : X @tab X
42969 @tab Contains header with version and mode info, simplifying playback.
42970 CRI ADX : X @tab X
42972 @tab Audio-only format used in console video games.
42973 CRI AIX : @tab X
42975 CRI HCA : @tab X
42976 @tab Audio-only format used in console video games.
42977 Discworld II BMV : @tab X
42979 Interplay C93 : @tab X
42980 @tab Used in the game Cyberia from Interplay.
42981 Delphine Software International CIN : @tab X
42983 @tab Multimedia format used by Delphine Software games.
42984 Digital Speech Standard (DSS) : @tab X
42986 CD+G : @tab X
42987 @tab Video format used by CD+G karaoke disks
42988 Phantom Cine : @tab X
42990 Commodore CDXL : @tab X
42991 @tab Amiga CD video format
42992 Core Audio Format : X @tab X
42994 @tab Apple Core Audio Format
42995 CRC testing format : X @tab
42997 Creative Voice : X @tab X
42998 @tab Created for the Sound Blaster Pro.
42999 CRYO APC : @tab X
43001 @tab Audio format used in some games by CRYO Interactive Entertainment.
43002 D-Cinema audio : X @tab X
43004 Deluxe Paint Animation : @tab X
43005 DCSTR : @tab X
43006 DFA : @tab X
43007 @tab This format is used in Chronomaster game
43008 DirectDraw Surface : @tab X
43010 DSD Stream File (DSF) : @tab X
43011 DV video : X @tab X
43012 DXA : @tab X
43013 @tab This format is used in the non-Windows version of the Feeble Files
43014 game and different game cutscenes repacked for use with ScummVM.
43015 Electronic Arts cdata : @tab X
43017 Electronic Arts Multimedia : @tab X
43018 @tab Used in various EA games; files have extensions like WVE and UV2.
43019 Ensoniq Paris Audio File : @tab X
43021 FFM (FFserver live feed) : X @tab X
43022 Flash (SWF) : X @tab X
43023 Flash 9 (AVM2) : X @tab X
43024 @tab Only embedded audio is decoded.
43025 FLI/FLC/FLX animation : @tab X
43027 @tab .fli/.flc files
43028 Flash Video (FLV) : X @tab X
43030 @tab Macromedia Flash video files
43031 framecrc testing format : X @tab
43033 FunCom ISS : @tab X
43034 @tab Audio format used in various games from FunCom like The Longest Journey.
43035 G.723.1 : X @tab X
43037 G.726 : @tab X @tab Both left- and right-
43038 justified.
43039 G.729 BIT : X @tab X
43040 G.729 raw : @tab X
43041 GENH : @tab X
43042 @tab Audio format for various games.
43043 GIF Animation : X @tab X
43045 GXF : X @tab X
43046 @tab General eXchange Format SMPTE 360M, used by Thomson Grass Valley
43047 playout servers.
43048 HNM : @tab X
43050 @tab Only version 4 supported, used in some games from Cryo Interactive
43051 iCEDraw File : @tab X
43053 ICO : X @tab X
43054 @tab Microsoft Windows ICO
43055 id Quake II CIN video : @tab X
43057 id RoQ : X @tab X
43058 @tab Used in Quake III, Jedi Knight 2 and other computer games.
43059 IEC61937 encapsulation : X @tab X
43061 IFF : @tab X
43062 @tab Interchange File Format
43063 IFV : @tab X
43065 @tab A format used by some old CCTV DVRs.
43066 iLBC : X @tab X
43068 Interplay MVE : @tab X
43069 @tab Format used in various Interplay computer games.
43070 Iterated Systems ClearVideo : @tab X
43072 @tab I-frames only
43073 IV8 : @tab X
43075 @tab A format generated by IndigoVision 8000 video server.
43076 IVF (On2) : X @tab X
43078 @tab A format used by libvpx
43079 Internet Video Recording : @tab X
43081 IRCAM : X @tab X
43082 LATM : X @tab X
43083 LMLM4 : @tab X
43084 @tab Used by Linux Media Labs MPEG-4 PCI boards
43085 LOAS : @tab X
43087 @tab contains LATM multiplexed AAC audio
43088 LRC : X @tab X
43090 LVF : @tab X
43091 LXF : @tab X
43092 @tab VR native stream format, used by Leitch/Harris' video servers.
43093 Magic Lantern Video (MLV) : @tab X
43095 Matroska : X @tab X
43096 Matroska audio : X @tab
43097 FFmpeg metadata : X @tab X
43098 @tab Metadata in text format.
43099 MAXIS XA : @tab X
43101 @tab Used in Sim City 3000; file extension .xa.
43102 MCA : @tab X
43104 @tab Used in some games from Capcom; file extension .mca.
43105 MD Studio : @tab X
43107 Metal Gear Solid: The Twin Snakes : @tab X
43108 Megalux Frame : @tab X
43109 @tab Used by Megalux Ultimate Paint
43110 Mobotix .mxg : @tab X
43112 Monkey's Audio : @tab X
43113 Motion Pixels MVI : @tab X
43114 MOV/QuickTime/MP4 : X @tab X
43115 @tab 3GP, 3GP2, PSP, iPod variants supported
43116 MP2 : X @tab X
43118 MP3 : X @tab X
43119 MPEG-1 System : X @tab X
43120 @tab muxed audio and video, VCD format supported
43121 MPEG-PS (program stream) : X @tab X
43123 @tab also known as C<VOB> file, SVCD and DVD format supported
43124 MPEG-TS (transport stream) : X @tab X
43126 @tab also known as DVB Transport Stream
43127 MPEG-4 : X @tab X
43129 @tab MPEG-4 is a variant of QuickTime.
43130 MSF : @tab X
43132 @tab Audio format used on the PS3.
43133 Mirillis FIC video : @tab X
43135 @tab No cursor rendering.
43136 MIDI Sample Dump Standard : @tab X
43138 MIME multipart JPEG : X @tab
43139 MSN TCP webcam : @tab X
43140 @tab Used by MSN Messenger webcam streams.
43141 MTV : @tab X
43143 Musepack : @tab X
43144 Musepack SV8 : @tab X
43145 Material eXchange Format (MXF) : X @tab X
43146 @tab SMPTE 377M, used by D-Cinema, broadcast industry.
43147 Material eXchange Format (MXF), D-10 Mapping : X @tab X
43149 @tab SMPTE 386M, D-10/IMX Mapping.
43150 NC camera feed : @tab X
43152 @tab NC (AVIP NC4600) camera streams
43153 NIST SPeech HEader REsources : @tab X
43155 Computerized Speech Lab NSP : @tab X
43156 NTT TwinVQ (VQF) : @tab X
43157 @tab Nippon Telegraph and Telephone Corporation TwinVQ.
43158 Nullsoft Streaming Video : @tab X
43160 NuppelVideo : @tab X
43161 NUT : X @tab X
43162 @tab NUT Open Container Format
43163 Ogg : X @tab X
43165 Playstation Portable PMP : @tab X
43166 Portable Voice Format : @tab X
43167 TechnoTrend PVA : @tab X
43168 @tab Used by TechnoTrend DVB PCI boards.
43169 QCP : @tab X
43171 raw ADTS (AAC) : X @tab X
43172 raw AC-3 : X @tab X
43173 raw AMR-NB : @tab X
43174 raw AMR-WB : @tab X
43175 raw aptX : X @tab X
43176 raw aptX HD : X @tab X
43177 raw Chinese AVS video : X @tab X
43178 raw Dirac : X @tab X
43179 raw DNxHD : X @tab X
43180 raw DTS : X @tab X
43181 raw DTS-HD : @tab X
43182 raw E-AC-3 : X @tab X
43183 raw FLAC : X @tab X
43184 raw GSM : @tab X
43185 raw H.261 : X @tab X
43186 raw H.263 : X @tab X
43187 raw H.264 : X @tab X
43188 raw HEVC : X @tab X
43189 raw Ingenient MJPEG : @tab X
43190 raw MJPEG : X @tab X
43191 raw MLP : @tab X
43192 raw MPEG : @tab X
43193 raw MPEG-1 : @tab X
43194 raw MPEG-2 : @tab X
43195 raw MPEG-4 : X @tab X
43196 raw NULL : X @tab
43197 raw video : X @tab X
43198 raw id RoQ : X @tab
43199 raw OBU : X @tab X
43200 raw SBC : X @tab X
43201 raw Shorten : @tab X
43202 raw TAK : @tab X
43203 raw TrueHD : X @tab X
43204 raw VC-1 : X @tab X
43205 raw PCM A-law : X @tab X
43206 raw PCM mu-law : X @tab X
43207 raw PCM Archimedes VIDC : X @tab X
43208 raw PCM signed 8 bit : X @tab X
43209 raw PCM signed 16 bit big-endian : X @tab X
43210 raw PCM signed 16 bit little-endian : X @tab X
43211 raw PCM signed 24 bit big-endian : X @tab X
43212 raw PCM signed 24 bit little-endian : X @tab X
43213 raw PCM signed 32 bit big-endian : X @tab X
43214 raw PCM signed 32 bit little-endian : X @tab X
43215 raw PCM signed 64 bit big-endian : X @tab X
43216 raw PCM signed 64 bit little-endian : X @tab X
43217 raw PCM unsigned 8 bit : X @tab X
43218 raw PCM unsigned 16 bit big-endian : X @tab X
43219 raw PCM unsigned 16 bit little-endian : X @tab X
43220 raw PCM unsigned 24 bit big-endian : X @tab X
43221 raw PCM unsigned 24 bit little-endian : X @tab X
43222 raw PCM unsigned 32 bit big-endian : X @tab X
43223 raw PCM unsigned 32 bit little-endian : X @tab X
43224 raw PCM 16.8 floating point little-endian : @tab X
43225 raw PCM 24.0 floating point little-endian : @tab X
43226 raw PCM floating-point 32 bit big-endian : X @tab X
43227 raw PCM floating-point 32 bit little-endian : X @tab X
43228 raw PCM floating-point 64 bit big-endian : X @tab X
43229 raw PCM floating-point 64 bit little-endian : X @tab X
43230 RDT : @tab X
43231 REDCODE R3D : @tab X
43232 @tab File format used by RED Digital cameras, contains JPEG 2000 frames and PCM audio.
43233 RealMedia : X @tab X
43235 Redirector : @tab X
43236 RedSpark : @tab X
43237 Renderware TeXture Dictionary : @tab X
43238 Resolume DXV : @tab X
43239 RF64 : @tab X
43240 RL2 : @tab X
43241 @tab Audio and video format used in some games by Entertainment Software Partners.
43242 RPL/ARMovie : @tab X
43244 Lego Mindstorms RSO : X @tab X
43245 RSD : @tab X
43246 RTMP : X @tab X
43247 @tab Output is performed by publishing stream to RTMP server
43248 RTP : X @tab X
43250 RTSP : X @tab X
43251 Sample Dump eXchange : @tab X
43252 SAP : X @tab X
43253 SBG : @tab X
43254 SDP : @tab X
43255 SER : @tab X
43256 Sega FILM/CPK : X @tab X
43257 @tab Used in many Sega Saturn console games.
43258 Silicon Graphics Movie : @tab X
43260 Sierra SOL : @tab X
43261 @tab .sol files used in Sierra Online games.
43262 Sierra VMD : @tab X
43264 @tab Used in Sierra CD-ROM games.
43265 Smacker : @tab X
43267 @tab Multimedia format used by many games.
43268 SMJPEG : X @tab X
43270 @tab Used in certain Loki game ports.
43271 SMPTE 337M encapsulation : @tab X
43273 Smush : @tab X
43274 @tab Multimedia format used in some LucasArts games.
43275 Sony OpenMG (OMA) : X @tab X
43277 @tab Audio format used in Sony Sonic Stage and Sony Vegas.
43278 Sony PlayStation STR : @tab X
43280 Sony Wave64 (W64) : X @tab X
43281 SoX native format : X @tab X
43282 SUN AU format : X @tab X
43283 SUP raw PGS subtitles : X @tab X
43284 SVAG : @tab X
43285 @tab Audio format used in Konami PS2 games.
43286 TDSC : @tab X
43288 Text files : @tab X
43289 THP : @tab X
43290 @tab Used on the Nintendo GameCube.
43291 Tiertex Limited SEQ : @tab X
43293 @tab Tiertex .seq files used in the DOS CD-ROM version of the game Flashback.
43294 True Audio : X @tab X
43296 VAG : @tab X
43297 @tab Audio format used in many Sony PS2 games.
43298 VC-1 test bitstream : X @tab X
43300 Vidvox Hap : X @tab X
43301 Vivo : @tab X
43302 VPK : @tab X
43303 @tab Audio format used in Sony PS games.
43304 WAV : X @tab X
43306 WavPack : X @tab X
43307 WebM : X @tab X
43308 Windows Televison (WTV) : X @tab X
43309 Wing Commander III movie : @tab X
43310 @tab Multimedia format used in Origin's Wing Commander III computer game.
43311 Westwood Studios audio : X @tab X
43313 @tab Multimedia format used in Westwood Studios games.
43314 Westwood Studios VQA : @tab X
43316 @tab Multimedia format used in Westwood Studios games.
43317 Wideband Single-bit Data (WSD) : @tab X
43319 WVE : @tab X
43320 XMV : @tab X
43321 @tab Microsoft video container used in Xbox games.
43322 XVAG : @tab X
43324 @tab Audio format used on the PS3.
43325 xWMA : @tab X
43327 @tab Microsoft audio container used by XAudio 2.
43328 eXtended BINary text (XBIN) : @tab X
43330 YUV4MPEG pipe : X @tab X
43331 Psygnosis YOP : @tab X
43332 "X" means that the feature in that column (encoding / decoding) is
43334 supported.
43335 Image Formats
43337 FFmpeg can read and write images for each frame of a video sequence.
43338 The following image formats are supported:
43339 Name : Encoding @tab Decoding @tab Comments
43341 .Y.U.V : X @tab X
43342 @tab one raw file per component
43343 Alias PIX : X @tab X
43345 @tab Alias/Wavefront PIX image format
43346 animated GIF : X @tab X
43348 APNG : X @tab X
43349 @tab Animated Portable Network Graphics
43350 BMP : X @tab X
43352 @tab Microsoft BMP image
43353 BRender PIX : @tab X
43355 @tab Argonaut BRender 3D engine image format.
43356 CRI : @tab X
43358 @tab Cintel RAW
43359 DPX : X @tab X
43361 @tab Digital Picture Exchange
43362 EXR : @tab X
43364 @tab OpenEXR
43365 FITS : X @tab X
43367 @tab Flexible Image Transport System
43368 IMG : @tab X
43370 @tab GEM Raster image
43371 JPEG : X @tab X
43373 @tab Progressive JPEG is not supported.
43374 JPEG 2000 : X @tab X
43376 JPEG-LS : X @tab X
43377 LJPEG : X @tab
43378 @tab Lossless JPEG
43379 MSP : @tab X
43381 @tab Microsoft Paint image
43382 PAM : X @tab X
43384 @tab PAM is a PNM extension with alpha support.
43385 PBM : X @tab X
43387 @tab Portable BitMap image
43388 PCD : @tab X
43390 @tab PhotoCD
43391 PCX : X @tab X
43393 @tab PC Paintbrush
43394 PFM : X @tab X
43396 @tab Portable FloatMap image
43397 PGM : X @tab X
43399 @tab Portable GrayMap image
43400 PGMYUV : X @tab X
43402 @tab PGM with U and V components in YUV 4:2:0
43403 PGX : @tab X
43405 @tab PGX file decoder
43406 PIC : @tab X
43408 @tab Pictor/PC Paint
43409 PNG : X @tab X
43411 @tab Portable Network Graphics image
43412 PPM : X @tab X
43414 @tab Portable PixelMap image
43415 PSD : @tab X
43417 @tab Photoshop
43418 PTX : @tab X
43420 @tab V.Flash PTX format
43421 SGI : X @tab X
43423 @tab SGI RGB image format
43424 Sun Rasterfile : X @tab X
43426 @tab Sun RAS image format
43427 TIFF : X @tab X
43429 @tab YUV, JPEG and some extension is not supported yet.
43430 Truevision Targa : X @tab X
43432 @tab Targa (.TGA) image format
43433 WebP : E @tab X
43435 @tab WebP image format, encoding supported through external library libwebp
43436 XBM : X @tab X
43438 @tab X BitMap image format
43439 XFace : X @tab X
43441 @tab X-Face image format
43442 XPM : @tab X
43444 @tab X PixMap image format
43445 XWD : X @tab X
43447 @tab X Window Dump image format
43448 "X" means that the feature in that column (encoding / decoding) is
43450 supported.
43451 "E" means that support is provided through an external library.
43453 Video Codecs
43455 Name : Encoding @tab Decoding @tab Comments
43456 4X Movie : @tab X
43457 @tab Used in certain computer games.
43458 8088flex TMV : @tab X
43460 A64 multicolor : X @tab
43461 @tab Creates video suitable to be played on a commodore 64 (multicolor mode).
43462 Amazing Studio PAF Video : @tab X
43464 American Laser Games MM : @tab X
43465 @tab Used in games like Mad Dog McCree.
43466 Amuse Graphics Movie : @tab X
43468 AMV Video : X @tab X
43469 @tab Used in Chinese MP3 players.
43470 ANSI/ASCII art : @tab X
43472 Apple Intermediate Codec : @tab X
43473 Apple MJPEG-B : @tab X
43474 Apple Pixlet : @tab X
43475 Apple ProRes : X @tab X
43476 @tab fourcc: apch,apcn,apcs,apco,ap4h,ap4x
43477 Apple QuickDraw : @tab X
43479 @tab fourcc: qdrw
43480 Argonaut Video : @tab X
43482 @tab Used in some Argonaut games.
43483 Asus v1 : X @tab X
43485 @tab fourcc: ASV1
43486 Asus v2 : X @tab X
43488 @tab fourcc: ASV2
43489 ATI VCR1 : @tab X
43491 @tab fourcc: VCR1
43492 ATI VCR2 : @tab X
43494 @tab fourcc: VCR2
43495 Auravision Aura : @tab X
43497 Auravision Aura 2 : @tab X
43498 Autodesk Animator Flic video : @tab X
43499 Autodesk RLE : @tab X
43500 @tab fourcc: AASC
43501 AV1 : E @tab E
43503 @tab Supported through external libraries libaom, libdav1d, librav1e and libsvtav1
43504 Avid 1:1 10-bit RGB Packer : X @tab X
43506 @tab fourcc: AVrp
43507 AVS (Audio Video Standard) video : @tab X
43509 @tab Video encoding used by the Creature Shock game.
43510 AVS2-P2/IEEE1857.4 : E @tab E
43512 @tab Supported through external libraries libxavs2 and libdavs2
43513 AVS3-P2/IEEE1857.10 : @tab E
43515 @tab Supported through external library libuavs3d
43516 AYUV : X @tab X
43518 @tab Microsoft uncompressed packed 4:4:4:4
43519 Beam Software VB : @tab X
43521 Bethesda VID video : @tab X
43522 @tab Used in some games from Bethesda Softworks.
43523 Bink Video : @tab X
43525 BitJazz SheerVideo : @tab X
43526 Bitmap Brothers JV video : @tab X
43527 y41p Brooktree uncompressed 4:1:1 12-bit : X @tab X
43528 Brooktree ProSumer Video : @tab X
43529 @tab fourcc: BT20
43530 Brute Force & Ignorance : @tab X
43532 @tab Used in the game Flash Traffic: City of Angels.
43533 C93 video : @tab X
43535 @tab Codec used in Cyberia game.
43536 CamStudio : @tab X
43538 @tab fourcc: CSCD
43539 CD+G : @tab X
43541 @tab Video codec for CD+G karaoke disks
43542 CDXL : @tab X
43544 @tab Amiga CD video codec
43545 Chinese AVS video : E @tab X
43547 @tab AVS1-P2, JiZhun profile, encoding through external library libxavs
43548 Delphine Software International CIN video : @tab X
43550 @tab Codec used in Delphine Software International games.
43551 Discworld II BMV Video : @tab X
43553 CineForm HD : X @tab X
43554 Canopus HQ : @tab X
43555 Canopus HQA : @tab X
43556 Canopus HQX : @tab X
43557 Canopus Lossless Codec : @tab X
43558 CDToons : @tab X
43559 @tab Codec used in various Broderbund games.
43560 Cinepak : @tab X
43562 Cirrus Logic AccuPak : X @tab X
43563 @tab fourcc: CLJR
43564 CPiA Video Format : @tab X
43566 Creative YUV (CYUV) : @tab X
43567 DFA : @tab X
43568 @tab Codec used in Chronomaster game.
43569 Dirac : E @tab X
43571 @tab supported though the native vc2 (Dirac Pro) encoder
43572 Deluxe Paint Animation : @tab X
43574 DNxHD : X @tab X
43575 @tab aka SMPTE VC3
43576 Duck TrueMotion 1.0 : @tab X
43578 @tab fourcc: DUCK
43579 Duck TrueMotion 2.0 : @tab X
43581 @tab fourcc: TM20
43582 Duck TrueMotion 2.0 RT : @tab X
43584 @tab fourcc: TR20
43585 DV (Digital Video) : X @tab X
43587 Dxtory capture format : @tab X
43588 Feeble Files/ScummVM DXA : @tab X
43589 @tab Codec originally used in Feeble Files game.
43590 Electronic Arts CMV video : @tab X
43592 @tab Used in NHL 95 game.
43593 Electronic Arts Madcow video : @tab X
43595 Electronic Arts TGV video : @tab X
43596 Electronic Arts TGQ video : @tab X
43597 Electronic Arts TQI video : @tab X
43598 Escape 124 : @tab X
43599 Escape 130 : @tab X
43600 FFmpeg video codec #1 : X @tab X
43601 @tab lossless codec (fourcc: FFV1)
43602 Flash Screen Video v1 : X @tab X
43604 @tab fourcc: FSV1
43605 Flash Screen Video v2 : X @tab X
43607 Flash Video (FLV) : X @tab X
43608 @tab Sorenson H.263 used in Flash
43609 FM Screen Capture Codec : @tab X
43611 Forward Uncompressed : @tab X
43612 Fraps : @tab X
43613 Go2Meeting : @tab X
43614 @tab fourcc: G2M2, G2M3
43615 Go2Webinar : @tab X
43617 @tab fourcc: G2M4
43618 Gremlin Digital Video : @tab X
43620 H.261 : X @tab X
43621 H.263 / H.263-1996 : X @tab X
43622 H.263+ / H.263-1998 / H.263 version 2 : X @tab X
43623 H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10 : E @tab X
43624 @tab encoding supported through external library libx264 and OpenH264
43625 HEVC : X @tab X
43627 @tab encoding supported through external library libx265 and libkvazaar
43628 HNM version 4 : @tab X
43630 HuffYUV : X @tab X
43631 HuffYUV FFmpeg variant : X @tab X
43632 IBM Ultimotion : @tab X
43633 @tab fourcc: ULTI
43634 id Cinematic video : @tab X
43636 @tab Used in Quake II.
43637 id RoQ video : X @tab X
43639 @tab Used in Quake III, Jedi Knight 2, other computer games.
43640 IFF ILBM : @tab X
43642 @tab IFF interleaved bitmap
43643 IFF ByteRun1 : @tab X
43645 @tab IFF run length encoded bitmap
43646 Infinity IMM4 : @tab X
43648 Intel H.263 : @tab X
43649 Intel Indeo 2 : @tab X
43650 Intel Indeo 3 : @tab X
43651 Intel Indeo 4 : @tab X
43652 Intel Indeo 5 : @tab X
43653 Interplay C93 : @tab X
43654 @tab Used in the game Cyberia from Interplay.
43655 Interplay MVE video : @tab X
43657 @tab Used in Interplay .MVE files.
43658 J2K : X @tab X
43660 Karl Morton's video codec : @tab X
43661 @tab Codec used in Worms games.
43662 Kega Game Video (KGV1) : @tab X
43664 @tab Kega emulator screen capture codec.
43665 Lagarith : @tab X
43667 LCL (LossLess Codec Library) MSZH : @tab X
43668 LCL (LossLess Codec Library) ZLIB : E @tab E
43669 LOCO : @tab X
43670 LucasArts SANM/Smush : @tab X
43671 @tab Used in LucasArts games / SMUSH animations.
43672 lossless MJPEG : X @tab X
43674 MagicYUV Video : X @tab X
43675 Mandsoft Screen Capture Codec : @tab X
43676 Microsoft ATC Screen : @tab X
43677 @tab Also known as Microsoft Screen 3.
43678 Microsoft Expression Encoder Screen : @tab X
43680 @tab Also known as Microsoft Titanium Screen 2.
43681 Microsoft RLE : @tab X
43683 Microsoft Screen 1 : @tab X
43684 @tab Also known as Windows Media Video V7 Screen.
43685 Microsoft Screen 2 : @tab X
43687 @tab Also known as Windows Media Video V9 Screen.
43688 Microsoft Video 1 : @tab X
43690 Mimic : @tab X
43691 @tab Used in MSN Messenger Webcam streams.
43692 Miro VideoXL : @tab X
43694 @tab fourcc: VIXL
43695 MJPEG (Motion JPEG) : X @tab X
43697 Mobotix MxPEG video : @tab X
43698 Motion Pixels video : @tab X
43699 MPEG-1 video : X @tab X
43700 MPEG-2 video : X @tab X
43701 MPEG-4 part 2 : X @tab X
43702 @tab libxvidcore can be used alternatively for encoding.
43703 MPEG-4 part 2 Microsoft variant version 1 : @tab X
43705 MPEG-4 part 2 Microsoft variant version 2 : X @tab X
43706 MPEG-4 part 2 Microsoft variant version 3 : X @tab X
43707 Newtek SpeedHQ : X @tab X
43708 Nintendo Gamecube THP video : @tab X
43709 NotchLC : @tab X
43710 NuppelVideo/RTjpeg : @tab X
43711 @tab Video encoding used in NuppelVideo files.
43712 On2 VP3 : @tab X
43714 @tab still experimental
43715 On2 VP4 : @tab X
43717 @tab fourcc: VP40
43718 On2 VP5 : @tab X
43720 @tab fourcc: VP50
43721 On2 VP6 : @tab X
43723 @tab fourcc: VP60,VP61,VP62
43724 On2 VP7 : @tab X
43726 @tab fourcc: VP70,VP71
43727 VP8 : E @tab X
43729 @tab fourcc: VP80, encoding supported through external library libvpx
43730 VP9 : E @tab X
43732 @tab encoding supported through external library libvpx
43733 Pinnacle TARGA CineWave YUV16 : @tab X
43735 @tab fourcc: Y216
43736 Q-team QPEG : @tab X
43738 @tab fourccs: QPEG, Q1.0, Q1.1
43739 QuickTime 8BPS video : @tab X
43741 QuickTime Animation (RLE) video : X @tab X
43742 @tab fourcc: 'rle '
43743 QuickTime Graphics (SMC) : X @tab X
43745 @tab fourcc: 'smc '
43746 QuickTime video (RPZA) : X @tab X
43748 @tab fourcc: rpza
43749 R10K AJA Kona 10-bit RGB Codec : X @tab X
43751 R210 Quicktime Uncompressed RGB 10-bit : X @tab X
43752 Raw Video : X @tab X
43753 RealVideo 1.0 : X @tab X
43754 RealVideo 2.0 : X @tab X
43755 RealVideo 3.0 : @tab X
43756 @tab still far from ideal
43757 RealVideo 4.0 : @tab X
43759 Renderware TXD (TeXture Dictionary) : @tab X
43760 @tab Texture dictionaries used by the Renderware Engine.
43761 RL2 video : @tab X
43763 @tab used in some games by Entertainment Software Partners
43764 ScreenPressor : @tab X
43766 Screenpresso : @tab X
43767 Screen Recorder Gold Codec : @tab X
43768 Sierra VMD video : @tab X
43769 @tab Used in Sierra VMD files.
43770 Silicon Graphics Motion Video Compressor 1 (MVC1) : @tab X
43772 Silicon Graphics Motion Video Compressor 2 (MVC2) : @tab X
43773 Silicon Graphics RLE 8-bit video : @tab X
43774 Smacker video : @tab X
43775 @tab Video encoding used in Smacker.
43776 SMPTE VC-1 : @tab X
43778 Snow : X @tab X
43779 @tab experimental wavelet codec (fourcc: SNOW)
43780 Sony PlayStation MDEC (Motion DECoder) : @tab X
43782 Sorenson Vector Quantizer 1 : X @tab X
43783 @tab fourcc: SVQ1
43784 Sorenson Vector Quantizer 3 : @tab X
43786 @tab fourcc: SVQ3
43787 Sunplus JPEG (SP5X) : @tab X
43789 @tab fourcc: SP5X
43790 TechSmith Screen Capture Codec : @tab X
43792 @tab fourcc: TSCC
43793 TechSmith Screen Capture Codec 2 : @tab X
43795 @tab fourcc: TSC2
43796 Theora : E @tab X
43798 @tab encoding supported through external library libtheora
43799 Tiertex Limited SEQ video : @tab X
43801 @tab Codec used in DOS CD-ROM FlashBack game.
43802 Ut Video : X @tab X
43804 v210 QuickTime uncompressed 4:2:2 10-bit : X @tab X
43805 v308 QuickTime uncompressed 4:4:4 : X @tab X
43806 v408 QuickTime uncompressed 4:4:4:4 : X @tab X
43807 v410 QuickTime uncompressed 4:4:4 10-bit : X @tab X
43808 VBLE Lossless Codec : @tab X
43809 VMware Screen Codec / VMware Video : @tab X
43810 @tab Codec used in videos captured by VMware.
43811 Westwood Studios VQA (Vector Quantized Animation) video : @tab
43813 X
43814 Windows Media Image : @tab X
43815 Windows Media Video 7 : X @tab X
43816 Windows Media Video 8 : X @tab X
43817 Windows Media Video 9 : @tab X
43818 @tab not completely working
43819 Wing Commander III / Xan : @tab X
43821 @tab Used in Wing Commander III .MVE files.
43822 Wing Commander IV / Xan : @tab X
43824 @tab Used in Wing Commander IV.
43825 Winnov WNV1 : @tab X
43827 WMV7 : X @tab X
43828 YAMAHA SMAF : X @tab X
43829 Psygnosis YOP Video : @tab X
43830 yuv4 : X @tab X
43831 @tab libquicktime uncompressed packed 4:2:0
43832 ZeroCodec Lossless Video : @tab X
43834 ZLIB : X @tab X
43835 @tab part of LCL, encoder experimental
43836 Zip Motion Blocks Video : X @tab X
43838 @tab Encoder works only in PAL8.
43839 "X" means that the feature in that column (encoding / decoding) is
43841 supported.
43842 "E" means that support is provided through an external library.
43844 Audio Codecs
43846 Name : Encoding @tab Decoding @tab Comments
43847 8SVX exponential : @tab X
43848 8SVX fibonacci : @tab X
43849 AAC : EX @tab X
43850 @tab encoding supported through internal encoder and external library libfdk-aac
43851 AAC+ : E @tab IX
43853 @tab encoding supported through external library libfdk-aac
43854 AC-3 : IX @tab IX
43856 ACELP.KELVIN : @tab X
43857 ADPCM 4X Movie : @tab X
43858 ADPCM Yamaha AICA : @tab X
43859 ADPCM AmuseGraphics Movie : @tab X
43860 ADPCM Argonaut Games : X @tab X
43861 ADPCM CDROM XA : @tab X
43862 ADPCM Creative Technology : @tab X
43863 @tab 16 -E<gt> 4, 8 -E<gt> 4, 8 -E<gt> 3, 8 -E<gt> 2
43864 ADPCM Electronic Arts : @tab X
43866 @tab Used in various EA titles.
43867 ADPCM Electronic Arts Maxis CDROM XS : @tab X
43869 @tab Used in Sim City 3000.
43870 ADPCM Electronic Arts R1 : @tab X
43872 ADPCM Electronic Arts R2 : @tab X
43873 ADPCM Electronic Arts R3 : @tab X
43874 ADPCM Electronic Arts XAS : @tab X
43875 ADPCM G.722 : X @tab X
43876 ADPCM G.726 : X @tab X
43877 ADPCM IMA Acorn Replay : @tab X
43878 ADPCM IMA AMV : X @tab X
43879 @tab Used in AMV files
43880 ADPCM IMA Cunning Developments : @tab X
43882 ADPCM IMA Electronic Arts EACS : @tab X
43883 ADPCM IMA Electronic Arts SEAD : @tab X
43884 ADPCM IMA Funcom : @tab X
43885 ADPCM IMA High Voltage Software ALP : X @tab X
43886 ADPCM IMA QuickTime : X @tab X
43887 ADPCM IMA Simon & Schuster Interactive : X @tab X
43888 ADPCM IMA Ubisoft APM : X @tab X
43889 ADPCM IMA Loki SDL MJPEG : @tab X
43890 ADPCM IMA WAV : X @tab X
43891 ADPCM IMA Westwood : @tab X
43892 ADPCM ISS IMA : @tab X
43893 @tab Used in FunCom games.
43894 ADPCM IMA Dialogic : @tab X
43896 ADPCM IMA Duck DK3 : @tab X
43897 @tab Used in some Sega Saturn console games.
43898 ADPCM IMA Duck DK4 : @tab X
43900 @tab Used in some Sega Saturn console games.
43901 ADPCM IMA Radical : @tab X
43903 ADPCM Microsoft : X @tab X
43904 ADPCM MS IMA : X @tab X
43905 ADPCM Nintendo Gamecube AFC : @tab X
43906 ADPCM Nintendo Gamecube DTK : @tab X
43907 ADPCM Nintendo THP : @tab X
43908 ADPCM Playstation : @tab X
43909 ADPCM QT IMA : X @tab X
43910 ADPCM SEGA CRI ADX : X @tab X
43911 @tab Used in Sega Dreamcast games.
43912 ADPCM Shockwave Flash : X @tab X
43914 ADPCM Sound Blaster Pro 2-bit : @tab X
43915 ADPCM Sound Blaster Pro 2.6-bit : @tab X
43916 ADPCM Sound Blaster Pro 4-bit : @tab X
43917 ADPCM VIMA : @tab X
43918 @tab Used in LucasArts SMUSH animations.
43919 ADPCM Westwood Studios IMA : X @tab X
43921 @tab Used in Westwood Studios games like Command and Conquer.
43922 ADPCM Yamaha : X @tab X
43924 ADPCM Zork : @tab X
43925 AMR-NB : E @tab X
43926 @tab encoding supported through external library libopencore-amrnb
43927 AMR-WB : E @tab X
43929 @tab encoding supported through external library libvo-amrwbenc
43930 Amazing Studio PAF Audio : @tab X
43932 Apple lossless audio : X @tab X
43933 @tab QuickTime fourcc 'alac'
43934 aptX : X @tab X
43936 @tab Used in Bluetooth A2DP
43937 aptX HD : X @tab X
43939 @tab Used in Bluetooth A2DP
43940 ATRAC1 : @tab X
43942 ATRAC3 : @tab X
43943 ATRAC3+ : @tab X
43944 ATRAC9 : @tab X
43945 Bink Audio : @tab X
43946 @tab Used in Bink and Smacker files in many games.
43947 CELT : @tab E
43949 @tab decoding supported through external library libcelt
43950 codec2 : E @tab E
43952 @tab en/decoding supported through external library libcodec2
43953 CRI HCA : @tab X
43955 Delphine Software International CIN audio : @tab X
43956 @tab Codec used in Delphine Software International games.
43957 Digital Speech Standard - Standard Play mode (DSS SP) : @tab X
43959 Discworld II BMV Audio : @tab X
43960 COOK : @tab X
43961 @tab All versions except 5.1 are supported.
43962 DCA (DTS Coherent Acoustics) : X @tab X
43964 @tab supported extensions: XCh, XXCH, X96, XBR, XLL, LBR (partially)
43965 Dolby E : @tab X
43967 DPCM Gremlin : @tab X
43968 DPCM id RoQ : X @tab X
43969 @tab Used in Quake III, Jedi Knight 2 and other computer games.
43970 DPCM Interplay : @tab X
43972 @tab Used in various Interplay computer games.
43973 DPCM Squareroot-Delta-Exact : @tab X
43975 @tab Used in various games.
43976 DPCM Sierra Online : @tab X
43978 @tab Used in Sierra Online game audio files.
43979 DPCM Sol : @tab X
43981 DPCM Xan : @tab X
43982 @tab Used in Origin's Wing Commander IV AVI files.
43983 DPCM Xilam DERF : @tab X
43985 DSD (Direct Stream Digital), least significant bit first : @tab X
43986 DSD (Direct Stream Digital), most significant bit first : @tab X
43987 DSD (Direct Stream Digital), least significant bit first, planar :
43988 @tab X
43989 DSD (Direct Stream Digital), most significant bit first, planar :
43990 @tab X
43991 DSP Group TrueSpeech : @tab X
43992 DST (Direct Stream Transfer) : @tab X
43993 DV audio : @tab X
43994 Enhanced AC-3 : X @tab X
43995 EVRC (Enhanced Variable Rate Codec) : @tab X
43996 FLAC (Free Lossless Audio Codec) : X @tab IX
43997 G.723.1 : X @tab X
43998 G.729 : @tab X
43999 GSM : E @tab X
44000 @tab encoding supported through external library libgsm
44001 GSM Microsoft variant : E @tab X
44003 @tab encoding supported through external library libgsm
44004 IAC (Indeo Audio Coder) : @tab X
44006 iLBC (Internet Low Bitrate Codec) : E @tab EX
44007 @tab encoding and decoding supported through external library libilbc
44008 IMC (Intel Music Coder) : @tab X
44010 Interplay ACM : @tab X
44011 MACE (Macintosh Audio Compression/Expansion) 3:1 : @tab X
44012 MACE (Macintosh Audio Compression/Expansion) 6:1 : @tab X
44013 MLP (Meridian Lossless Packing) : X @tab X
44014 @tab Used in DVD-Audio discs.
44015 Monkey's Audio : @tab X
44017 MP1 (MPEG audio layer 1) : @tab IX
44018 MP2 (MPEG audio layer 2) : IX @tab IX
44019 @tab encoding supported also through external library TwoLAME
44020 MP3 (MPEG audio layer 3) : E @tab IX
44022 @tab encoding supported through external library LAME, ADU MP3 and MP3onMP4 also supported
44023 MPEG-4 Audio Lossless Coding (ALS) : @tab X
44025 Musepack SV7 : @tab X
44026 Musepack SV8 : @tab X
44027 Nellymoser Asao : X @tab X
44028 On2 AVC (Audio for Video Codec) : @tab X
44029 Opus : E @tab X
44030 @tab encoding supported through external library libopus
44031 PCM A-law : X @tab X
44033 PCM mu-law : X @tab X
44034 PCM Archimedes VIDC : X @tab X
44035 PCM signed 8-bit planar : X @tab X
44036 PCM signed 16-bit big-endian planar : X @tab X
44037 PCM signed 16-bit little-endian planar : X @tab X
44038 PCM signed 24-bit little-endian planar : X @tab X
44039 PCM signed 32-bit little-endian planar : X @tab X
44040 PCM 32-bit floating point big-endian : X @tab X
44041 PCM 32-bit floating point little-endian : X @tab X
44042 PCM 64-bit floating point big-endian : X @tab X
44043 PCM 64-bit floating point little-endian : X @tab X
44044 PCM D-Cinema audio signed 24-bit : X @tab X
44045 PCM signed 8-bit : X @tab X
44046 PCM signed 16-bit big-endian : X @tab X
44047 PCM signed 16-bit little-endian : X @tab X
44048 PCM signed 24-bit big-endian : X @tab X
44049 PCM signed 24-bit little-endian : X @tab X
44050 PCM signed 32-bit big-endian : X @tab X
44051 PCM signed 32-bit little-endian : X @tab X
44052 PCM signed 16/20/24-bit big-endian in MPEG-TS : @tab X
44053 PCM unsigned 8-bit : X @tab X
44054 PCM unsigned 16-bit big-endian : X @tab X
44055 PCM unsigned 16-bit little-endian : X @tab X
44056 PCM unsigned 24-bit big-endian : X @tab X
44057 PCM unsigned 24-bit little-endian : X @tab X
44058 PCM unsigned 32-bit big-endian : X @tab X
44059 PCM unsigned 32-bit little-endian : X @tab X
44060 QCELP / PureVoice : @tab X
44061 QDesign Music Codec 1 : @tab X
44062 QDesign Music Codec 2 : @tab X
44063 @tab There are still some distortions.
44064 RealAudio 1.0 (14.4K) : X @tab X
44066 @tab Real 14400 bit/s codec
44067 RealAudio 2.0 (28.8K) : @tab X
44069 @tab Real 28800 bit/s codec
44070 RealAudio 3.0 (dnet) : IX @tab X
44072 @tab Real low bitrate AC-3 codec
44073 RealAudio Lossless : @tab X
44075 RealAudio SIPR / ACELP.NET : @tab X
44076 SBC (low-complexity subband codec) : X @tab X
44077 @tab Used in Bluetooth A2DP
44078 Shorten : @tab X
44080 Sierra VMD audio : @tab X
44081 @tab Used in Sierra VMD files.
44082 Smacker audio : @tab X
44084 SMPTE 302M AES3 audio : X @tab X
44085 Sonic : X @tab X
44086 @tab experimental codec
44087 Sonic lossless : X @tab X
44089 @tab experimental codec
44090 Speex : E @tab EX
44092 @tab supported through external library libspeex
44093 TAK (Tom's lossless Audio Kompressor) : @tab X
44095 True Audio (TTA) : X @tab X
44096 TrueHD : X @tab X
44097 @tab Used in HD-DVD and Blu-Ray discs.
44098 TwinVQ (VQF flavor) : @tab X
44100 VIMA : @tab X
44101 @tab Used in LucasArts SMUSH animations.
44102 Vorbis : E @tab X
44104 @tab A native but very primitive encoder exists.
44105 Voxware MetaSound : @tab X
44107 WavPack : X @tab X
44108 Westwood Audio (SND1) : @tab X
44109 Windows Media Audio 1 : X @tab X
44110 Windows Media Audio 2 : X @tab X
44111 Windows Media Audio Lossless : @tab X
44112 Windows Media Audio Pro : @tab X
44113 Windows Media Audio Voice : @tab X
44114 Xbox Media Audio 1 : @tab X
44115 Xbox Media Audio 2 : @tab X
44116 "X" means that the feature in that column (encoding / decoding) is
44118 supported.
44119 "E" means that support is provided through an external library.
44121 "I" means that an integer-only version is available, too (ensures high
44123 performance on systems without hardware floating point support).
44124 Subtitle Formats
44126 Name : Muxing @tab Demuxing @tab Encoding @tab Decoding
44127 3GPP Timed Text : @tab @tab X @tab X
44128 AQTitle : @tab X @tab @tab X
44129 DVB : X @tab X @tab X @tab X
44130 DVB teletext : @tab X @tab @tab E
44131 DVD : X @tab X @tab X @tab X
44132 JACOsub : X @tab X @tab @tab X
44133 MicroDVD : X @tab X @tab @tab X
44134 MPL2 : @tab X @tab @tab X
44135 MPsub (MPlayer) : @tab X @tab @tab X
44136 PGS : @tab @tab @tab X
44137 PJS (Phoenix) : @tab X @tab @tab X
44138 RealText : @tab X @tab @tab X
44139 SAMI : @tab X @tab @tab X
44140 Spruce format (STL) : @tab X @tab @tab X
44141 SSA/ASS : X @tab X @tab X @tab X
44142 SubRip (SRT) : X @tab X @tab X @tab X
44143 SubViewer v1 : @tab X @tab @tab X
44144 SubViewer : @tab X @tab @tab X
44145 TED Talks captions : @tab X @tab @tab X
44146 TTML : X @tab @tab X @tab
44147 VobSub (IDX+SUB) : @tab X @tab @tab X
44148 VPlayer : @tab X @tab @tab X
44149 WebVTT : X @tab X @tab X @tab X
44150 XSUB : @tab @tab X @tab X
44151 "X" means that the feature is supported.
44153 "E" means that support is provided through an external library.
44155 Network Protocols
44157 Name : Support
44158 AMQP : E
44159 file : X
44160 FTP : X
44161 Gopher : X
44162 Gophers : X
44163 HLS : X
44164 HTTP : X
44165 HTTPS : X
44166 Icecast : X
44167 MMSH : X
44168 MMST : X
44169 pipe : X
44170 Pro-MPEG FEC : X
44171 RTMP : X
44172 RTMPE : X
44173 RTMPS : X
44174 RTMPT : X
44175 RTMPTE : X
44176 RTMPTS : X
44177 RTP : X
44178 SAMBA : E
44179 SCTP : X
44180 SFTP : E
44181 TCP : X
44182 TLS : X
44183 UDP : X
44184 ZMQ : E
44185 "X" means that the protocol is supported.
44187 "E" means that support is provided through an external library.
44189 Input/Output Devices
44191 Name : Input @tab Output
44192 ALSA : X @tab X
44193 BKTR : X @tab
44194 caca : @tab X
44195 DV1394 : X @tab
44196 Lavfi virtual device : X @tab
44197 Linux framebuffer : X @tab X
44198 JACK : X @tab
44199 LIBCDIO : X
44200 LIBDC1394 : X @tab
44201 OpenAL : X
44202 OpenGL : @tab X
44203 OSS : X @tab X
44204 PulseAudio : X @tab X
44205 SDL : @tab X
44206 Video4Linux2 : X @tab X
44207 VfW capture : X @tab
44208 X11 grabbing : X @tab
44209 Win32 grabbing : X @tab
44210 "X" means that input/output is supported.
44212 Timecode
44214 Codec/format : Read @tab Write
44215 AVI : X @tab X
44216 DV : X @tab X
44217 GXF : X @tab X
44218 MOV : X @tab X
44219 MPEG1/2 : X @tab X
44220 MXF : X @tab X
44221
44223 ffmpeg(1), ffplay(1), ffprobe(1), ffmpeg-utils(1), ffmpeg-scaler(1),
44224 ffmpeg-resampler(1), ffmpeg-codecs(1), ffmpeg-bitstream-filters(1),
44225 ffmpeg-formats(1), ffmpeg-devices(1), ffmpeg-protocols(1),
44226 ffmpeg-filters(1)
44227
44229 The FFmpeg developers.
44230 For details about the authorship, see the Git history of the project
44232 (git://source.ffmpeg.org/ffmpeg), e.g. by typing the command git log in
44233 the FFmpeg source directory, or browsing the online repository at
44234 <http://source.ffmpeg.org>.
44235 Maintainers for the specific components are listed in the file
44237 MAINTAINERS in the source code tree.
44238 FFMPEG-ALL(1)
