1 @chapter Filtering Introduction
2 @c man begin FILTERING INTRODUCTION
4 Filtering in FFmpeg is enabled through the libavfilter library.
6 In libavfilter, a filter can have multiple inputs and multiple
8 To illustrate the sorts of things that are possible, we consider the
13 input --> split ---------------------> overlay --> output
16 +-----> crop --> vflip -------+
19 This filtergraph splits the input stream in two streams, then sends one
20 stream through the crop filter and the vflip filter, before merging it
21 back with the other stream by overlaying it on top. You can use the
22 following command to achieve this:
25 ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
28 The result will be that the top half of the video is mirrored
29 onto the bottom half of the output video.
31 Filters in the same linear chain are separated by commas, and distinct
32 linear chains of filters are separated by semicolons. In our example,
33 @var{crop,vflip} are in one linear chain, @var{split} and
34 @var{overlay} are separately in another. The points where the linear
35 chains join are labelled by names enclosed in square brackets. In the
36 example, the split filter generates two outputs that are associated to
37 the labels @var{[main]} and @var{[tmp]}.
39 The stream sent to the second output of @var{split}, labelled as
40 @var{[tmp]}, is processed through the @var{crop} filter, which crops
41 away the lower half part of the video, and then vertically flipped. The
42 @var{overlay} filter takes in input the first unchanged output of the
43 split filter (which was labelled as @var{[main]}), and overlay on its
44 lower half the output generated by the @var{crop,vflip} filterchain.
46 Some filters take in input a list of parameters: they are specified
47 after the filter name and an equal sign, and are separated from each other
50 There exist so-called @var{source filters} that do not have an
51 audio/video input, and @var{sink filters} that will not have audio/video
54 @c man end FILTERING INTRODUCTION
57 @c man begin GRAPH2DOT
59 The @file{graph2dot} program included in the FFmpeg @file{tools}
60 directory can be used to parse a filtergraph description and issue a
61 corresponding textual representation in the dot language.
68 to see how to use @file{graph2dot}.
70 You can then pass the dot description to the @file{dot} program (from
71 the graphviz suite of programs) and obtain a graphical representation
74 For example the sequence of commands:
76 echo @var{GRAPH_DESCRIPTION} | \
77 tools/graph2dot -o graph.tmp && \
78 dot -Tpng graph.tmp -o graph.png && \
82 can be used to create and display an image representing the graph
83 described by the @var{GRAPH_DESCRIPTION} string. Note that this string must be
84 a complete self-contained graph, with its inputs and outputs explicitly defined.
85 For example if your command line is of the form:
87 ffmpeg -i infile -vf scale=640:360 outfile
89 your @var{GRAPH_DESCRIPTION} string will need to be of the form:
91 nullsrc,scale=640:360,nullsink
93 you may also need to set the @var{nullsrc} parameters and add a @var{format}
94 filter in order to simulate a specific input file.
98 @chapter Filtergraph description
99 @c man begin FILTERGRAPH DESCRIPTION
101 A filtergraph is a directed graph of connected filters. It can contain
102 cycles, and there can be multiple links between a pair of
103 filters. Each link has one input pad on one side connecting it to one
104 filter from which it takes its input, and one output pad on the other
105 side connecting it to one filter accepting its output.
107 Each filter in a filtergraph is an instance of a filter class
108 registered in the application, which defines the features and the
109 number of input and output pads of the filter.
111 A filter with no input pads is called a "source", and a filter with no
112 output pads is called a "sink".
114 @anchor{Filtergraph syntax}
115 @section Filtergraph syntax
117 A filtergraph has a textual representation, which is recognized by the
118 @option{-filter}/@option{-vf}/@option{-af} and
119 @option{-filter_complex} options in @command{ffmpeg} and
120 @option{-vf}/@option{-af} in @command{ffplay}, and by the
121 @code{avfilter_graph_parse_ptr()} function defined in
122 @file{libavfilter/avfilter.h}.
124 A filterchain consists of a sequence of connected filters, each one
125 connected to the previous one in the sequence. A filterchain is
126 represented by a list of ","-separated filter descriptions.
128 A filtergraph consists of a sequence of filterchains. A sequence of
129 filterchains is represented by a list of ";"-separated filterchain
132 A filter is represented by a string of the form:
133 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
135 @var{filter_name} is the name of the filter class of which the
136 described filter is an instance of, and has to be the name of one of
137 the filter classes registered in the program.
138 The name of the filter class is optionally followed by a string
141 @var{arguments} is a string which contains the parameters used to
142 initialize the filter instance. It may have one of two forms:
146 A ':'-separated list of @var{key=value} pairs.
149 A ':'-separated list of @var{value}. In this case, the keys are assumed to be
150 the option names in the order they are declared. E.g. the @code{fade} filter
151 declares three options in this order -- @option{type}, @option{start_frame} and
152 @option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
153 @var{in} is assigned to the option @option{type}, @var{0} to
154 @option{start_frame} and @var{30} to @option{nb_frames}.
157 A ':'-separated list of mixed direct @var{value} and long @var{key=value}
158 pairs. The direct @var{value} must precede the @var{key=value} pairs, and
159 follow the same constraints order of the previous point. The following
160 @var{key=value} pairs can be set in any preferred order.
164 If the option value itself is a list of items (e.g. the @code{format} filter
165 takes a list of pixel formats), the items in the list are usually separated by
168 The list of arguments can be quoted using the character @samp{'} as initial
169 and ending mark, and the character @samp{\} for escaping the characters
170 within the quoted text; otherwise the argument string is considered
171 terminated when the next special character (belonging to the set
172 @samp{[]=;,}) is encountered.
174 The name and arguments of the filter are optionally preceded and
175 followed by a list of link labels.
176 A link label allows one to name a link and associate it to a filter output
177 or input pad. The preceding labels @var{in_link_1}
178 ... @var{in_link_N}, are associated to the filter input pads,
179 the following labels @var{out_link_1} ... @var{out_link_M}, are
180 associated to the output pads.
182 When two link labels with the same name are found in the
183 filtergraph, a link between the corresponding input and output pad is
186 If an output pad is not labelled, it is linked by default to the first
187 unlabelled input pad of the next filter in the filterchain.
188 For example in the filterchain
190 nullsrc, split[L1], [L2]overlay, nullsink
192 the split filter instance has two output pads, and the overlay filter
193 instance two input pads. The first output pad of split is labelled
194 "L1", the first input pad of overlay is labelled "L2", and the second
195 output pad of split is linked to the second input pad of overlay,
196 which are both unlabelled.
198 In a filter description, if the input label of the first filter is not
199 specified, "in" is assumed; if the output label of the last filter is not
200 specified, "out" is assumed.
202 In a complete filterchain all the unlabelled filter input and output
203 pads must be connected. A filtergraph is considered valid if all the
204 filter input and output pads of all the filterchains are connected.
206 Libavfilter will automatically insert @ref{scale} filters where format
207 conversion is required. It is possible to specify swscale flags
208 for those automatically inserted scalers by prepending
209 @code{sws_flags=@var{flags};}
210 to the filtergraph description.
212 Here is a BNF description of the filtergraph syntax:
214 @var{NAME} ::= sequence of alphanumeric characters and '_'
215 @var{LINKLABEL} ::= "[" @var{NAME} "]"
216 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
217 @var{FILTER_ARGUMENTS} ::= sequence of chars (possibly quoted)
218 @var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
219 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
220 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
223 @section Notes on filtergraph escaping
225 Filtergraph description composition entails several levels of
226 escaping. See @ref{quoting_and_escaping,,the "Quoting and escaping"
227 section in the ffmpeg-utils(1) manual,ffmpeg-utils} for more
228 information about the employed escaping procedure.
230 A first level escaping affects the content of each filter option
231 value, which may contain the special character @code{:} used to
232 separate values, or one of the escaping characters @code{\'}.
234 A second level escaping affects the whole filter description, which
235 may contain the escaping characters @code{\'} or the special
236 characters @code{[],;} used by the filtergraph description.
238 Finally, when you specify a filtergraph on a shell commandline, you
239 need to perform a third level escaping for the shell special
240 characters contained within it.
242 For example, consider the following string to be embedded in
243 the @ref{drawtext} filter description @option{text} value:
245 this is a 'string': may contain one, or more, special characters
248 This string contains the @code{'} special escaping character, and the
249 @code{:} special character, so it needs to be escaped in this way:
251 text=this is a \'string\'\: may contain one, or more, special characters
254 A second level of escaping is required when embedding the filter
255 description in a filtergraph description, in order to escape all the
256 filtergraph special characters. Thus the example above becomes:
258 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
260 (note that in addition to the @code{\'} escaping special characters,
261 also @code{,} needs to be escaped).
263 Finally an additional level of escaping is needed when writing the
264 filtergraph description in a shell command, which depends on the
265 escaping rules of the adopted shell. For example, assuming that
266 @code{\} is special and needs to be escaped with another @code{\}, the
267 previous string will finally result in:
269 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
272 @chapter Timeline editing
274 Some filters support a generic @option{enable} option. For the filters
275 supporting timeline editing, this option can be set to an expression which is
276 evaluated before sending a frame to the filter. If the evaluation is non-zero,
277 the filter will be enabled, otherwise the frame will be sent unchanged to the
278 next filter in the filtergraph.
280 The expression accepts the following values:
283 timestamp expressed in seconds, NAN if the input timestamp is unknown
286 sequential number of the input frame, starting from 0
289 the position in the file of the input frame, NAN if unknown
293 width and height of the input frame if video
296 Additionally, these filters support an @option{enable} command that can be used
297 to re-define the expression.
299 Like any other filtering option, the @option{enable} option follows the same
302 For example, to enable a blur filter (@ref{smartblur}) from 10 seconds to 3
303 minutes, and a @ref{curves} filter starting at 3 seconds:
305 smartblur = enable='between(t,10,3*60)',
306 curves = enable='gte(t,3)' : preset=cross_process
309 See @code{ffmpeg -filters} to view which filters have timeline support.
311 @c man end FILTERGRAPH DESCRIPTION
313 @chapter Audio Filters
314 @c man begin AUDIO FILTERS
316 When you configure your FFmpeg build, you can disable any of the
317 existing filters using @code{--disable-filters}.
318 The configure output will show the audio filters included in your
321 Below is a description of the currently available audio filters.
325 A compressor is mainly used to reduce the dynamic range of a signal.
326 Especially modern music is mostly compressed at a high ratio to
327 improve the overall loudness. It's done to get the highest attention
328 of a listener, "fatten" the sound and bring more "power" to the track.
329 If a signal is compressed too much it may sound dull or "dead"
330 afterwards or it may start to "pump" (which could be a powerful effect
331 but can also destroy a track completely).
332 The right compression is the key to reach a professional sound and is
333 the high art of mixing and mastering. Because of its complex settings
334 it may take a long time to get the right feeling for this kind of effect.
336 Compression is done by detecting the volume above a chosen level
337 @code{threshold} and dividing it by the factor set with @code{ratio}.
338 So if you set the threshold to -12dB and your signal reaches -6dB a ratio
339 of 2:1 will result in a signal at -9dB. Because an exact manipulation of
340 the signal would cause distortion of the waveform the reduction can be
341 levelled over the time. This is done by setting "Attack" and "Release".
342 @code{attack} determines how long the signal has to rise above the threshold
343 before any reduction will occur and @code{release} sets the time the signal
344 has to fall below the threshold to reduce the reduction again. Shorter signals
345 than the chosen attack time will be left untouched.
346 The overall reduction of the signal can be made up afterwards with the
347 @code{makeup} setting. So compressing the peaks of a signal about 6dB and
348 raising the makeup to this level results in a signal twice as loud than the
349 source. To gain a softer entry in the compression the @code{knee} flattens the
350 hard edge at the threshold in the range of the chosen decibels.
352 The filter accepts the following options:
356 Set input gain. Default is 1. Range is between 0.015625 and 64.
359 If a signal of second stream rises above this level it will affect the gain
360 reduction of the first stream.
361 By default it is 0.125. Range is between 0.00097563 and 1.
364 Set a ratio by which the signal is reduced. 1:2 means that if the level
365 rose 4dB above the threshold, it will be only 2dB above after the reduction.
366 Default is 2. Range is between 1 and 20.
369 Amount of milliseconds the signal has to rise above the threshold before gain
370 reduction starts. Default is 20. Range is between 0.01 and 2000.
373 Amount of milliseconds the signal has to fall below the threshold before
374 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
377 Set the amount by how much signal will be amplified after processing.
378 Default is 2. Range is from 1 and 64.
381 Curve the sharp knee around the threshold to enter gain reduction more softly.
382 Default is 2.82843. Range is between 1 and 8.
385 Choose if the @code{average} level between all channels of input stream
386 or the louder(@code{maximum}) channel of input stream affects the
387 reduction. Default is @code{average}.
390 Should the exact signal be taken in case of @code{peak} or an RMS one in case
391 of @code{rms}. Default is @code{rms} which is mostly smoother.
394 How much to use compressed signal in output. Default is 1.
395 Range is between 0 and 1.
400 Apply cross fade from one input audio stream to another input audio stream.
401 The cross fade is applied for specified duration near the end of first stream.
403 The filter accepts the following options:
407 Specify the number of samples for which the cross fade effect has to last.
408 At the end of the cross fade effect the first input audio will be completely
409 silent. Default is 44100.
412 Specify the duration of the cross fade effect. See
413 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
414 for the accepted syntax.
415 By default the duration is determined by @var{nb_samples}.
416 If set this option is used instead of @var{nb_samples}.
419 Should first stream end overlap with second stream start. Default is enabled.
422 Set curve for cross fade transition for first stream.
425 Set curve for cross fade transition for second stream.
427 For description of available curve types see @ref{afade} filter description.
434 Cross fade from one input to another:
436 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
440 Cross fade from one input to another but without overlapping:
442 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
448 Reduce audio bit resolution.
450 This filter is bit crusher with enhanced functionality. A bit crusher
451 is used to audibly reduce number of bits an audio signal is sampled
452 with. This doesn't change the bit depth at all, it just produces the
453 effect. Material reduced in bit depth sounds more harsh and "digital".
454 This filter is able to even round to continuous values instead of discrete
456 Additionally it has a D/C offset which results in different crushing of
457 the lower and the upper half of the signal.
458 An Anti-Aliasing setting is able to produce "softer" crushing sounds.
460 Another feature of this filter is the logarithmic mode.
461 This setting switches from linear distances between bits to logarithmic ones.
462 The result is a much more "natural" sounding crusher which doesn't gate low
463 signals for example. The human ear has a logarithmic perception, too
464 so this kind of crushing is much more pleasant.
465 Logarithmic crushing is also able to get anti-aliased.
467 The filter accepts the following options:
483 Can be linear: @code{lin} or logarithmic: @code{log}.
492 Set sample reduction.
495 Enable LFO. By default disabled.
506 Delay one or more audio channels.
508 Samples in delayed channel are filled with silence.
510 The filter accepts the following option:
514 Set list of delays in milliseconds for each channel separated by '|'.
515 At least one delay greater than 0 should be provided.
516 Unused delays will be silently ignored. If number of given delays is
517 smaller than number of channels all remaining channels will not be delayed.
518 If you want to delay exact number of samples, append 'S' to number.
525 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
526 the second channel (and any other channels that may be present) unchanged.
532 Delay second channel by 500 samples, the third channel by 700 samples and leave
533 the first channel (and any other channels that may be present) unchanged.
541 Apply echoing to the input audio.
543 Echoes are reflected sound and can occur naturally amongst mountains
544 (and sometimes large buildings) when talking or shouting; digital echo
545 effects emulate this behaviour and are often used to help fill out the
546 sound of a single instrument or vocal. The time difference between the
547 original signal and the reflection is the @code{delay}, and the
548 loudness of the reflected signal is the @code{decay}.
549 Multiple echoes can have different delays and decays.
551 A description of the accepted parameters follows.
555 Set input gain of reflected signal. Default is @code{0.6}.
558 Set output gain of reflected signal. Default is @code{0.3}.
561 Set list of time intervals in milliseconds between original signal and reflections
562 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
563 Default is @code{1000}.
566 Set list of loudnesses of reflected signals separated by '|'.
567 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
568 Default is @code{0.5}.
575 Make it sound as if there are twice as many instruments as are actually playing:
577 aecho=0.8:0.88:60:0.4
581 If delay is very short, then it sound like a (metallic) robot playing music:
587 A longer delay will sound like an open air concert in the mountains:
589 aecho=0.8:0.9:1000:0.3
593 Same as above but with one more mountain:
595 aecho=0.8:0.9:1000|1800:0.3|0.25
600 Audio emphasis filter creates or restores material directly taken from LPs or
601 emphased CDs with different filter curves. E.g. to store music on vinyl the
602 signal has to be altered by a filter first to even out the disadvantages of
603 this recording medium.
604 Once the material is played back the inverse filter has to be applied to
605 restore the distortion of the frequency response.
607 The filter accepts the following options:
617 Set filter mode. For restoring material use @code{reproduction} mode, otherwise
618 use @code{production} mode. Default is @code{reproduction} mode.
621 Set filter type. Selects medium. Can be one of the following:
633 select Compact Disc (CD).
639 select 50µs (FM-KF).
641 select 75µs (FM-KF).
647 Modify an audio signal according to the specified expressions.
649 This filter accepts one or more expressions (one for each channel),
650 which are evaluated and used to modify a corresponding audio signal.
652 It accepts the following parameters:
656 Set the '|'-separated expressions list for each separate channel. If
657 the number of input channels is greater than the number of
658 expressions, the last specified expression is used for the remaining
661 @item channel_layout, c
662 Set output channel layout. If not specified, the channel layout is
663 specified by the number of expressions. If set to @samp{same}, it will
664 use by default the same input channel layout.
667 Each expression in @var{exprs} can contain the following constants and functions:
671 channel number of the current expression
674 number of the evaluated sample, starting from 0
680 time of the evaluated sample expressed in seconds
683 @item nb_out_channels
684 input and output number of channels
687 the value of input channel with number @var{CH}
690 Note: this filter is slow. For faster processing you should use a
699 aeval=val(ch)/2:c=same
703 Invert phase of the second channel:
712 Apply fade-in/out effect to input audio.
714 A description of the accepted parameters follows.
718 Specify the effect type, can be either @code{in} for fade-in, or
719 @code{out} for a fade-out effect. Default is @code{in}.
721 @item start_sample, ss
722 Specify the number of the start sample for starting to apply the fade
723 effect. Default is 0.
726 Specify the number of samples for which the fade effect has to last. At
727 the end of the fade-in effect the output audio will have the same
728 volume as the input audio, at the end of the fade-out transition
729 the output audio will be silence. Default is 44100.
732 Specify the start time of the fade effect. Default is 0.
733 The value must be specified as a time duration; see
734 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
735 for the accepted syntax.
736 If set this option is used instead of @var{start_sample}.
739 Specify the duration of the fade effect. See
740 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
741 for the accepted syntax.
742 At the end of the fade-in effect the output audio will have the same
743 volume as the input audio, at the end of the fade-out transition
744 the output audio will be silence.
745 By default the duration is determined by @var{nb_samples}.
746 If set this option is used instead of @var{nb_samples}.
749 Set curve for fade transition.
751 It accepts the following values:
754 select triangular, linear slope (default)
756 select quarter of sine wave
758 select half of sine wave
760 select exponential sine wave
764 select inverted parabola
778 select inverted quarter of sine wave
780 select inverted half of sine wave
782 select double-exponential seat
784 select double-exponential sigmoid
792 Fade in first 15 seconds of audio:
798 Fade out last 25 seconds of a 900 seconds audio:
800 afade=t=out:st=875:d=25
805 Apply arbitrary expressions to samples in frequency domain.
809 Set frequency domain real expression for each separate channel separated
810 by '|'. Default is "1".
811 If the number of input channels is greater than the number of
812 expressions, the last specified expression is used for the remaining
816 Set frequency domain imaginary expression for each separate channel
817 separated by '|'. If not set, @var{real} option is used.
819 Each expression in @var{real} and @var{imag} can contain the following
827 current frequency bin number
830 number of available bins
833 channel number of the current expression
845 It accepts the following values:
861 Default is @code{w4096}
864 Set window function. Default is @code{hann}.
867 Set window overlap. If set to 1, the recommended overlap for selected
868 window function will be picked. Default is @code{0.75}.
875 Leave almost only low frequencies in audio:
877 afftfilt="1-clip((b/nb)*b,0,1)"
884 Set output format constraints for the input audio. The framework will
885 negotiate the most appropriate format to minimize conversions.
887 It accepts the following parameters:
891 A '|'-separated list of requested sample formats.
894 A '|'-separated list of requested sample rates.
896 @item channel_layouts
897 A '|'-separated list of requested channel layouts.
899 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
900 for the required syntax.
903 If a parameter is omitted, all values are allowed.
905 Force the output to either unsigned 8-bit or signed 16-bit stereo
907 aformat=sample_fmts=u8|s16:channel_layouts=stereo
912 A gate is mainly used to reduce lower parts of a signal. This kind of signal
913 processing reduces disturbing noise between useful signals.
915 Gating is done by detecting the volume below a chosen level @var{threshold}
916 and dividing it by the factor set with @var{ratio}. The bottom of the noise
917 floor is set via @var{range}. Because an exact manipulation of the signal
918 would cause distortion of the waveform the reduction can be levelled over
919 time. This is done by setting @var{attack} and @var{release}.
921 @var{attack} determines how long the signal has to fall below the threshold
922 before any reduction will occur and @var{release} sets the time the signal
923 has to rise above the threshold to reduce the reduction again.
924 Shorter signals than the chosen attack time will be left untouched.
928 Set input level before filtering.
929 Default is 1. Allowed range is from 0.015625 to 64.
932 Set the level of gain reduction when the signal is below the threshold.
933 Default is 0.06125. Allowed range is from 0 to 1.
936 If a signal rises above this level the gain reduction is released.
937 Default is 0.125. Allowed range is from 0 to 1.
940 Set a ratio by which the signal is reduced.
941 Default is 2. Allowed range is from 1 to 9000.
944 Amount of milliseconds the signal has to rise above the threshold before gain
946 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
949 Amount of milliseconds the signal has to fall below the threshold before the
950 reduction is increased again. Default is 250 milliseconds.
951 Allowed range is from 0.01 to 9000.
954 Set amount of amplification of signal after processing.
955 Default is 1. Allowed range is from 1 to 64.
958 Curve the sharp knee around the threshold to enter gain reduction more softly.
959 Default is 2.828427125. Allowed range is from 1 to 8.
962 Choose if exact signal should be taken for detection or an RMS like one.
963 Default is @code{rms}. Can be @code{peak} or @code{rms}.
966 Choose if the average level between all channels or the louder channel affects
968 Default is @code{average}. Can be @code{average} or @code{maximum}.
973 The limiter prevents an input signal from rising over a desired threshold.
974 This limiter uses lookahead technology to prevent your signal from distorting.
975 It means that there is a small delay after the signal is processed. Keep in mind
976 that the delay it produces is the attack time you set.
978 The filter accepts the following options:
982 Set input gain. Default is 1.
985 Set output gain. Default is 1.
988 Don't let signals above this level pass the limiter. Default is 1.
991 The limiter will reach its attenuation level in this amount of time in
992 milliseconds. Default is 5 milliseconds.
995 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
996 Default is 50 milliseconds.
999 When gain reduction is always needed ASC takes care of releasing to an
1000 average reduction level rather than reaching a reduction of 0 in the release
1004 Select how much the release time is affected by ASC, 0 means nearly no changes
1005 in release time while 1 produces higher release times.
1008 Auto level output signal. Default is enabled.
1009 This normalizes audio back to 0dB if enabled.
1012 Depending on picked setting it is recommended to upsample input 2x or 4x times
1013 with @ref{aresample} before applying this filter.
1017 Apply a two-pole all-pass filter with central frequency (in Hz)
1018 @var{frequency}, and filter-width @var{width}.
1019 An all-pass filter changes the audio's frequency to phase relationship
1020 without changing its frequency to amplitude relationship.
1022 The filter accepts the following options:
1026 Set frequency in Hz.
1029 Set method to specify band-width of filter.
1042 Specify the band-width of a filter in width_type units.
1049 The filter accepts the following options:
1053 Set the number of loops.
1056 Set maximal number of samples.
1059 Set first sample of loop.
1065 Merge two or more audio streams into a single multi-channel stream.
1067 The filter accepts the following options:
1072 Set the number of inputs. Default is 2.
1076 If the channel layouts of the inputs are disjoint, and therefore compatible,
1077 the channel layout of the output will be set accordingly and the channels
1078 will be reordered as necessary. If the channel layouts of the inputs are not
1079 disjoint, the output will have all the channels of the first input then all
1080 the channels of the second input, in that order, and the channel layout of
1081 the output will be the default value corresponding to the total number of
1084 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
1085 is FC+BL+BR, then the output will be in 5.1, with the channels in the
1086 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
1087 first input, b1 is the first channel of the second input).
1089 On the other hand, if both input are in stereo, the output channels will be
1090 in the default order: a1, a2, b1, b2, and the channel layout will be
1091 arbitrarily set to 4.0, which may or may not be the expected value.
1093 All inputs must have the same sample rate, and format.
1095 If inputs do not have the same duration, the output will stop with the
1098 @subsection Examples
1102 Merge two mono files into a stereo stream:
1104 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
1108 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
1110 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
1116 Mixes multiple audio inputs into a single output.
1118 Note that this filter only supports float samples (the @var{amerge}
1119 and @var{pan} audio filters support many formats). If the @var{amix}
1120 input has integer samples then @ref{aresample} will be automatically
1121 inserted to perform the conversion to float samples.
1125 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
1127 will mix 3 input audio streams to a single output with the same duration as the
1128 first input and a dropout transition time of 3 seconds.
1130 It accepts the following parameters:
1134 The number of inputs. If unspecified, it defaults to 2.
1137 How to determine the end-of-stream.
1141 The duration of the longest input. (default)
1144 The duration of the shortest input.
1147 The duration of the first input.
1151 @item dropout_transition
1152 The transition time, in seconds, for volume renormalization when an input
1153 stream ends. The default value is 2 seconds.
1157 @section anequalizer
1159 High-order parametric multiband equalizer for each channel.
1161 It accepts the following parameters:
1165 This option string is in format:
1166 "c@var{chn} f=@var{cf} w=@var{w} g=@var{g} t=@var{f} | ..."
1167 Each equalizer band is separated by '|'.
1171 Set channel number to which equalization will be applied.
1172 If input doesn't have that channel the entry is ignored.
1175 Set central frequency for band.
1176 If input doesn't have that frequency the entry is ignored.
1179 Set band width in hertz.
1182 Set band gain in dB.
1185 Set filter type for band, optional, can be:
1189 Butterworth, this is default.
1200 With this option activated frequency response of anequalizer is displayed
1204 Set video stream size. Only useful if curves option is activated.
1207 Set max gain that will be displayed. Only useful if curves option is activated.
1208 Setting this to a reasonable value makes it possible to display gain which is derived from
1209 neighbour bands which are too close to each other and thus produce higher gain
1210 when both are activated.
1213 Set frequency scale used to draw frequency response in video output.
1214 Can be linear or logarithmic. Default is logarithmic.
1217 Set color for each channel curve which is going to be displayed in video stream.
1218 This is list of color names separated by space or by '|'.
1219 Unrecognised or missing colors will be replaced by white color.
1222 @subsection Examples
1226 Lower gain by 10 of central frequency 200Hz and width 100 Hz
1227 for first 2 channels using Chebyshev type 1 filter:
1229 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1233 @subsection Commands
1235 This filter supports the following commands:
1238 Alter existing filter parameters.
1239 Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1241 @var{fN} is existing filter number, starting from 0, if no such filter is available
1243 @var{freq} set new frequency parameter.
1244 @var{width} set new width parameter in herz.
1245 @var{gain} set new gain parameter in dB.
1247 Full filter invocation with asendcmd may look like this:
1248 asendcmd=c='4.0 anequalizer change 0|f=200|w=50|g=1',anequalizer=...
1253 Pass the audio source unchanged to the output.
1257 Pad the end of an audio stream with silence.
1259 This can be used together with @command{ffmpeg} @option{-shortest} to
1260 extend audio streams to the same length as the video stream.
1262 A description of the accepted options follows.
1266 Set silence packet size. Default value is 4096.
1269 Set the number of samples of silence to add to the end. After the
1270 value is reached, the stream is terminated. This option is mutually
1271 exclusive with @option{whole_len}.
1274 Set the minimum total number of samples in the output audio stream. If
1275 the value is longer than the input audio length, silence is added to
1276 the end, until the value is reached. This option is mutually exclusive
1277 with @option{pad_len}.
1280 If neither the @option{pad_len} nor the @option{whole_len} option is
1281 set, the filter will add silence to the end of the input stream
1284 @subsection Examples
1288 Add 1024 samples of silence to the end of the input:
1294 Make sure the audio output will contain at least 10000 samples, pad
1295 the input with silence if required:
1297 apad=whole_len=10000
1301 Use @command{ffmpeg} to pad the audio input with silence, so that the
1302 video stream will always result the shortest and will be converted
1303 until the end in the output file when using the @option{shortest}
1306 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1311 Add a phasing effect to the input audio.
1313 A phaser filter creates series of peaks and troughs in the frequency spectrum.
1314 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1316 A description of the accepted parameters follows.
1320 Set input gain. Default is 0.4.
1323 Set output gain. Default is 0.74
1326 Set delay in milliseconds. Default is 3.0.
1329 Set decay. Default is 0.4.
1332 Set modulation speed in Hz. Default is 0.5.
1335 Set modulation type. Default is triangular.
1337 It accepts the following values:
1346 Audio pulsator is something between an autopanner and a tremolo.
1347 But it can produce funny stereo effects as well. Pulsator changes the volume
1348 of the left and right channel based on a LFO (low frequency oscillator) with
1349 different waveforms and shifted phases.
1350 This filter have the ability to define an offset between left and right
1351 channel. An offset of 0 means that both LFO shapes match each other.
1352 The left and right channel are altered equally - a conventional tremolo.
1353 An offset of 50% means that the shape of the right channel is exactly shifted
1354 in phase (or moved backwards about half of the frequency) - pulsator acts as
1355 an autopanner. At 1 both curves match again. Every setting in between moves the
1356 phase shift gapless between all stages and produces some "bypassing" sounds with
1357 sine and triangle waveforms. The more you set the offset near 1 (starting from
1358 the 0.5) the faster the signal passes from the left to the right speaker.
1360 The filter accepts the following options:
1364 Set input gain. By default it is 1. Range is [0.015625 - 64].
1367 Set output gain. By default it is 1. Range is [0.015625 - 64].
1370 Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1371 sawup or sawdown. Default is sine.
1374 Set modulation. Define how much of original signal is affected by the LFO.
1377 Set left channel offset. Default is 0. Allowed range is [0 - 1].
1380 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1383 Set pulse width. Default is 1. Allowed range is [0 - 2].
1386 Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1389 Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1393 Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1397 Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1398 if timing is set to hz.
1404 Resample the input audio to the specified parameters, using the
1405 libswresample library. If none are specified then the filter will
1406 automatically convert between its input and output.
1408 This filter is also able to stretch/squeeze the audio data to make it match
1409 the timestamps or to inject silence / cut out audio to make it match the
1410 timestamps, do a combination of both or do neither.
1412 The filter accepts the syntax
1413 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1414 expresses a sample rate and @var{resampler_options} is a list of
1415 @var{key}=@var{value} pairs, separated by ":". See the
1416 ffmpeg-resampler manual for the complete list of supported options.
1418 @subsection Examples
1422 Resample the input audio to 44100Hz:
1428 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1429 samples per second compensation:
1431 aresample=async=1000
1437 Reverse an audio clip.
1439 Warning: This filter requires memory to buffer the entire clip, so trimming
1442 @subsection Examples
1446 Take the first 5 seconds of a clip, and reverse it.
1448 atrim=end=5,areverse
1452 @section asetnsamples
1454 Set the number of samples per each output audio frame.
1456 The last output packet may contain a different number of samples, as
1457 the filter will flush all the remaining samples when the input audio
1460 The filter accepts the following options:
1464 @item nb_out_samples, n
1465 Set the number of frames per each output audio frame. The number is
1466 intended as the number of samples @emph{per each channel}.
1467 Default value is 1024.
1470 If set to 1, the filter will pad the last audio frame with zeroes, so
1471 that the last frame will contain the same number of samples as the
1472 previous ones. Default value is 1.
1475 For example, to set the number of per-frame samples to 1234 and
1476 disable padding for the last frame, use:
1478 asetnsamples=n=1234:p=0
1483 Set the sample rate without altering the PCM data.
1484 This will result in a change of speed and pitch.
1486 The filter accepts the following options:
1489 @item sample_rate, r
1490 Set the output sample rate. Default is 44100 Hz.
1495 Show a line containing various information for each input audio frame.
1496 The input audio is not modified.
1498 The shown line contains a sequence of key/value pairs of the form
1499 @var{key}:@var{value}.
1501 The following values are shown in the output:
1505 The (sequential) number of the input frame, starting from 0.
1508 The presentation timestamp of the input frame, in time base units; the time base
1509 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1512 The presentation timestamp of the input frame in seconds.
1515 position of the frame in the input stream, -1 if this information in
1516 unavailable and/or meaningless (for example in case of synthetic audio)
1525 The sample rate for the audio frame.
1528 The number of samples (per channel) in the frame.
1531 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1532 audio, the data is treated as if all the planes were concatenated.
1534 @item plane_checksums
1535 A list of Adler-32 checksums for each data plane.
1541 Display time domain statistical information about the audio channels.
1542 Statistics are calculated and displayed for each audio channel and,
1543 where applicable, an overall figure is also given.
1545 It accepts the following option:
1548 Short window length in seconds, used for peak and trough RMS measurement.
1549 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1553 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1554 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1557 Available keys for each channel are:
1588 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1589 this @code{lavfi.astats.Overall.Peak_count}.
1591 For description what each key means read below.
1594 Set number of frame after which stats are going to be recalculated.
1595 Default is disabled.
1598 A description of each shown parameter follows:
1602 Mean amplitude displacement from zero.
1605 Minimal sample level.
1608 Maximal sample level.
1610 @item Min difference
1611 Minimal difference between two consecutive samples.
1613 @item Max difference
1614 Maximal difference between two consecutive samples.
1616 @item Mean difference
1617 Mean difference between two consecutive samples.
1618 The average of each difference between two consecutive samples.
1622 Standard peak and RMS level measured in dBFS.
1626 Peak and trough values for RMS level measured over a short window.
1629 Standard ratio of peak to RMS level (note: not in dB).
1632 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1633 (i.e. either @var{Min level} or @var{Max level}).
1636 Number of occasions (not the number of samples) that the signal attained either
1637 @var{Min level} or @var{Max level}.
1640 Overall bit depth of audio. Number of bits used for each sample.
1645 Synchronize audio data with timestamps by squeezing/stretching it and/or
1646 dropping samples/adding silence when needed.
1648 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1650 It accepts the following parameters:
1654 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1655 by default. When disabled, time gaps are covered with silence.
1658 The minimum difference between timestamps and audio data (in seconds) to trigger
1659 adding/dropping samples. The default value is 0.1. If you get an imperfect
1660 sync with this filter, try setting this parameter to 0.
1663 The maximum compensation in samples per second. Only relevant with compensate=1.
1664 The default value is 500.
1667 Assume that the first PTS should be this value. The time base is 1 / sample
1668 rate. This allows for padding/trimming at the start of the stream. By default,
1669 no assumption is made about the first frame's expected PTS, so no padding or
1670 trimming is done. For example, this could be set to 0 to pad the beginning with
1671 silence if an audio stream starts after the video stream or to trim any samples
1672 with a negative PTS due to encoder delay.
1680 The filter accepts exactly one parameter, the audio tempo. If not
1681 specified then the filter will assume nominal 1.0 tempo. Tempo must
1682 be in the [0.5, 2.0] range.
1684 @subsection Examples
1688 Slow down audio to 80% tempo:
1694 To speed up audio to 125% tempo:
1702 Trim the input so that the output contains one continuous subpart of the input.
1704 It accepts the following parameters:
1707 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1708 sample with the timestamp @var{start} will be the first sample in the output.
1711 Specify time of the first audio sample that will be dropped, i.e. the
1712 audio sample immediately preceding the one with the timestamp @var{end} will be
1713 the last sample in the output.
1716 Same as @var{start}, except this option sets the start timestamp in samples
1720 Same as @var{end}, except this option sets the end timestamp in samples instead
1724 The maximum duration of the output in seconds.
1727 The number of the first sample that should be output.
1730 The number of the first sample that should be dropped.
1733 @option{start}, @option{end}, and @option{duration} are expressed as time
1734 duration specifications; see
1735 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1737 Note that the first two sets of the start/end options and the @option{duration}
1738 option look at the frame timestamp, while the _sample options simply count the
1739 samples that pass through the filter. So start/end_pts and start/end_sample will
1740 give different results when the timestamps are wrong, inexact or do not start at
1741 zero. Also note that this filter does not modify the timestamps. If you wish
1742 to have the output timestamps start at zero, insert the asetpts filter after the
1745 If multiple start or end options are set, this filter tries to be greedy and
1746 keep all samples that match at least one of the specified constraints. To keep
1747 only the part that matches all the constraints at once, chain multiple atrim
1750 The defaults are such that all the input is kept. So it is possible to set e.g.
1751 just the end values to keep everything before the specified time.
1756 Drop everything except the second minute of input:
1758 ffmpeg -i INPUT -af atrim=60:120
1762 Keep only the first 1000 samples:
1764 ffmpeg -i INPUT -af atrim=end_sample=1000
1771 Apply a two-pole Butterworth band-pass filter with central
1772 frequency @var{frequency}, and (3dB-point) band-width width.
1773 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1774 instead of the default: constant 0dB peak gain.
1775 The filter roll off at 6dB per octave (20dB per decade).
1777 The filter accepts the following options:
1781 Set the filter's central frequency. Default is @code{3000}.
1784 Constant skirt gain if set to 1. Defaults to 0.
1787 Set method to specify band-width of filter.
1800 Specify the band-width of a filter in width_type units.
1805 Apply a two-pole Butterworth band-reject filter with central
1806 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1807 The filter roll off at 6dB per octave (20dB per decade).
1809 The filter accepts the following options:
1813 Set the filter's central frequency. Default is @code{3000}.
1816 Set method to specify band-width of filter.
1829 Specify the band-width of a filter in width_type units.
1834 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1835 shelving filter with a response similar to that of a standard
1836 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1838 The filter accepts the following options:
1842 Give the gain at 0 Hz. Its useful range is about -20
1843 (for a large cut) to +20 (for a large boost).
1844 Beware of clipping when using a positive gain.
1847 Set the filter's central frequency and so can be used
1848 to extend or reduce the frequency range to be boosted or cut.
1849 The default value is @code{100} Hz.
1852 Set method to specify band-width of filter.
1865 Determine how steep is the filter's shelf transition.
1870 Apply a biquad IIR filter with the given coefficients.
1871 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1872 are the numerator and denominator coefficients respectively.
1875 Bauer stereo to binaural transformation, which improves headphone listening of
1876 stereo audio records.
1878 It accepts the following parameters:
1882 Pre-defined crossfeed level.
1886 Default level (fcut=700, feed=50).
1889 Chu Moy circuit (fcut=700, feed=60).
1892 Jan Meier circuit (fcut=650, feed=95).
1897 Cut frequency (in Hz).
1906 Remap input channels to new locations.
1908 It accepts the following parameters:
1911 Map channels from input to output. The argument is a '|'-separated list of
1912 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1913 @var{in_channel} form. @var{in_channel} can be either the name of the input
1914 channel (e.g. FL for front left) or its index in the input channel layout.
1915 @var{out_channel} is the name of the output channel or its index in the output
1916 channel layout. If @var{out_channel} is not given then it is implicitly an
1917 index, starting with zero and increasing by one for each mapping.
1919 @item channel_layout
1920 The channel layout of the output stream.
1923 If no mapping is present, the filter will implicitly map input channels to
1924 output channels, preserving indices.
1926 For example, assuming a 5.1+downmix input MOV file,
1928 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1930 will create an output WAV file tagged as stereo from the downmix channels of
1933 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1935 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1938 @section channelsplit
1940 Split each channel from an input audio stream into a separate output stream.
1942 It accepts the following parameters:
1944 @item channel_layout
1945 The channel layout of the input stream. The default is "stereo".
1948 For example, assuming a stereo input MP3 file,
1950 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1952 will create an output Matroska file with two audio streams, one containing only
1953 the left channel and the other the right channel.
1955 Split a 5.1 WAV file into per-channel files:
1957 ffmpeg -i in.wav -filter_complex
1958 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1959 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1960 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1965 Add a chorus effect to the audio.
1967 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1969 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1970 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1971 The modulation depth defines the range the modulated delay is played before or after
1972 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1973 sound tuned around the original one, like in a chorus where some vocals are slightly
1976 It accepts the following parameters:
1979 Set input gain. Default is 0.4.
1982 Set output gain. Default is 0.4.
1985 Set delays. A typical delay is around 40ms to 60ms.
1997 @subsection Examples
2003 chorus=0.7:0.9:55:0.4:0.25:2
2009 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
2013 Fuller sounding chorus with three delays:
2015 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
2020 Compress or expand the audio's dynamic range.
2022 It accepts the following parameters:
2028 A list of times in seconds for each channel over which the instantaneous level
2029 of the input signal is averaged to determine its volume. @var{attacks} refers to
2030 increase of volume and @var{decays} refers to decrease of volume. For most
2031 situations, the attack time (response to the audio getting louder) should be
2032 shorter than the decay time, because the human ear is more sensitive to sudden
2033 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
2034 a typical value for decay is 0.8 seconds.
2035 If specified number of attacks & decays is lower than number of channels, the last
2036 set attack/decay will be used for all remaining channels.
2039 A list of points for the transfer function, specified in dB relative to the
2040 maximum possible signal amplitude. Each key points list must be defined using
2041 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
2042 @code{x0/y0 x1/y1 x2/y2 ....}
2044 The input values must be in strictly increasing order but the transfer function
2045 does not have to be monotonically rising. The point @code{0/0} is assumed but
2046 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
2047 function are @code{-70/-70|-60/-20}.
2050 Set the curve radius in dB for all joints. It defaults to 0.01.
2053 Set the additional gain in dB to be applied at all points on the transfer
2054 function. This allows for easy adjustment of the overall gain.
2058 Set an initial volume, in dB, to be assumed for each channel when filtering
2059 starts. This permits the user to supply a nominal level initially, so that, for
2060 example, a very large gain is not applied to initial signal levels before the
2061 companding has begun to operate. A typical value for audio which is initially
2062 quiet is -90 dB. It defaults to 0.
2065 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
2066 delayed before being fed to the volume adjuster. Specifying a delay
2067 approximately equal to the attack/decay times allows the filter to effectively
2068 operate in predictive rather than reactive mode. It defaults to 0.
2072 @subsection Examples
2076 Make music with both quiet and loud passages suitable for listening to in a
2079 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
2082 Another example for audio with whisper and explosion parts:
2084 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
2088 A noise gate for when the noise is at a lower level than the signal:
2090 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
2094 Here is another noise gate, this time for when the noise is at a higher level
2095 than the signal (making it, in some ways, similar to squelch):
2097 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
2101 2:1 compression starting at -6dB:
2103 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
2107 2:1 compression starting at -9dB:
2109 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
2113 2:1 compression starting at -12dB:
2115 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
2119 2:1 compression starting at -18dB:
2121 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
2125 3:1 compression starting at -15dB:
2127 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
2133 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
2139 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
2143 Hard limiter at -6dB:
2145 compand=attacks=0:points=-80/-80|-6/-6|20/-6
2149 Hard limiter at -12dB:
2151 compand=attacks=0:points=-80/-80|-12/-12|20/-12
2155 Hard noise gate at -35 dB:
2157 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
2163 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
2167 @section compensationdelay
2169 Compensation Delay Line is a metric based delay to compensate differing
2170 positions of microphones or speakers.
2172 For example, you have recorded guitar with two microphones placed in
2173 different location. Because the front of sound wave has fixed speed in
2174 normal conditions, the phasing of microphones can vary and depends on
2175 their location and interposition. The best sound mix can be achieved when
2176 these microphones are in phase (synchronized). Note that distance of
2177 ~30 cm between microphones makes one microphone to capture signal in
2178 antiphase to another microphone. That makes the final mix sounding moody.
2179 This filter helps to solve phasing problems by adding different delays
2180 to each microphone track and make them synchronized.
2182 The best result can be reached when you take one track as base and
2183 synchronize other tracks one by one with it.
2184 Remember that synchronization/delay tolerance depends on sample rate, too.
2185 Higher sample rates will give more tolerance.
2187 It accepts the following parameters:
2191 Set millimeters distance. This is compensation distance for fine tuning.
2195 Set cm distance. This is compensation distance for tightening distance setup.
2199 Set meters distance. This is compensation distance for hard distance setup.
2203 Set dry amount. Amount of unprocessed (dry) signal.
2207 Set wet amount. Amount of processed (wet) signal.
2211 Set temperature degree in Celsius. This is the temperature of the environment.
2215 @section crystalizer
2216 Simple algorithm to expand audio dynamic range.
2218 The filter accepts the following options:
2222 Sets the intensity of effect (default: 2.0). Must be in range between 0.0
2223 (unchanged sound) to 10.0 (maximum effect).
2226 Enable clipping. By default is enabled.
2230 Apply a DC shift to the audio.
2232 This can be useful to remove a DC offset (caused perhaps by a hardware problem
2233 in the recording chain) from the audio. The effect of a DC offset is reduced
2234 headroom and hence volume. The @ref{astats} filter can be used to determine if
2235 a signal has a DC offset.
2239 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
2243 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
2244 used to prevent clipping.
2248 Dynamic Audio Normalizer.
2250 This filter applies a certain amount of gain to the input audio in order
2251 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
2252 contrast to more "simple" normalization algorithms, the Dynamic Audio
2253 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
2254 This allows for applying extra gain to the "quiet" sections of the audio
2255 while avoiding distortions or clipping the "loud" sections. In other words:
2256 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
2257 sections, in the sense that the volume of each section is brought to the
2258 same target level. Note, however, that the Dynamic Audio Normalizer achieves
2259 this goal *without* applying "dynamic range compressing". It will retain 100%
2260 of the dynamic range *within* each section of the audio file.
2264 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
2265 Default is 500 milliseconds.
2266 The Dynamic Audio Normalizer processes the input audio in small chunks,
2267 referred to as frames. This is required, because a peak magnitude has no
2268 meaning for just a single sample value. Instead, we need to determine the
2269 peak magnitude for a contiguous sequence of sample values. While a "standard"
2270 normalizer would simply use the peak magnitude of the complete file, the
2271 Dynamic Audio Normalizer determines the peak magnitude individually for each
2272 frame. The length of a frame is specified in milliseconds. By default, the
2273 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
2274 been found to give good results with most files.
2275 Note that the exact frame length, in number of samples, will be determined
2276 automatically, based on the sampling rate of the individual input audio file.
2279 Set the Gaussian filter window size. In range from 3 to 301, must be odd
2280 number. Default is 31.
2281 Probably the most important parameter of the Dynamic Audio Normalizer is the
2282 @code{window size} of the Gaussian smoothing filter. The filter's window size
2283 is specified in frames, centered around the current frame. For the sake of
2284 simplicity, this must be an odd number. Consequently, the default value of 31
2285 takes into account the current frame, as well as the 15 preceding frames and
2286 the 15 subsequent frames. Using a larger window results in a stronger
2287 smoothing effect and thus in less gain variation, i.e. slower gain
2288 adaptation. Conversely, using a smaller window results in a weaker smoothing
2289 effect and thus in more gain variation, i.e. faster gain adaptation.
2290 In other words, the more you increase this value, the more the Dynamic Audio
2291 Normalizer will behave like a "traditional" normalization filter. On the
2292 contrary, the more you decrease this value, the more the Dynamic Audio
2293 Normalizer will behave like a dynamic range compressor.
2296 Set the target peak value. This specifies the highest permissible magnitude
2297 level for the normalized audio input. This filter will try to approach the
2298 target peak magnitude as closely as possible, but at the same time it also
2299 makes sure that the normalized signal will never exceed the peak magnitude.
2300 A frame's maximum local gain factor is imposed directly by the target peak
2301 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
2302 It is not recommended to go above this value.
2305 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
2306 The Dynamic Audio Normalizer determines the maximum possible (local) gain
2307 factor for each input frame, i.e. the maximum gain factor that does not
2308 result in clipping or distortion. The maximum gain factor is determined by
2309 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
2310 additionally bounds the frame's maximum gain factor by a predetermined
2311 (global) maximum gain factor. This is done in order to avoid excessive gain
2312 factors in "silent" or almost silent frames. By default, the maximum gain
2313 factor is 10.0, For most inputs the default value should be sufficient and
2314 it usually is not recommended to increase this value. Though, for input
2315 with an extremely low overall volume level, it may be necessary to allow even
2316 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
2317 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
2318 Instead, a "sigmoid" threshold function will be applied. This way, the
2319 gain factors will smoothly approach the threshold value, but never exceed that
2323 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
2324 By default, the Dynamic Audio Normalizer performs "peak" normalization.
2325 This means that the maximum local gain factor for each frame is defined
2326 (only) by the frame's highest magnitude sample. This way, the samples can
2327 be amplified as much as possible without exceeding the maximum signal
2328 level, i.e. without clipping. Optionally, however, the Dynamic Audio
2329 Normalizer can also take into account the frame's root mean square,
2330 abbreviated RMS. In electrical engineering, the RMS is commonly used to
2331 determine the power of a time-varying signal. It is therefore considered
2332 that the RMS is a better approximation of the "perceived loudness" than
2333 just looking at the signal's peak magnitude. Consequently, by adjusting all
2334 frames to a constant RMS value, a uniform "perceived loudness" can be
2335 established. If a target RMS value has been specified, a frame's local gain
2336 factor is defined as the factor that would result in exactly that RMS value.
2337 Note, however, that the maximum local gain factor is still restricted by the
2338 frame's highest magnitude sample, in order to prevent clipping.
2341 Enable channels coupling. By default is enabled.
2342 By default, the Dynamic Audio Normalizer will amplify all channels by the same
2343 amount. This means the same gain factor will be applied to all channels, i.e.
2344 the maximum possible gain factor is determined by the "loudest" channel.
2345 However, in some recordings, it may happen that the volume of the different
2346 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
2347 In this case, this option can be used to disable the channel coupling. This way,
2348 the gain factor will be determined independently for each channel, depending
2349 only on the individual channel's highest magnitude sample. This allows for
2350 harmonizing the volume of the different channels.
2353 Enable DC bias correction. By default is disabled.
2354 An audio signal (in the time domain) is a sequence of sample values.
2355 In the Dynamic Audio Normalizer these sample values are represented in the
2356 -1.0 to 1.0 range, regardless of the original input format. Normally, the
2357 audio signal, or "waveform", should be centered around the zero point.
2358 That means if we calculate the mean value of all samples in a file, or in a
2359 single frame, then the result should be 0.0 or at least very close to that
2360 value. If, however, there is a significant deviation of the mean value from
2361 0.0, in either positive or negative direction, this is referred to as a
2362 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
2363 Audio Normalizer provides optional DC bias correction.
2364 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
2365 the mean value, or "DC correction" offset, of each input frame and subtract
2366 that value from all of the frame's sample values which ensures those samples
2367 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
2368 boundaries, the DC correction offset values will be interpolated smoothly
2369 between neighbouring frames.
2372 Enable alternative boundary mode. By default is disabled.
2373 The Dynamic Audio Normalizer takes into account a certain neighbourhood
2374 around each frame. This includes the preceding frames as well as the
2375 subsequent frames. However, for the "boundary" frames, located at the very
2376 beginning and at the very end of the audio file, not all neighbouring
2377 frames are available. In particular, for the first few frames in the audio
2378 file, the preceding frames are not known. And, similarly, for the last few
2379 frames in the audio file, the subsequent frames are not known. Thus, the
2380 question arises which gain factors should be assumed for the missing frames
2381 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
2382 to deal with this situation. The default boundary mode assumes a gain factor
2383 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
2384 "fade out" at the beginning and at the end of the input, respectively.
2387 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
2388 By default, the Dynamic Audio Normalizer does not apply "traditional"
2389 compression. This means that signal peaks will not be pruned and thus the
2390 full dynamic range will be retained within each local neighbourhood. However,
2391 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
2392 normalization algorithm with a more "traditional" compression.
2393 For this purpose, the Dynamic Audio Normalizer provides an optional compression
2394 (thresholding) function. If (and only if) the compression feature is enabled,
2395 all input frames will be processed by a soft knee thresholding function prior
2396 to the actual normalization process. Put simply, the thresholding function is
2397 going to prune all samples whose magnitude exceeds a certain threshold value.
2398 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
2399 value. Instead, the threshold value will be adjusted for each individual
2401 In general, smaller parameters result in stronger compression, and vice versa.
2402 Values below 3.0 are not recommended, because audible distortion may appear.
2407 Make audio easier to listen to on headphones.
2409 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
2410 so that when listened to on headphones the stereo image is moved from
2411 inside your head (standard for headphones) to outside and in front of
2412 the listener (standard for speakers).
2418 Apply a two-pole peaking equalisation (EQ) filter. With this
2419 filter, the signal-level at and around a selected frequency can
2420 be increased or decreased, whilst (unlike bandpass and bandreject
2421 filters) that at all other frequencies is unchanged.
2423 In order to produce complex equalisation curves, this filter can
2424 be given several times, each with a different central frequency.
2426 The filter accepts the following options:
2430 Set the filter's central frequency in Hz.
2433 Set method to specify band-width of filter.
2446 Specify the band-width of a filter in width_type units.
2449 Set the required gain or attenuation in dB.
2450 Beware of clipping when using a positive gain.
2453 @subsection Examples
2456 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2458 equalizer=f=1000:width_type=h:width=200:g=-10
2462 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2464 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2468 @section extrastereo
2470 Linearly increases the difference between left and right channels which
2471 adds some sort of "live" effect to playback.
2473 The filter accepts the following options:
2477 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2478 (average of both channels), with 1.0 sound will be unchanged, with
2479 -1.0 left and right channels will be swapped.
2482 Enable clipping. By default is enabled.
2485 @section firequalizer
2486 Apply FIR Equalization using arbitrary frequency response.
2488 The filter accepts the following option:
2492 Set gain curve equation (in dB). The expression can contain variables:
2495 the evaluated frequency
2499 channel number, set to 0 when multichannels evaluation is disabled
2501 channel id, see libavutil/channel_layout.h, set to the first channel id when
2502 multichannels evaluation is disabled
2506 channel_layout, see libavutil/channel_layout.h
2511 @item gain_interpolate(f)
2512 interpolate gain on frequency f based on gain_entry
2513 @item cubic_interpolate(f)
2514 same as gain_interpolate, but smoother
2516 This option is also available as command. Default is @code{gain_interpolate(f)}.
2519 Set gain entry for gain_interpolate function. The expression can
2523 store gain entry at frequency f with value g
2525 This option is also available as command.
2528 Set filter delay in seconds. Higher value means more accurate.
2529 Default is @code{0.01}.
2532 Set filter accuracy in Hz. Lower value means more accurate.
2533 Default is @code{5}.
2536 Set window function. Acceptable values are:
2539 rectangular window, useful when gain curve is already smooth
2541 hann window (default)
2547 3-terms continuous 1st derivative nuttall window
2549 minimum 3-terms discontinuous nuttall window
2551 4-terms continuous 1st derivative nuttall window
2553 minimum 4-terms discontinuous nuttall (blackman-nuttall) window
2555 blackman-harris window
2561 If enabled, use fixed number of audio samples. This improves speed when
2562 filtering with large delay. Default is disabled.
2565 Enable multichannels evaluation on gain. Default is disabled.
2568 Enable zero phase mode by subtracting timestamp to compensate delay.
2569 Default is disabled.
2572 Set scale used by gain. Acceptable values are:
2575 linear frequency, linear gain
2577 linear frequency, logarithmic (in dB) gain (default)
2579 logarithmic (in octave scale where 20 Hz is 0) frequency, linear gain
2581 logarithmic frequency, logarithmic gain
2585 Set file for dumping, suitable for gnuplot.
2588 Set scale for dumpfile. Acceptable values are same with scale option.
2592 Enable 2-channel convolution using complex FFT. This improves speed significantly.
2593 Default is disabled.
2596 @subsection Examples
2601 firequalizer=gain='if(lt(f,1000), 0, -INF)'
2604 lowpass at 1000 Hz with gain_entry:
2606 firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
2609 custom equalization:
2611 firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
2614 higher delay with zero phase to compensate delay:
2616 firequalizer=delay=0.1:fixed=on:zero_phase=on
2619 lowpass on left channel, highpass on right channel:
2621 firequalizer=gain='if(eq(chid,1), gain_interpolate(f), if(eq(chid,2), gain_interpolate(1e6+f), 0))'
2622 :gain_entry='entry(1000, 0); entry(1001,-INF); entry(1e6+1000,0)':multi=on
2627 Apply a flanging effect to the audio.
2629 The filter accepts the following options:
2633 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2636 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2639 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2643 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2644 Default value is 71.
2647 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2650 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2651 Default value is @var{sinusoidal}.
2654 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2655 Default value is 25.
2658 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2659 Default is @var{linear}.
2664 Decodes High Definition Compatible Digital (HDCD) data. A 16-bit PCM stream with
2665 embedded HDCD codes is expanded into a 20-bit PCM stream.
2667 The filter supports the Peak Extend and Low-level Gain Adjustment features
2668 of HDCD, and detects the Transient Filter flag.
2671 ffmpeg -i HDCD16.flac -af hdcd OUT24.flac
2674 When using the filter with wav, note the default encoding for wav is 16-bit,
2675 so the resulting 20-bit stream will be truncated back to 16-bit. Use something
2676 like @command{-acodec pcm_s24le} after the filter to get 24-bit PCM output.
2678 ffmpeg -i HDCD16.wav -af hdcd OUT16.wav
2679 ffmpeg -i HDCD16.wav -af hdcd -acodec pcm_s24le OUT24.wav
2682 The filter accepts the following options:
2685 @item disable_autoconvert
2686 Disable any automatic format conversion or resampling in the filter graph.
2688 @item process_stereo
2689 Process the stereo channels together. If target_gain does not match between
2690 channels, consider it invalid and use the last valid target_gain.
2693 Set the code detect timer period in ms.
2696 Always extend peaks above -3dBFS even if PE isn't signaled.
2699 Replace audio with a solid tone and adjust the amplitude to signal some
2700 specific aspect of the decoding process. The output file can be loaded in
2701 an audio editor alongside the original to aid analysis.
2703 @code{analyze_mode=pe:force_pe=true} can be used to see all samples above the PE level.
2710 Gain adjustment level at each sample
2712 Samples where peak extend occurs
2714 Samples where the code detect timer is active
2716 Samples where the target gain does not match between channels
2722 Apply a high-pass filter with 3dB point frequency.
2723 The filter can be either single-pole, or double-pole (the default).
2724 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2726 The filter accepts the following options:
2730 Set frequency in Hz. Default is 3000.
2733 Set number of poles. Default is 2.
2736 Set method to specify band-width of filter.
2749 Specify the band-width of a filter in width_type units.
2750 Applies only to double-pole filter.
2751 The default is 0.707q and gives a Butterworth response.
2756 Join multiple input streams into one multi-channel stream.
2758 It accepts the following parameters:
2762 The number of input streams. It defaults to 2.
2764 @item channel_layout
2765 The desired output channel layout. It defaults to stereo.
2768 Map channels from inputs to output. The argument is a '|'-separated list of
2769 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2770 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2771 can be either the name of the input channel (e.g. FL for front left) or its
2772 index in the specified input stream. @var{out_channel} is the name of the output
2776 The filter will attempt to guess the mappings when they are not specified
2777 explicitly. It does so by first trying to find an unused matching input channel
2778 and if that fails it picks the first unused input channel.
2780 Join 3 inputs (with properly set channel layouts):
2782 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2785 Build a 5.1 output from 6 single-channel streams:
2787 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2788 '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'
2794 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2796 To enable compilation of this filter you need to configure FFmpeg with
2797 @code{--enable-ladspa}.
2801 Specifies the name of LADSPA plugin library to load. If the environment
2802 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2803 each one of the directories specified by the colon separated list in
2804 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2805 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2806 @file{/usr/lib/ladspa/}.
2809 Specifies the plugin within the library. Some libraries contain only
2810 one plugin, but others contain many of them. If this is not set filter
2811 will list all available plugins within the specified library.
2814 Set the '|' separated list of controls which are zero or more floating point
2815 values that determine the behavior of the loaded plugin (for example delay,
2817 Controls need to be defined using the following syntax:
2818 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2819 @var{valuei} is the value set on the @var{i}-th control.
2820 Alternatively they can be also defined using the following syntax:
2821 @var{value0}|@var{value1}|@var{value2}|..., where
2822 @var{valuei} is the value set on the @var{i}-th control.
2823 If @option{controls} is set to @code{help}, all available controls and
2824 their valid ranges are printed.
2826 @item sample_rate, s
2827 Specify the sample rate, default to 44100. Only used if plugin have
2831 Set the number of samples per channel per each output frame, default
2832 is 1024. Only used if plugin have zero inputs.
2835 Set the minimum duration of the sourced audio. See
2836 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2837 for the accepted syntax.
2838 Note that the resulting duration may be greater than the specified duration,
2839 as the generated audio is always cut at the end of a complete frame.
2840 If not specified, or the expressed duration is negative, the audio is
2841 supposed to be generated forever.
2842 Only used if plugin have zero inputs.
2846 @subsection Examples
2850 List all available plugins within amp (LADSPA example plugin) library:
2856 List all available controls and their valid ranges for @code{vcf_notch}
2857 plugin from @code{VCF} library:
2859 ladspa=f=vcf:p=vcf_notch:c=help
2863 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2866 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2870 Add reverberation to the audio using TAP-plugins
2871 (Tom's Audio Processing plugins):
2873 ladspa=file=tap_reverb:tap_reverb
2877 Generate white noise, with 0.2 amplitude:
2879 ladspa=file=cmt:noise_source_white:c=c0=.2
2883 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2884 @code{C* Audio Plugin Suite} (CAPS) library:
2886 ladspa=file=caps:Click:c=c1=20'
2890 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2892 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2896 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2897 @code{SWH Plugins} collection:
2899 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2903 Attenuate low frequencies using Multiband EQ from Steve Harris
2904 @code{SWH Plugins} collection:
2906 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2910 @subsection Commands
2912 This filter supports the following commands:
2915 Modify the @var{N}-th control value.
2917 If the specified value is not valid, it is ignored and prior one is kept.
2922 EBU R128 loudness normalization. Includes both dynamic and linear normalization modes.
2923 Support for both single pass (livestreams, files) and double pass (files) modes.
2924 This algorithm can target IL, LRA, and maximum true peak.
2926 The filter accepts the following options:
2930 Set integrated loudness target.
2931 Range is -70.0 - -5.0. Default value is -24.0.
2934 Set loudness range target.
2935 Range is 1.0 - 20.0. Default value is 7.0.
2938 Set maximum true peak.
2939 Range is -9.0 - +0.0. Default value is -2.0.
2941 @item measured_I, measured_i
2942 Measured IL of input file.
2943 Range is -99.0 - +0.0.
2945 @item measured_LRA, measured_lra
2946 Measured LRA of input file.
2947 Range is 0.0 - 99.0.
2949 @item measured_TP, measured_tp
2950 Measured true peak of input file.
2951 Range is -99.0 - +99.0.
2953 @item measured_thresh
2954 Measured threshold of input file.
2955 Range is -99.0 - +0.0.
2958 Set offset gain. Gain is applied before the true-peak limiter.
2959 Range is -99.0 - +99.0. Default is +0.0.
2962 Normalize linearly if possible.
2963 measured_I, measured_LRA, measured_TP, and measured_thresh must also
2964 to be specified in order to use this mode.
2965 Options are true or false. Default is true.
2968 Treat mono input files as "dual-mono". If a mono file is intended for playback
2969 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
2970 If set to @code{true}, this option will compensate for this effect.
2971 Multi-channel input files are not affected by this option.
2972 Options are true or false. Default is false.
2975 Set print format for stats. Options are summary, json, or none.
2976 Default value is none.
2981 Apply a low-pass filter with 3dB point frequency.
2982 The filter can be either single-pole or double-pole (the default).
2983 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2985 The filter accepts the following options:
2989 Set frequency in Hz. Default is 500.
2992 Set number of poles. Default is 2.
2995 Set method to specify band-width of filter.
3008 Specify the band-width of a filter in width_type units.
3009 Applies only to double-pole filter.
3010 The default is 0.707q and gives a Butterworth response.
3016 Mix channels with specific gain levels. The filter accepts the output
3017 channel layout followed by a set of channels definitions.
3019 This filter is also designed to efficiently remap the channels of an audio
3022 The filter accepts parameters of the form:
3023 "@var{l}|@var{outdef}|@var{outdef}|..."
3027 output channel layout or number of channels
3030 output channel specification, of the form:
3031 "@var{out_name}=[@var{gain}*]@var{in_name}[(+-)[@var{gain}*]@var{in_name}...]"
3034 output channel to define, either a channel name (FL, FR, etc.) or a channel
3035 number (c0, c1, etc.)
3038 multiplicative coefficient for the channel, 1 leaving the volume unchanged
3041 input channel to use, see out_name for details; it is not possible to mix
3042 named and numbered input channels
3045 If the `=' in a channel specification is replaced by `<', then the gains for
3046 that specification will be renormalized so that the total is 1, thus
3047 avoiding clipping noise.
3049 @subsection Mixing examples
3051 For example, if you want to down-mix from stereo to mono, but with a bigger
3052 factor for the left channel:
3054 pan=1c|c0=0.9*c0+0.1*c1
3057 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
3058 7-channels surround:
3060 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
3063 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
3064 that should be preferred (see "-ac" option) unless you have very specific
3067 @subsection Remapping examples
3069 The channel remapping will be effective if, and only if:
3072 @item gain coefficients are zeroes or ones,
3073 @item only one input per channel output,
3076 If all these conditions are satisfied, the filter will notify the user ("Pure
3077 channel mapping detected"), and use an optimized and lossless method to do the
3080 For example, if you have a 5.1 source and want a stereo audio stream by
3081 dropping the extra channels:
3083 pan="stereo| c0=FL | c1=FR"
3086 Given the same source, you can also switch front left and front right channels
3087 and keep the input channel layout:
3089 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
3092 If the input is a stereo audio stream, you can mute the front left channel (and
3093 still keep the stereo channel layout) with:
3098 Still with a stereo audio stream input, you can copy the right channel in both
3099 front left and right:
3101 pan="stereo| c0=FR | c1=FR"
3106 ReplayGain scanner filter. This filter takes an audio stream as an input and
3107 outputs it unchanged.
3108 At end of filtering it displays @code{track_gain} and @code{track_peak}.
3112 Convert the audio sample format, sample rate and channel layout. It is
3113 not meant to be used directly.
3116 Apply time-stretching and pitch-shifting with librubberband.
3118 The filter accepts the following options:
3122 Set tempo scale factor.
3125 Set pitch scale factor.
3128 Set transients detector.
3129 Possible values are:
3138 Possible values are:
3147 Possible values are:
3154 Set processing window size.
3155 Possible values are:
3164 Possible values are:
3171 Enable formant preservation when shift pitching.
3172 Possible values are:
3180 Possible values are:
3189 Possible values are:
3196 @section sidechaincompress
3198 This filter acts like normal compressor but has the ability to compress
3199 detected signal using second input signal.
3200 It needs two input streams and returns one output stream.
3201 First input stream will be processed depending on second stream signal.
3202 The filtered signal then can be filtered with other filters in later stages of
3203 processing. See @ref{pan} and @ref{amerge} filter.
3205 The filter accepts the following options:
3209 Set input gain. Default is 1. Range is between 0.015625 and 64.
3212 If a signal of second stream raises above this level it will affect the gain
3213 reduction of first stream.
3214 By default is 0.125. Range is between 0.00097563 and 1.
3217 Set a ratio about which the signal is reduced. 1:2 means that if the level
3218 raised 4dB above the threshold, it will be only 2dB above after the reduction.
3219 Default is 2. Range is between 1 and 20.
3222 Amount of milliseconds the signal has to rise above the threshold before gain
3223 reduction starts. Default is 20. Range is between 0.01 and 2000.
3226 Amount of milliseconds the signal has to fall below the threshold before
3227 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
3230 Set the amount by how much signal will be amplified after processing.
3231 Default is 2. Range is from 1 and 64.
3234 Curve the sharp knee around the threshold to enter gain reduction more softly.
3235 Default is 2.82843. Range is between 1 and 8.
3238 Choose if the @code{average} level between all channels of side-chain stream
3239 or the louder(@code{maximum}) channel of side-chain stream affects the
3240 reduction. Default is @code{average}.
3243 Should the exact signal be taken in case of @code{peak} or an RMS one in case
3244 of @code{rms}. Default is @code{rms} which is mainly smoother.
3247 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
3250 How much to use compressed signal in output. Default is 1.
3251 Range is between 0 and 1.
3254 @subsection Examples
3258 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
3259 depending on the signal of 2nd input and later compressed signal to be
3260 merged with 2nd input:
3262 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3266 @section sidechaingate
3268 A sidechain gate acts like a normal (wideband) gate but has the ability to
3269 filter the detected signal before sending it to the gain reduction stage.
3270 Normally a gate uses the full range signal to detect a level above the
3272 For example: If you cut all lower frequencies from your sidechain signal
3273 the gate will decrease the volume of your track only if not enough highs
3274 appear. With this technique you are able to reduce the resonation of a
3275 natural drum or remove "rumbling" of muted strokes from a heavily distorted
3277 It needs two input streams and returns one output stream.
3278 First input stream will be processed depending on second stream signal.
3280 The filter accepts the following options:
3284 Set input level before filtering.
3285 Default is 1. Allowed range is from 0.015625 to 64.
3288 Set the level of gain reduction when the signal is below the threshold.
3289 Default is 0.06125. Allowed range is from 0 to 1.
3292 If a signal rises above this level the gain reduction is released.
3293 Default is 0.125. Allowed range is from 0 to 1.
3296 Set a ratio about which the signal is reduced.
3297 Default is 2. Allowed range is from 1 to 9000.
3300 Amount of milliseconds the signal has to rise above the threshold before gain
3302 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
3305 Amount of milliseconds the signal has to fall below the threshold before the
3306 reduction is increased again. Default is 250 milliseconds.
3307 Allowed range is from 0.01 to 9000.
3310 Set amount of amplification of signal after processing.
3311 Default is 1. Allowed range is from 1 to 64.
3314 Curve the sharp knee around the threshold to enter gain reduction more softly.
3315 Default is 2.828427125. Allowed range is from 1 to 8.
3318 Choose if exact signal should be taken for detection or an RMS like one.
3319 Default is rms. Can be peak or rms.
3322 Choose if the average level between all channels or the louder channel affects
3324 Default is average. Can be average or maximum.
3327 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3330 @section silencedetect
3332 Detect silence in an audio stream.
3334 This filter logs a message when it detects that the input audio volume is less
3335 or equal to a noise tolerance value for a duration greater or equal to the
3336 minimum detected noise duration.
3338 The printed times and duration are expressed in seconds.
3340 The filter accepts the following options:
3344 Set silence duration until notification (default is 2 seconds).
3347 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
3348 specified value) or amplitude ratio. Default is -60dB, or 0.001.
3351 @subsection Examples
3355 Detect 5 seconds of silence with -50dB noise tolerance:
3357 silencedetect=n=-50dB:d=5
3361 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3362 tolerance in @file{silence.mp3}:
3364 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3368 @section silenceremove
3370 Remove silence from the beginning, middle or end of the audio.
3372 The filter accepts the following options:
3376 This value is used to indicate if audio should be trimmed at beginning of
3377 the audio. A value of zero indicates no silence should be trimmed from the
3378 beginning. When specifying a non-zero value, it trims audio up until it
3379 finds non-silence. Normally, when trimming silence from beginning of audio
3380 the @var{start_periods} will be @code{1} but it can be increased to higher
3381 values to trim all audio up to specific count of non-silence periods.
3382 Default value is @code{0}.
3384 @item start_duration
3385 Specify the amount of time that non-silence must be detected before it stops
3386 trimming audio. By increasing the duration, bursts of noises can be treated
3387 as silence and trimmed off. Default value is @code{0}.
3389 @item start_threshold
3390 This indicates what sample value should be treated as silence. For digital
3391 audio, a value of @code{0} may be fine but for audio recorded from analog,
3392 you may wish to increase the value to account for background noise.
3393 Can be specified in dB (in case "dB" is appended to the specified value)
3394 or amplitude ratio. Default value is @code{0}.
3397 Set the count for trimming silence from the end of audio.
3398 To remove silence from the middle of a file, specify a @var{stop_periods}
3399 that is negative. This value is then treated as a positive value and is
3400 used to indicate the effect should restart processing as specified by
3401 @var{start_periods}, making it suitable for removing periods of silence
3402 in the middle of the audio.
3403 Default value is @code{0}.
3406 Specify a duration of silence that must exist before audio is not copied any
3407 more. By specifying a higher duration, silence that is wanted can be left in
3409 Default value is @code{0}.
3411 @item stop_threshold
3412 This is the same as @option{start_threshold} but for trimming silence from
3414 Can be specified in dB (in case "dB" is appended to the specified value)
3415 or amplitude ratio. Default value is @code{0}.
3418 This indicates that @var{stop_duration} length of audio should be left intact
3419 at the beginning of each period of silence.
3420 For example, if you want to remove long pauses between words but do not want
3421 to remove the pauses completely. Default value is @code{0}.
3424 Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
3425 and works better with digital silence which is exactly 0.
3426 Default value is @code{rms}.
3429 Set ratio used to calculate size of window for detecting silence.
3430 Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
3433 @subsection Examples
3437 The following example shows how this filter can be used to start a recording
3438 that does not contain the delay at the start which usually occurs between
3439 pressing the record button and the start of the performance:
3441 silenceremove=1:5:0.02
3445 Trim all silence encountered from beginning to end where there is more than 1
3446 second of silence in audio:
3448 silenceremove=0:0:0:-1:1:-90dB
3454 SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3455 loudspeakers around the user for binaural listening via headphones (audio
3456 formats up to 9 channels supported).
3457 The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3458 SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3459 Austrian Academy of Sciences.
3461 To enable compilation of this filter you need to configure FFmpeg with
3462 @code{--enable-netcdf}.
3464 The filter accepts the following options:
3468 Set the SOFA file used for rendering.
3471 Set gain applied to audio. Value is in dB. Default is 0.
3474 Set rotation of virtual loudspeakers in deg. Default is 0.
3477 Set elevation of virtual speakers in deg. Default is 0.
3480 Set distance in meters between loudspeakers and the listener with near-field
3481 HRTFs. Default is 1.
3484 Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3485 processing audio in time domain which is slow.
3486 @var{freq} is processing audio in frequency domain which is fast.
3487 Default is @var{freq}.
3490 Set custom positions of virtual loudspeakers. Syntax for this option is:
3491 <CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...].
3492 Each virtual loudspeaker is described with short channel name following with
3493 azimuth and elevation in degreees.
3494 Each virtual loudspeaker description is separated by '|'.
3495 For example to override front left and front right channel positions use:
3496 'speakers=FL 45 15|FR 345 15'.
3497 Descriptions with unrecognised channel names are ignored.
3500 @subsection Examples
3504 Using ClubFritz6 sofa file:
3506 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3510 Using ClubFritz12 sofa file and bigger radius with small rotation:
3512 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
3516 Similar as above but with custom speaker positions for front left, front right, back left and back right
3517 and also with custom gain:
3519 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|BL 135|BR 225:gain=28"
3523 @section stereotools
3525 This filter has some handy utilities to manage stereo signals, for converting
3526 M/S stereo recordings to L/R signal while having control over the parameters
3527 or spreading the stereo image of master track.
3529 The filter accepts the following options:
3533 Set input level before filtering for both channels. Defaults is 1.
3534 Allowed range is from 0.015625 to 64.
3537 Set output level after filtering for both channels. Defaults is 1.
3538 Allowed range is from 0.015625 to 64.
3541 Set input balance between both channels. Default is 0.
3542 Allowed range is from -1 to 1.
3545 Set output balance between both channels. Default is 0.
3546 Allowed range is from -1 to 1.
3549 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3550 clipping. Disabled by default.
3553 Mute the left channel. Disabled by default.
3556 Mute the right channel. Disabled by default.
3559 Change the phase of the left channel. Disabled by default.
3562 Change the phase of the right channel. Disabled by default.
3565 Set stereo mode. Available values are:
3569 Left/Right to Left/Right, this is default.
3572 Left/Right to Mid/Side.
3575 Mid/Side to Left/Right.
3578 Left/Right to Left/Left.
3581 Left/Right to Right/Right.
3584 Left/Right to Left + Right.
3587 Left/Right to Right/Left.
3591 Set level of side signal. Default is 1.
3592 Allowed range is from 0.015625 to 64.
3595 Set balance of side signal. Default is 0.
3596 Allowed range is from -1 to 1.
3599 Set level of the middle signal. Default is 1.
3600 Allowed range is from 0.015625 to 64.
3603 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3606 Set stereo base between mono and inversed channels. Default is 0.
3607 Allowed range is from -1 to 1.
3610 Set delay in milliseconds how much to delay left from right channel and
3611 vice versa. Default is 0. Allowed range is from -20 to 20.
3614 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3617 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3620 @subsection Examples
3624 Apply karaoke like effect:
3626 stereotools=mlev=0.015625
3630 Convert M/S signal to L/R:
3632 "stereotools=mode=ms>lr"
3636 @section stereowiden
3638 This filter enhance the stereo effect by suppressing signal common to both
3639 channels and by delaying the signal of left into right and vice versa,
3640 thereby widening the stereo effect.
3642 The filter accepts the following options:
3646 Time in milliseconds of the delay of left signal into right and vice versa.
3647 Default is 20 milliseconds.
3650 Amount of gain in delayed signal into right and vice versa. Gives a delay
3651 effect of left signal in right output and vice versa which gives widening
3652 effect. Default is 0.3.
3655 Cross feed of left into right with inverted phase. This helps in suppressing
3656 the mono. If the value is 1 it will cancel all the signal common to both
3657 channels. Default is 0.3.
3660 Set level of input signal of original channel. Default is 0.8.
3665 Boost or cut treble (upper) frequencies of the audio using a two-pole
3666 shelving filter with a response similar to that of a standard
3667 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3669 The filter accepts the following options:
3673 Give the gain at whichever is the lower of ~22 kHz and the
3674 Nyquist frequency. Its useful range is about -20 (for a large cut)
3675 to +20 (for a large boost). Beware of clipping when using a positive gain.
3678 Set the filter's central frequency and so can be used
3679 to extend or reduce the frequency range to be boosted or cut.
3680 The default value is @code{3000} Hz.
3683 Set method to specify band-width of filter.
3696 Determine how steep is the filter's shelf transition.
3701 Sinusoidal amplitude modulation.
3703 The filter accepts the following options:
3707 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3708 (20 Hz or lower) will result in a tremolo effect.
3709 This filter may also be used as a ring modulator by specifying
3710 a modulation frequency higher than 20 Hz.
3711 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3714 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3715 Default value is 0.5.
3720 Sinusoidal phase modulation.
3722 The filter accepts the following options:
3726 Modulation frequency in Hertz.
3727 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3730 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3731 Default value is 0.5.
3736 Adjust the input audio volume.
3738 It accepts the following parameters:
3742 Set audio volume expression.
3744 Output values are clipped to the maximum value.
3746 The output audio volume is given by the relation:
3748 @var{output_volume} = @var{volume} * @var{input_volume}
3751 The default value for @var{volume} is "1.0".
3754 This parameter represents the mathematical precision.
3756 It determines which input sample formats will be allowed, which affects the
3757 precision of the volume scaling.
3761 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3763 32-bit floating-point; this limits input sample format to FLT. (default)
3765 64-bit floating-point; this limits input sample format to DBL.
3769 Choose the behaviour on encountering ReplayGain side data in input frames.
3773 Remove ReplayGain side data, ignoring its contents (the default).
3776 Ignore ReplayGain side data, but leave it in the frame.
3779 Prefer the track gain, if present.
3782 Prefer the album gain, if present.
3785 @item replaygain_preamp
3786 Pre-amplification gain in dB to apply to the selected replaygain gain.
3788 Default value for @var{replaygain_preamp} is 0.0.
3791 Set when the volume expression is evaluated.
3793 It accepts the following values:
3796 only evaluate expression once during the filter initialization, or
3797 when the @samp{volume} command is sent
3800 evaluate expression for each incoming frame
3803 Default value is @samp{once}.
3806 The volume expression can contain the following parameters.
3810 frame number (starting at zero)
3813 @item nb_consumed_samples
3814 number of samples consumed by the filter
3816 number of samples in the current frame
3818 original frame position in the file
3824 PTS at start of stream
3826 time at start of stream
3832 last set volume value
3835 Note that when @option{eval} is set to @samp{once} only the
3836 @var{sample_rate} and @var{tb} variables are available, all other
3837 variables will evaluate to NAN.
3839 @subsection Commands
3841 This filter supports the following commands:
3844 Modify the volume expression.
3845 The command accepts the same syntax of the corresponding option.
3847 If the specified expression is not valid, it is kept at its current
3849 @item replaygain_noclip
3850 Prevent clipping by limiting the gain applied.
3852 Default value for @var{replaygain_noclip} is 1.
3856 @subsection Examples
3860 Halve the input audio volume:
3864 volume=volume=-6.0206dB
3867 In all the above example the named key for @option{volume} can be
3868 omitted, for example like in:
3874 Increase input audio power by 6 decibels using fixed-point precision:
3876 volume=volume=6dB:precision=fixed
3880 Fade volume after time 10 with an annihilation period of 5 seconds:
3882 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3886 @section volumedetect
3888 Detect the volume of the input video.
3890 The filter has no parameters. The input is not modified. Statistics about
3891 the volume will be printed in the log when the input stream end is reached.
3893 In particular it will show the mean volume (root mean square), maximum
3894 volume (on a per-sample basis), and the beginning of a histogram of the
3895 registered volume values (from the maximum value to a cumulated 1/1000 of
3898 All volumes are in decibels relative to the maximum PCM value.
3900 @subsection Examples
3902 Here is an excerpt of the output:
3904 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3905 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3906 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3907 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3908 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3909 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3910 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3911 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3912 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3918 The mean square energy is approximately -27 dB, or 10^-2.7.
3920 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3922 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3925 In other words, raising the volume by +4 dB does not cause any clipping,
3926 raising it by +5 dB causes clipping for 6 samples, etc.
3928 @c man end AUDIO FILTERS
3930 @chapter Audio Sources
3931 @c man begin AUDIO SOURCES
3933 Below is a description of the currently available audio sources.
3937 Buffer audio frames, and make them available to the filter chain.
3939 This source is mainly intended for a programmatic use, in particular
3940 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3942 It accepts the following parameters:
3946 The timebase which will be used for timestamps of submitted frames. It must be
3947 either a floating-point number or in @var{numerator}/@var{denominator} form.
3950 The sample rate of the incoming audio buffers.
3953 The sample format of the incoming audio buffers.
3954 Either a sample format name or its corresponding integer representation from
3955 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3957 @item channel_layout
3958 The channel layout of the incoming audio buffers.
3959 Either a channel layout name from channel_layout_map in
3960 @file{libavutil/channel_layout.c} or its corresponding integer representation
3961 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3964 The number of channels of the incoming audio buffers.
3965 If both @var{channels} and @var{channel_layout} are specified, then they
3970 @subsection Examples
3973 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3976 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3977 Since the sample format with name "s16p" corresponds to the number
3978 6 and the "stereo" channel layout corresponds to the value 0x3, this is
3981 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3986 Generate an audio signal specified by an expression.
3988 This source accepts in input one or more expressions (one for each
3989 channel), which are evaluated and used to generate a corresponding
3992 This source accepts the following options:
3996 Set the '|'-separated expressions list for each separate channel. In case the
3997 @option{channel_layout} option is not specified, the selected channel layout
3998 depends on the number of provided expressions. Otherwise the last
3999 specified expression is applied to the remaining output channels.
4001 @item channel_layout, c
4002 Set the channel layout. The number of channels in the specified layout
4003 must be equal to the number of specified expressions.
4006 Set the minimum duration of the sourced audio. See
4007 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
4008 for the accepted syntax.
4009 Note that the resulting duration may be greater than the specified
4010 duration, as the generated audio is always cut at the end of a
4013 If not specified, or the expressed duration is negative, the audio is
4014 supposed to be generated forever.
4017 Set the number of samples per channel per each output frame,
4020 @item sample_rate, s
4021 Specify the sample rate, default to 44100.
4024 Each expression in @var{exprs} can contain the following constants:
4028 number of the evaluated sample, starting from 0
4031 time of the evaluated sample expressed in seconds, starting from 0
4038 @subsection Examples
4048 Generate a sin signal with frequency of 440 Hz, set sample rate to
4051 aevalsrc="sin(440*2*PI*t):s=8000"
4055 Generate a two channels signal, specify the channel layout (Front
4056 Center + Back Center) explicitly:
4058 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
4062 Generate white noise:
4064 aevalsrc="-2+random(0)"
4068 Generate an amplitude modulated signal:
4070 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
4074 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
4076 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
4083 The null audio source, return unprocessed audio frames. It is mainly useful
4084 as a template and to be employed in analysis / debugging tools, or as
4085 the source for filters which ignore the input data (for example the sox
4088 This source accepts the following options:
4092 @item channel_layout, cl
4094 Specifies the channel layout, and can be either an integer or a string
4095 representing a channel layout. The default value of @var{channel_layout}
4098 Check the channel_layout_map definition in
4099 @file{libavutil/channel_layout.c} for the mapping between strings and
4100 channel layout values.
4102 @item sample_rate, r
4103 Specifies the sample rate, and defaults to 44100.
4106 Set the number of samples per requested frames.
4110 @subsection Examples
4114 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
4116 anullsrc=r=48000:cl=4
4120 Do the same operation with a more obvious syntax:
4122 anullsrc=r=48000:cl=mono
4126 All the parameters need to be explicitly defined.
4130 Synthesize a voice utterance using the libflite library.
4132 To enable compilation of this filter you need to configure FFmpeg with
4133 @code{--enable-libflite}.
4135 Note that the flite library is not thread-safe.
4137 The filter accepts the following options:
4142 If set to 1, list the names of the available voices and exit
4143 immediately. Default value is 0.
4146 Set the maximum number of samples per frame. Default value is 512.
4149 Set the filename containing the text to speak.
4152 Set the text to speak.
4155 Set the voice to use for the speech synthesis. Default value is
4156 @code{kal}. See also the @var{list_voices} option.
4159 @subsection Examples
4163 Read from file @file{speech.txt}, and synthesize the text using the
4164 standard flite voice:
4166 flite=textfile=speech.txt
4170 Read the specified text selecting the @code{slt} voice:
4172 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4176 Input text to ffmpeg:
4178 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4182 Make @file{ffplay} speak the specified text, using @code{flite} and
4183 the @code{lavfi} device:
4185 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
4189 For more information about libflite, check:
4190 @url{http://www.speech.cs.cmu.edu/flite/}
4194 Generate a noise audio signal.
4196 The filter accepts the following options:
4199 @item sample_rate, r
4200 Specify the sample rate. Default value is 48000 Hz.
4203 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
4207 Specify the duration of the generated audio stream. Not specifying this option
4208 results in noise with an infinite length.
4210 @item color, colour, c
4211 Specify the color of noise. Available noise colors are white, pink, and brown.
4212 Default color is white.
4215 Specify a value used to seed the PRNG.
4218 Set the number of samples per each output frame, default is 1024.
4221 @subsection Examples
4226 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
4228 anoisesrc=d=60:c=pink:r=44100:a=0.5
4234 Generate an audio signal made of a sine wave with amplitude 1/8.
4236 The audio signal is bit-exact.
4238 The filter accepts the following options:
4243 Set the carrier frequency. Default is 440 Hz.
4245 @item beep_factor, b
4246 Enable a periodic beep every second with frequency @var{beep_factor} times
4247 the carrier frequency. Default is 0, meaning the beep is disabled.
4249 @item sample_rate, r
4250 Specify the sample rate, default is 44100.
4253 Specify the duration of the generated audio stream.
4255 @item samples_per_frame
4256 Set the number of samples per output frame.
4258 The expression can contain the following constants:
4262 The (sequential) number of the output audio frame, starting from 0.
4265 The PTS (Presentation TimeStamp) of the output audio frame,
4266 expressed in @var{TB} units.
4269 The PTS of the output audio frame, expressed in seconds.
4272 The timebase of the output audio frames.
4275 Default is @code{1024}.
4278 @subsection Examples
4283 Generate a simple 440 Hz sine wave:
4289 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4293 sine=frequency=220:beep_factor=4:duration=5
4297 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4300 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4304 @c man end AUDIO SOURCES
4306 @chapter Audio Sinks
4307 @c man begin AUDIO SINKS
4309 Below is a description of the currently available audio sinks.
4311 @section abuffersink
4313 Buffer audio frames, and make them available to the end of filter chain.
4315 This sink is mainly intended for programmatic use, in particular
4316 through the interface defined in @file{libavfilter/buffersink.h}
4317 or the options system.
4319 It accepts a pointer to an AVABufferSinkContext structure, which
4320 defines the incoming buffers' formats, to be passed as the opaque
4321 parameter to @code{avfilter_init_filter} for initialization.
4324 Null audio sink; do absolutely nothing with the input audio. It is
4325 mainly useful as a template and for use in analysis / debugging
4328 @c man end AUDIO SINKS
4330 @chapter Video Filters
4331 @c man begin VIDEO FILTERS
4333 When you configure your FFmpeg build, you can disable any of the
4334 existing filters using @code{--disable-filters}.
4335 The configure output will show the video filters included in your
4338 Below is a description of the currently available video filters.
4340 @section alphaextract
4342 Extract the alpha component from the input as a grayscale video. This
4343 is especially useful with the @var{alphamerge} filter.
4347 Add or replace the alpha component of the primary input with the
4348 grayscale value of a second input. This is intended for use with
4349 @var{alphaextract} to allow the transmission or storage of frame
4350 sequences that have alpha in a format that doesn't support an alpha
4353 For example, to reconstruct full frames from a normal YUV-encoded video
4354 and a separate video created with @var{alphaextract}, you might use:
4356 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
4359 Since this filter is designed for reconstruction, it operates on frame
4360 sequences without considering timestamps, and terminates when either
4361 input reaches end of stream. This will cause problems if your encoding
4362 pipeline drops frames. If you're trying to apply an image as an
4363 overlay to a video stream, consider the @var{overlay} filter instead.
4367 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
4368 and libavformat to work. On the other hand, it is limited to ASS (Advanced
4369 Substation Alpha) subtitles files.
4371 This filter accepts the following option in addition to the common options from
4372 the @ref{subtitles} filter:
4376 Set the shaping engine
4378 Available values are:
4381 The default libass shaping engine, which is the best available.
4383 Fast, font-agnostic shaper that can do only substitutions
4385 Slower shaper using OpenType for substitutions and positioning
4388 The default is @code{auto}.
4392 Apply an Adaptive Temporal Averaging Denoiser to the video input.
4394 The filter accepts the following options:
4398 Set threshold A for 1st plane. Default is 0.02.
4399 Valid range is 0 to 0.3.
4402 Set threshold B for 1st plane. Default is 0.04.
4403 Valid range is 0 to 5.
4406 Set threshold A for 2nd plane. Default is 0.02.
4407 Valid range is 0 to 0.3.
4410 Set threshold B for 2nd plane. Default is 0.04.
4411 Valid range is 0 to 5.
4414 Set threshold A for 3rd plane. Default is 0.02.
4415 Valid range is 0 to 0.3.
4418 Set threshold B for 3rd plane. Default is 0.04.
4419 Valid range is 0 to 5.
4421 Threshold A is designed to react on abrupt changes in the input signal and
4422 threshold B is designed to react on continuous changes in the input signal.
4425 Set number of frames filter will use for averaging. Default is 33. Must be odd
4426 number in range [5, 129].
4429 Set what planes of frame filter will use for averaging. Default is all.
4434 Apply average blur filter.
4436 The filter accepts the following options:
4440 Set horizontal kernel size.
4443 Set which planes to filter. By default all planes are filtered.
4446 Set vertical kernel size, if zero it will be same as @code{sizeX}.
4447 Default is @code{0}.
4452 Compute the bounding box for the non-black pixels in the input frame
4455 This filter computes the bounding box containing all the pixels with a
4456 luminance value greater than the minimum allowed value.
4457 The parameters describing the bounding box are printed on the filter
4460 The filter accepts the following option:
4464 Set the minimal luminance value. Default is @code{16}.
4467 @section bitplanenoise
4469 Show and measure bit plane noise.
4471 The filter accepts the following options:
4475 Set which plane to analyze. Default is @code{1}.
4478 Filter out noisy pixels from @code{bitplane} set above.
4479 Default is disabled.
4482 @section blackdetect
4484 Detect video intervals that are (almost) completely black. Can be
4485 useful to detect chapter transitions, commercials, or invalid
4486 recordings. Output lines contains the time for the start, end and
4487 duration of the detected black interval expressed in seconds.
4489 In order to display the output lines, you need to set the loglevel at
4490 least to the AV_LOG_INFO value.
4492 The filter accepts the following options:
4495 @item black_min_duration, d
4496 Set the minimum detected black duration expressed in seconds. It must
4497 be a non-negative floating point number.
4499 Default value is 2.0.
4501 @item picture_black_ratio_th, pic_th
4502 Set the threshold for considering a picture "black".
4503 Express the minimum value for the ratio:
4505 @var{nb_black_pixels} / @var{nb_pixels}
4508 for which a picture is considered black.
4509 Default value is 0.98.
4511 @item pixel_black_th, pix_th
4512 Set the threshold for considering a pixel "black".
4514 The threshold expresses the maximum pixel luminance value for which a
4515 pixel is considered "black". The provided value is scaled according to
4516 the following equation:
4518 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
4521 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
4522 the input video format, the range is [0-255] for YUV full-range
4523 formats and [16-235] for YUV non full-range formats.
4525 Default value is 0.10.
4528 The following example sets the maximum pixel threshold to the minimum
4529 value, and detects only black intervals of 2 or more seconds:
4531 blackdetect=d=2:pix_th=0.00
4536 Detect frames that are (almost) completely black. Can be useful to
4537 detect chapter transitions or commercials. Output lines consist of
4538 the frame number of the detected frame, the percentage of blackness,
4539 the position in the file if known or -1 and the timestamp in seconds.
4541 In order to display the output lines, you need to set the loglevel at
4542 least to the AV_LOG_INFO value.
4544 This filter exports frame metadata @code{lavfi.blackframe.pblack}.
4545 The value represents the percentage of pixels in the picture that
4546 are below the threshold value.
4548 It accepts the following parameters:
4553 The percentage of the pixels that have to be below the threshold; it defaults to
4556 @item threshold, thresh
4557 The threshold below which a pixel value is considered black; it defaults to
4562 @section blend, tblend
4564 Blend two video frames into each other.
4566 The @code{blend} filter takes two input streams and outputs one
4567 stream, the first input is the "top" layer and second input is
4568 "bottom" layer. By default, the output terminates when the longest input terminates.
4570 The @code{tblend} (time blend) filter takes two consecutive frames
4571 from one single stream, and outputs the result obtained by blending
4572 the new frame on top of the old frame.
4574 A description of the accepted options follows.
4582 Set blend mode for specific pixel component or all pixel components in case
4583 of @var{all_mode}. Default value is @code{normal}.
4585 Available values for component modes are:
4626 Set blend opacity for specific pixel component or all pixel components in case
4627 of @var{all_opacity}. Only used in combination with pixel component blend modes.
4634 Set blend expression for specific pixel component or all pixel components in case
4635 of @var{all_expr}. Note that related mode options will be ignored if those are set.
4637 The expressions can use the following variables:
4641 The sequential number of the filtered frame, starting from @code{0}.
4645 the coordinates of the current sample
4649 the width and height of currently filtered plane
4653 Width and height scale depending on the currently filtered plane. It is the
4654 ratio between the corresponding luma plane number of pixels and the current
4655 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4656 @code{0.5,0.5} for chroma planes.
4659 Time of the current frame, expressed in seconds.
4662 Value of pixel component at current location for first video frame (top layer).
4665 Value of pixel component at current location for second video frame (bottom layer).
4669 Force termination when the shortest input terminates. Default is
4670 @code{0}. This option is only defined for the @code{blend} filter.
4673 Continue applying the last bottom frame after the end of the stream. A value of
4674 @code{0} disable the filter after the last frame of the bottom layer is reached.
4675 Default is @code{1}. This option is only defined for the @code{blend} filter.
4678 @subsection Examples
4682 Apply transition from bottom layer to top layer in first 10 seconds:
4684 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4688 Apply 1x1 checkerboard effect:
4690 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4694 Apply uncover left effect:
4696 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4700 Apply uncover down effect:
4702 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4706 Apply uncover up-left effect:
4708 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4712 Split diagonally video and shows top and bottom layer on each side:
4714 blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4718 Display differences between the current and the previous frame:
4720 tblend=all_mode=difference128
4726 Apply a boxblur algorithm to the input video.
4728 It accepts the following parameters:
4732 @item luma_radius, lr
4733 @item luma_power, lp
4734 @item chroma_radius, cr
4735 @item chroma_power, cp
4736 @item alpha_radius, ar
4737 @item alpha_power, ap
4741 A description of the accepted options follows.
4744 @item luma_radius, lr
4745 @item chroma_radius, cr
4746 @item alpha_radius, ar
4747 Set an expression for the box radius in pixels used for blurring the
4748 corresponding input plane.
4750 The radius value must be a non-negative number, and must not be
4751 greater than the value of the expression @code{min(w,h)/2} for the
4752 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4755 Default value for @option{luma_radius} is "2". If not specified,
4756 @option{chroma_radius} and @option{alpha_radius} default to the
4757 corresponding value set for @option{luma_radius}.
4759 The expressions can contain the following constants:
4763 The input width and height in pixels.
4767 The input chroma image width and height in pixels.
4771 The horizontal and vertical chroma subsample values. For example, for the
4772 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4775 @item luma_power, lp
4776 @item chroma_power, cp
4777 @item alpha_power, ap
4778 Specify how many times the boxblur filter is applied to the
4779 corresponding plane.
4781 Default value for @option{luma_power} is 2. If not specified,
4782 @option{chroma_power} and @option{alpha_power} default to the
4783 corresponding value set for @option{luma_power}.
4785 A value of 0 will disable the effect.
4788 @subsection Examples
4792 Apply a boxblur filter with the luma, chroma, and alpha radii
4795 boxblur=luma_radius=2:luma_power=1
4800 Set the luma radius to 2, and alpha and chroma radius to 0:
4802 boxblur=2:1:cr=0:ar=0
4806 Set the luma and chroma radii to a fraction of the video dimension:
4808 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4814 Deinterlace the input video ("bwdif" stands for "Bob Weaver
4815 Deinterlacing Filter").
4817 Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4818 interpolation algorithms.
4819 It accepts the following parameters:
4823 The interlacing mode to adopt. It accepts one of the following values:
4827 Output one frame for each frame.
4829 Output one frame for each field.
4832 The default value is @code{send_field}.
4835 The picture field parity assumed for the input interlaced video. It accepts one
4836 of the following values:
4840 Assume the top field is first.
4842 Assume the bottom field is first.
4844 Enable automatic detection of field parity.
4847 The default value is @code{auto}.
4848 If the interlacing is unknown or the decoder does not export this information,
4849 top field first will be assumed.
4852 Specify which frames to deinterlace. Accept one of the following
4857 Deinterlace all frames.
4859 Only deinterlace frames marked as interlaced.
4862 The default value is @code{all}.
4866 YUV colorspace color/chroma keying.
4868 The filter accepts the following options:
4872 The color which will be replaced with transparency.
4875 Similarity percentage with the key color.
4877 0.01 matches only the exact key color, while 1.0 matches everything.
4882 0.0 makes pixels either fully transparent, or not transparent at all.
4884 Higher values result in semi-transparent pixels, with a higher transparency
4885 the more similar the pixels color is to the key color.
4888 Signals that the color passed is already in YUV instead of RGB.
4890 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4891 This can be used to pass exact YUV values as hexadecimal numbers.
4894 @subsection Examples
4898 Make every green pixel in the input image transparent:
4900 ffmpeg -i input.png -vf chromakey=green out.png
4904 Overlay a greenscreen-video on top of a static black background.
4906 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
4912 Display CIE color diagram with pixels overlaid onto it.
4914 The filter accepts the following options:
4929 @item uhdtv, rec2020
4942 Set what gamuts to draw.
4944 See @code{system} option for available values.
4947 Set ciescope size, by default set to 512.
4950 Set intensity used to map input pixel values to CIE diagram.
4953 Set contrast used to draw tongue colors that are out of active color system gamut.
4956 Correct gamma displayed on scope, by default enabled.
4959 Show white point on CIE diagram, by default disabled.
4962 Set input gamma. Used only with XYZ input color space.
4967 Visualize information exported by some codecs.
4969 Some codecs can export information through frames using side-data or other
4970 means. For example, some MPEG based codecs export motion vectors through the
4971 @var{export_mvs} flag in the codec @option{flags2} option.
4973 The filter accepts the following option:
4977 Set motion vectors to visualize.
4979 Available flags for @var{mv} are:
4983 forward predicted MVs of P-frames
4985 forward predicted MVs of B-frames
4987 backward predicted MVs of B-frames
4991 Display quantization parameters using the chroma planes.
4994 Set motion vectors type to visualize. Includes MVs from all frames unless specified by @var{frame_type} option.
4996 Available flags for @var{mv_type} are:
5000 forward predicted MVs
5002 backward predicted MVs
5005 @item frame_type, ft
5006 Set frame type to visualize motion vectors of.
5008 Available flags for @var{frame_type} are:
5012 intra-coded frames (I-frames)
5014 predicted frames (P-frames)
5016 bi-directionally predicted frames (B-frames)
5020 @subsection Examples
5024 Visualize forward predicted MVs of all frames using @command{ffplay}:
5026 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
5030 Visualize multi-directionals MVs of P and B-Frames using @command{ffplay}:
5032 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
5036 @section colorbalance
5037 Modify intensity of primary colors (red, green and blue) of input frames.
5039 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
5040 regions for the red-cyan, green-magenta or blue-yellow balance.
5042 A positive adjustment value shifts the balance towards the primary color, a negative
5043 value towards the complementary color.
5045 The filter accepts the following options:
5051 Adjust red, green and blue shadows (darkest pixels).
5056 Adjust red, green and blue midtones (medium pixels).
5061 Adjust red, green and blue highlights (brightest pixels).
5063 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5066 @subsection Examples
5070 Add red color cast to shadows:
5077 RGB colorspace color keying.
5079 The filter accepts the following options:
5083 The color which will be replaced with transparency.
5086 Similarity percentage with the key color.
5088 0.01 matches only the exact key color, while 1.0 matches everything.
5093 0.0 makes pixels either fully transparent, or not transparent at all.
5095 Higher values result in semi-transparent pixels, with a higher transparency
5096 the more similar the pixels color is to the key color.
5099 @subsection Examples
5103 Make every green pixel in the input image transparent:
5105 ffmpeg -i input.png -vf colorkey=green out.png
5109 Overlay a greenscreen-video on top of a static background image.
5111 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
5115 @section colorlevels
5117 Adjust video input frames using levels.
5119 The filter accepts the following options:
5126 Adjust red, green, blue and alpha input black point.
5127 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5133 Adjust red, green, blue and alpha input white point.
5134 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
5136 Input levels are used to lighten highlights (bright tones), darken shadows
5137 (dark tones), change the balance of bright and dark tones.
5143 Adjust red, green, blue and alpha output black point.
5144 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
5150 Adjust red, green, blue and alpha output white point.
5151 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
5153 Output levels allows manual selection of a constrained output level range.
5156 @subsection Examples
5160 Make video output darker:
5162 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
5168 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
5172 Make video output lighter:
5174 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
5178 Increase brightness:
5180 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
5184 @section colorchannelmixer
5186 Adjust video input frames by re-mixing color channels.
5188 This filter modifies a color channel by adding the values associated to
5189 the other channels of the same pixels. For example if the value to
5190 modify is red, the output value will be:
5192 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
5195 The filter accepts the following options:
5202 Adjust contribution of input red, green, blue and alpha channels for output red channel.
5203 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
5209 Adjust contribution of input red, green, blue and alpha channels for output green channel.
5210 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
5216 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
5217 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
5223 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
5224 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
5226 Allowed ranges for options are @code{[-2.0, 2.0]}.
5229 @subsection Examples
5233 Convert source to grayscale:
5235 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
5238 Simulate sepia tones:
5240 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
5244 @section colormatrix
5246 Convert color matrix.
5248 The filter accepts the following options:
5253 Specify the source and destination color matrix. Both values must be
5256 The accepted values are:
5275 For example to convert from BT.601 to SMPTE-240M, use the command:
5277 colormatrix=bt601:smpte240m
5282 Convert colorspace, transfer characteristics or color primaries.
5283 Input video needs to have an even size.
5285 The filter accepts the following options:
5290 Specify all color properties at once.
5292 The accepted values are:
5322 Specify output colorspace.
5324 The accepted values are:
5333 BT.470BG or BT.601-6 625
5336 SMPTE-170M or BT.601-6 525
5345 BT.2020 with non-constant luminance
5351 Specify output transfer characteristics.
5353 The accepted values are:
5365 Constant gamma of 2.2
5368 Constant gamma of 2.8
5371 SMPTE-170M, BT.601-6 625 or BT.601-6 525
5389 BT.2020 for 10-bits content
5392 BT.2020 for 12-bits content
5398 Specify output color primaries.
5400 The accepted values are:
5409 BT.470BG or BT.601-6 625
5412 SMPTE-170M or BT.601-6 525
5433 Specify output color range.
5435 The accepted values are:
5438 TV (restricted) range
5441 MPEG (restricted) range
5452 Specify output color format.
5454 The accepted values are:
5457 YUV 4:2:0 planar 8-bits
5460 YUV 4:2:0 planar 10-bits
5463 YUV 4:2:0 planar 12-bits
5466 YUV 4:2:2 planar 8-bits
5469 YUV 4:2:2 planar 10-bits
5472 YUV 4:2:2 planar 12-bits
5475 YUV 4:4:4 planar 8-bits
5478 YUV 4:4:4 planar 10-bits
5481 YUV 4:4:4 planar 12-bits
5486 Do a fast conversion, which skips gamma/primary correction. This will take
5487 significantly less CPU, but will be mathematically incorrect. To get output
5488 compatible with that produced by the colormatrix filter, use fast=1.
5491 Specify dithering mode.
5493 The accepted values are:
5499 Floyd-Steinberg dithering
5503 Whitepoint adaptation mode.
5505 The accepted values are:
5508 Bradford whitepoint adaptation
5511 von Kries whitepoint adaptation
5514 identity whitepoint adaptation (i.e. no whitepoint adaptation)
5518 Override all input properties at once. Same accepted values as @ref{all}.
5521 Override input colorspace. Same accepted values as @ref{space}.
5524 Override input color primaries. Same accepted values as @ref{primaries}.
5527 Override input transfer characteristics. Same accepted values as @ref{trc}.
5530 Override input color range. Same accepted values as @ref{range}.
5534 The filter converts the transfer characteristics, color space and color
5535 primaries to the specified user values. The output value, if not specified,
5536 is set to a default value based on the "all" property. If that property is
5537 also not specified, the filter will log an error. The output color range and
5538 format default to the same value as the input color range and format. The
5539 input transfer characteristics, color space, color primaries and color range
5540 should be set on the input data. If any of these are missing, the filter will
5541 log an error and no conversion will take place.
5543 For example to convert the input to SMPTE-240M, use the command:
5545 colorspace=smpte240m
5548 @section convolution
5550 Apply convolution 3x3 or 5x5 filter.
5552 The filter accepts the following options:
5559 Set matrix for each plane.
5560 Matrix is sequence of 9 or 25 signed integers.
5566 Set multiplier for calculated value for each plane.
5572 Set bias for each plane. This value is added to the result of the multiplication.
5573 Useful for making the overall image brighter or darker. Default is 0.0.
5576 @subsection Examples
5582 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"
5588 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"
5594 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"
5600 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"
5606 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"
5612 Copy the input source unchanged to the output. This is mainly useful for
5617 Video filtering on GPU using Apple's CoreImage API on OSX.
5619 Hardware acceleration is based on an OpenGL context. Usually, this means it is
5620 processed by video hardware. However, software-based OpenGL implementations
5621 exist which means there is no guarantee for hardware processing. It depends on
5624 There are many filters and image generators provided by Apple that come with a
5625 large variety of options. The filter has to be referenced by its name along
5628 The coreimage filter accepts the following options:
5631 List all available filters and generators along with all their respective
5632 options as well as possible minimum and maximum values along with the default
5639 Specify all filters by their respective name and options.
5640 Use @var{list_filters} to determine all valid filter names and options.
5641 Numerical options are specified by a float value and are automatically clamped
5642 to their respective value range. Vector and color options have to be specified
5643 by a list of space separated float values. Character escaping has to be done.
5644 A special option name @code{default} is available to use default options for a
5647 It is required to specify either @code{default} or at least one of the filter options.
5648 All omitted options are used with their default values.
5649 The syntax of the filter string is as follows:
5651 filter=<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...][#<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...]][#...]
5655 Specify a rectangle where the output of the filter chain is copied into the
5656 input image. It is given by a list of space separated float values:
5658 output_rect=x\ y\ width\ height
5660 If not given, the output rectangle equals the dimensions of the input image.
5661 The output rectangle is automatically cropped at the borders of the input
5662 image. Negative values are valid for each component.
5664 output_rect=25\ 25\ 100\ 100
5668 Several filters can be chained for successive processing without GPU-HOST
5669 transfers allowing for fast processing of complex filter chains.
5670 Currently, only filters with zero (generators) or exactly one (filters) input
5671 image and one output image are supported. Also, transition filters are not yet
5674 Some filters generate output images with additional padding depending on the
5675 respective filter kernel. The padding is automatically removed to ensure the
5676 filter output has the same size as the input image.
5678 For image generators, the size of the output image is determined by the
5679 previous output image of the filter chain or the input image of the whole
5680 filterchain, respectively. The generators do not use the pixel information of
5681 this image to generate their output. However, the generated output is
5682 blended onto this image, resulting in partial or complete coverage of the
5685 The @ref{coreimagesrc} video source can be used for generating input images
5686 which are directly fed into the filter chain. By using it, providing input
5687 images by another video source or an input video is not required.
5689 @subsection Examples
5694 List all filters available:
5696 coreimage=list_filters=true
5700 Use the CIBoxBlur filter with default options to blur an image:
5702 coreimage=filter=CIBoxBlur@@default
5706 Use a filter chain with CISepiaTone at default values and CIVignetteEffect with
5707 its center at 100x100 and a radius of 50 pixels:
5709 coreimage=filter=CIBoxBlur@@default#CIVignetteEffect@@inputCenter=100\ 100@@inputRadius=50
5713 Use nullsrc and CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
5714 given as complete and escaped command-line for Apple's standard bash shell:
5716 ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
5722 Crop the input video to given dimensions.
5724 It accepts the following parameters:
5728 The width of the output video. It defaults to @code{iw}.
5729 This expression is evaluated only once during the filter
5730 configuration, or when the @samp{w} or @samp{out_w} command is sent.
5733 The height of the output video. It defaults to @code{ih}.
5734 This expression is evaluated only once during the filter
5735 configuration, or when the @samp{h} or @samp{out_h} command is sent.
5738 The horizontal position, in the input video, of the left edge of the output
5739 video. It defaults to @code{(in_w-out_w)/2}.
5740 This expression is evaluated per-frame.
5743 The vertical position, in the input video, of the top edge of the output video.
5744 It defaults to @code{(in_h-out_h)/2}.
5745 This expression is evaluated per-frame.
5748 If set to 1 will force the output display aspect ratio
5749 to be the same of the input, by changing the output sample aspect
5750 ratio. It defaults to 0.
5753 Enable exact cropping. If enabled, subsampled videos will be cropped at exact
5754 width/height/x/y as specified and will not be rounded to nearest smaller value.
5758 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
5759 expressions containing the following constants:
5764 The computed values for @var{x} and @var{y}. They are evaluated for
5769 The input width and height.
5773 These are the same as @var{in_w} and @var{in_h}.
5777 The output (cropped) width and height.
5781 These are the same as @var{out_w} and @var{out_h}.
5784 same as @var{iw} / @var{ih}
5787 input sample aspect ratio
5790 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
5794 horizontal and vertical chroma subsample values. For example for the
5795 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5798 The number of the input frame, starting from 0.
5801 the position in the file of the input frame, NAN if unknown
5804 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
5808 The expression for @var{out_w} may depend on the value of @var{out_h},
5809 and the expression for @var{out_h} may depend on @var{out_w}, but they
5810 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
5811 evaluated after @var{out_w} and @var{out_h}.
5813 The @var{x} and @var{y} parameters specify the expressions for the
5814 position of the top-left corner of the output (non-cropped) area. They
5815 are evaluated for each frame. If the evaluated value is not valid, it
5816 is approximated to the nearest valid value.
5818 The expression for @var{x} may depend on @var{y}, and the expression
5819 for @var{y} may depend on @var{x}.
5821 @subsection Examples
5825 Crop area with size 100x100 at position (12,34).
5830 Using named options, the example above becomes:
5832 crop=w=100:h=100:x=12:y=34
5836 Crop the central input area with size 100x100:
5842 Crop the central input area with size 2/3 of the input video:
5844 crop=2/3*in_w:2/3*in_h
5848 Crop the input video central square:
5855 Delimit the rectangle with the top-left corner placed at position
5856 100:100 and the right-bottom corner corresponding to the right-bottom
5857 corner of the input image.
5859 crop=in_w-100:in_h-100:100:100
5863 Crop 10 pixels from the left and right borders, and 20 pixels from
5864 the top and bottom borders
5866 crop=in_w-2*10:in_h-2*20
5870 Keep only the bottom right quarter of the input image:
5872 crop=in_w/2:in_h/2:in_w/2:in_h/2
5876 Crop height for getting Greek harmony:
5878 crop=in_w:1/PHI*in_w
5882 Apply trembling effect:
5884 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)
5888 Apply erratic camera effect depending on timestamp:
5890 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)"
5894 Set x depending on the value of y:
5896 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
5900 @subsection Commands
5902 This filter supports the following commands:
5908 Set width/height of the output video and the horizontal/vertical position
5910 The command accepts the same syntax of the corresponding option.
5912 If the specified expression is not valid, it is kept at its current
5918 Auto-detect the crop size.
5920 It calculates the necessary cropping parameters and prints the
5921 recommended parameters via the logging system. The detected dimensions
5922 correspond to the non-black area of the input video.
5924 It accepts the following parameters:
5929 Set higher black value threshold, which can be optionally specified
5930 from nothing (0) to everything (255 for 8-bit based formats). An intensity
5931 value greater to the set value is considered non-black. It defaults to 24.
5932 You can also specify a value between 0.0 and 1.0 which will be scaled depending
5933 on the bitdepth of the pixel format.
5936 The value which the width/height should be divisible by. It defaults to
5937 16. The offset is automatically adjusted to center the video. Use 2 to
5938 get only even dimensions (needed for 4:2:2 video). 16 is best when
5939 encoding to most video codecs.
5941 @item reset_count, reset
5942 Set the counter that determines after how many frames cropdetect will
5943 reset the previously detected largest video area and start over to
5944 detect the current optimal crop area. Default value is 0.
5946 This can be useful when channel logos distort the video area. 0
5947 indicates 'never reset', and returns the largest area encountered during
5954 Apply color adjustments using curves.
5956 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
5957 component (red, green and blue) has its values defined by @var{N} key points
5958 tied from each other using a smooth curve. The x-axis represents the pixel
5959 values from the input frame, and the y-axis the new pixel values to be set for
5962 By default, a component curve is defined by the two points @var{(0;0)} and
5963 @var{(1;1)}. This creates a straight line where each original pixel value is
5964 "adjusted" to its own value, which means no change to the image.
5966 The filter allows you to redefine these two points and add some more. A new
5967 curve (using a natural cubic spline interpolation) will be define to pass
5968 smoothly through all these new coordinates. The new defined points needs to be
5969 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
5970 be in the @var{[0;1]} interval. If the computed curves happened to go outside
5971 the vector spaces, the values will be clipped accordingly.
5973 The filter accepts the following options:
5977 Select one of the available color presets. This option can be used in addition
5978 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
5979 options takes priority on the preset values.
5980 Available presets are:
5983 @item color_negative
5986 @item increase_contrast
5988 @item linear_contrast
5989 @item medium_contrast
5991 @item strong_contrast
5994 Default is @code{none}.
5996 Set the master key points. These points will define a second pass mapping. It
5997 is sometimes called a "luminance" or "value" mapping. It can be used with
5998 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
5999 post-processing LUT.
6001 Set the key points for the red component.
6003 Set the key points for the green component.
6005 Set the key points for the blue component.
6007 Set the key points for all components (not including master).
6008 Can be used in addition to the other key points component
6009 options. In this case, the unset component(s) will fallback on this
6010 @option{all} setting.
6012 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
6014 Save Gnuplot script of the curves in specified file.
6017 To avoid some filtergraph syntax conflicts, each key points list need to be
6018 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
6020 @subsection Examples
6024 Increase slightly the middle level of blue:
6026 curves=blue='0/0 0.5/0.58 1/1'
6032 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'
6034 Here we obtain the following coordinates for each components:
6037 @code{(0;0.11) (0.42;0.51) (1;0.95)}
6039 @code{(0;0) (0.50;0.48) (1;1)}
6041 @code{(0;0.22) (0.49;0.44) (1;0.80)}
6045 The previous example can also be achieved with the associated built-in preset:
6047 curves=preset=vintage
6057 Use a Photoshop preset and redefine the points of the green component:
6059 curves=psfile='MyCurvesPresets/purple.acv':green='0/0 0.45/0.53 1/1'
6063 Check out the curves of the @code{cross_process} profile using @command{ffmpeg}
6064 and @command{gnuplot}:
6066 ffmpeg -f lavfi -i color -vf curves=cross_process:plot=/tmp/curves.plt -frames:v 1 -f null -
6067 gnuplot -p /tmp/curves.plt
6073 Video data analysis filter.
6075 This filter shows hexadecimal pixel values of part of video.
6077 The filter accepts the following options:
6081 Set output video size.
6084 Set x offset from where to pick pixels.
6087 Set y offset from where to pick pixels.
6090 Set scope mode, can be one of the following:
6093 Draw hexadecimal pixel values with white color on black background.
6096 Draw hexadecimal pixel values with input video pixel color on black
6100 Draw hexadecimal pixel values on color background picked from input video,
6101 the text color is picked in such way so its always visible.
6105 Draw rows and columns numbers on left and top of video.
6108 Set background opacity.
6113 Denoise frames using 2D DCT (frequency domain filtering).
6115 This filter is not designed for real time.
6117 The filter accepts the following options:
6121 Set the noise sigma constant.
6123 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
6124 coefficient (absolute value) below this threshold with be dropped.
6126 If you need a more advanced filtering, see @option{expr}.
6128 Default is @code{0}.
6131 Set number overlapping pixels for each block. Since the filter can be slow, you
6132 may want to reduce this value, at the cost of a less effective filter and the
6133 risk of various artefacts.
6135 If the overlapping value doesn't permit processing the whole input width or
6136 height, a warning will be displayed and according borders won't be denoised.
6138 Default value is @var{blocksize}-1, which is the best possible setting.
6141 Set the coefficient factor expression.
6143 For each coefficient of a DCT block, this expression will be evaluated as a
6144 multiplier value for the coefficient.
6146 If this is option is set, the @option{sigma} option will be ignored.
6148 The absolute value of the coefficient can be accessed through the @var{c}
6152 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
6153 @var{blocksize}, which is the width and height of the processed blocks.
6155 The default value is @var{3} (8x8) and can be raised to @var{4} for a
6156 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
6157 on the speed processing. Also, a larger block size does not necessarily means a
6161 @subsection Examples
6163 Apply a denoise with a @option{sigma} of @code{4.5}:
6168 The same operation can be achieved using the expression system:
6170 dctdnoiz=e='gte(c, 4.5*3)'
6173 Violent denoise using a block size of @code{16x16}:
6180 Remove banding artifacts from input video.
6181 It works by replacing banded pixels with average value of referenced pixels.
6183 The filter accepts the following options:
6190 Set banding detection threshold for each plane. Default is 0.02.
6191 Valid range is 0.00003 to 0.5.
6192 If difference between current pixel and reference pixel is less than threshold,
6193 it will be considered as banded.
6196 Banding detection range in pixels. Default is 16. If positive, random number
6197 in range 0 to set value will be used. If negative, exact absolute value
6199 The range defines square of four pixels around current pixel.
6202 Set direction in radians from which four pixel will be compared. If positive,
6203 random direction from 0 to set direction will be picked. If negative, exact of
6204 absolute value will be picked. For example direction 0, -PI or -2*PI radians
6205 will pick only pixels on same row and -PI/2 will pick only pixels on same
6209 If enabled, current pixel is compared with average value of all four
6210 surrounding pixels. The default is enabled. If disabled current pixel is
6211 compared with all four surrounding pixels. The pixel is considered banded
6212 if only all four differences with surrounding pixels are less than threshold.
6215 If enabled, current pixel is changed if and only if all pixel components are banded,
6216 e.g. banding detection threshold is triggered for all color components.
6217 The default is disabled.
6223 Drop duplicated frames at regular intervals.
6225 The filter accepts the following options:
6229 Set the number of frames from which one will be dropped. Setting this to
6230 @var{N} means one frame in every batch of @var{N} frames will be dropped.
6231 Default is @code{5}.
6234 Set the threshold for duplicate detection. If the difference metric for a frame
6235 is less than or equal to this value, then it is declared as duplicate. Default
6239 Set scene change threshold. Default is @code{15}.
6243 Set the size of the x and y-axis blocks used during metric calculations.
6244 Larger blocks give better noise suppression, but also give worse detection of
6245 small movements. Must be a power of two. Default is @code{32}.
6248 Mark main input as a pre-processed input and activate clean source input
6249 stream. This allows the input to be pre-processed with various filters to help
6250 the metrics calculation while keeping the frame selection lossless. When set to
6251 @code{1}, the first stream is for the pre-processed input, and the second
6252 stream is the clean source from where the kept frames are chosen. Default is
6256 Set whether or not chroma is considered in the metric calculations. Default is
6262 Apply deflate effect to the video.
6264 This filter replaces the pixel by the local(3x3) average by taking into account
6265 only values lower than the pixel.
6267 It accepts the following options:
6274 Limit the maximum change for each plane, default is 65535.
6275 If 0, plane will remain unchanged.
6280 Remove judder produced by partially interlaced telecined content.
6282 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
6283 source was partially telecined content then the output of @code{pullup,dejudder}
6284 will have a variable frame rate. May change the recorded frame rate of the
6285 container. Aside from that change, this filter will not affect constant frame
6288 The option available in this filter is:
6292 Specify the length of the window over which the judder repeats.
6294 Accepts any integer greater than 1. Useful values are:
6298 If the original was telecined from 24 to 30 fps (Film to NTSC).
6301 If the original was telecined from 25 to 30 fps (PAL to NTSC).
6304 If a mixture of the two.
6307 The default is @samp{4}.
6312 Suppress a TV station logo by a simple interpolation of the surrounding
6313 pixels. Just set a rectangle covering the logo and watch it disappear
6314 (and sometimes something even uglier appear - your mileage may vary).
6316 It accepts the following parameters:
6321 Specify the top left corner coordinates of the logo. They must be
6326 Specify the width and height of the logo to clear. They must be
6330 Specify the thickness of the fuzzy edge of the rectangle (added to
6331 @var{w} and @var{h}). The default value is 1. This option is
6332 deprecated, setting higher values should no longer be necessary and
6336 When set to 1, a green rectangle is drawn on the screen to simplify
6337 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
6338 The default value is 0.
6340 The rectangle is drawn on the outermost pixels which will be (partly)
6341 replaced with interpolated values. The values of the next pixels
6342 immediately outside this rectangle in each direction will be used to
6343 compute the interpolated pixel values inside the rectangle.
6347 @subsection Examples
6351 Set a rectangle covering the area with top left corner coordinates 0,0
6352 and size 100x77, and a band of size 10:
6354 delogo=x=0:y=0:w=100:h=77:band=10
6361 Attempt to fix small changes in horizontal and/or vertical shift. This
6362 filter helps remove camera shake from hand-holding a camera, bumping a
6363 tripod, moving on a vehicle, etc.
6365 The filter accepts the following options:
6373 Specify a rectangular area where to limit the search for motion
6375 If desired the search for motion vectors can be limited to a
6376 rectangular area of the frame defined by its top left corner, width
6377 and height. These parameters have the same meaning as the drawbox
6378 filter which can be used to visualise the position of the bounding
6381 This is useful when simultaneous movement of subjects within the frame
6382 might be confused for camera motion by the motion vector search.
6384 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
6385 then the full frame is used. This allows later options to be set
6386 without specifying the bounding box for the motion vector search.
6388 Default - search the whole frame.
6392 Specify the maximum extent of movement in x and y directions in the
6393 range 0-64 pixels. Default 16.
6396 Specify how to generate pixels to fill blanks at the edge of the
6397 frame. Available values are:
6400 Fill zeroes at blank locations
6402 Original image at blank locations
6404 Extruded edge value at blank locations
6406 Mirrored edge at blank locations
6408 Default value is @samp{mirror}.
6411 Specify the blocksize to use for motion search. Range 4-128 pixels,
6415 Specify the contrast threshold for blocks. Only blocks with more than
6416 the specified contrast (difference between darkest and lightest
6417 pixels) will be considered. Range 1-255, default 125.
6420 Specify the search strategy. Available values are:
6423 Set exhaustive search
6425 Set less exhaustive search.
6427 Default value is @samp{exhaustive}.
6430 If set then a detailed log of the motion search is written to the
6434 If set to 1, specify using OpenCL capabilities, only available if
6435 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
6441 Apply an exact inverse of the telecine operation. It requires a predefined
6442 pattern specified using the pattern option which must be the same as that passed
6443 to the telecine filter.
6445 This filter accepts the following options:
6454 The default value is @code{top}.
6458 A string of numbers representing the pulldown pattern you wish to apply.
6459 The default value is @code{23}.
6462 A number representing position of the first frame with respect to the telecine
6463 pattern. This is to be used if the stream is cut. The default value is @code{0}.
6468 Apply dilation effect to the video.
6470 This filter replaces the pixel by the local(3x3) maximum.
6472 It accepts the following options:
6479 Limit the maximum change for each plane, default is 65535.
6480 If 0, plane will remain unchanged.
6483 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6486 Flags to local 3x3 coordinates maps like this:
6495 Displace pixels as indicated by second and third input stream.
6497 It takes three input streams and outputs one stream, the first input is the
6498 source, and second and third input are displacement maps.
6500 The second input specifies how much to displace pixels along the
6501 x-axis, while the third input specifies how much to displace pixels
6503 If one of displacement map streams terminates, last frame from that
6504 displacement map will be used.
6506 Note that once generated, displacements maps can be reused over and over again.
6508 A description of the accepted options follows.
6512 Set displace behavior for pixels that are out of range.
6514 Available values are:
6517 Missing pixels are replaced by black pixels.
6520 Adjacent pixels will spread out to replace missing pixels.
6523 Out of range pixels are wrapped so they point to pixels of other side.
6525 Default is @samp{smear}.
6529 @subsection Examples
6533 Add ripple effect to rgb input of video size hd720:
6535 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
6539 Add wave effect to rgb input of video size hd720:
6541 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
6547 Draw a colored box on the input image.
6549 It accepts the following parameters:
6554 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
6558 The expressions which specify the width and height of the box; if 0 they are interpreted as
6559 the input width and height. It defaults to 0.
6562 Specify the color of the box to write. For the general syntax of this option,
6563 check the "Color" section in the ffmpeg-utils manual. If the special
6564 value @code{invert} is used, the box edge color is the same as the
6565 video with inverted luma.
6568 The expression which sets the thickness of the box edge. Default value is @code{3}.
6570 See below for the list of accepted constants.
6573 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6574 following constants:
6578 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6582 horizontal and vertical chroma subsample values. For example for the
6583 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6587 The input width and height.
6590 The input sample aspect ratio.
6594 The x and y offset coordinates where the box is drawn.
6598 The width and height of the drawn box.
6601 The thickness of the drawn box.
6603 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6604 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6608 @subsection Examples
6612 Draw a black box around the edge of the input image:
6618 Draw a box with color red and an opacity of 50%:
6620 drawbox=10:20:200:60:red@@0.5
6623 The previous example can be specified as:
6625 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
6629 Fill the box with pink color:
6631 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
6635 Draw a 2-pixel red 2.40:1 mask:
6637 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
6643 Draw a grid on the input image.
6645 It accepts the following parameters:
6650 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
6654 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
6655 input width and height, respectively, minus @code{thickness}, so image gets
6656 framed. Default to 0.
6659 Specify the color of the grid. For the general syntax of this option,
6660 check the "Color" section in the ffmpeg-utils manual. If the special
6661 value @code{invert} is used, the grid color is the same as the
6662 video with inverted luma.
6665 The expression which sets the thickness of the grid line. Default value is @code{1}.
6667 See below for the list of accepted constants.
6670 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6671 following constants:
6675 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6679 horizontal and vertical chroma subsample values. For example for the
6680 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6684 The input grid cell width and height.
6687 The input sample aspect ratio.
6691 The x and y coordinates of some point of grid intersection (meant to configure offset).
6695 The width and height of the drawn cell.
6698 The thickness of the drawn cell.
6700 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6701 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6705 @subsection Examples
6709 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
6711 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
6715 Draw a white 3x3 grid with an opacity of 50%:
6717 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
6724 Draw a text string or text from a specified file on top of a video, using the
6725 libfreetype library.
6727 To enable compilation of this filter, you need to configure FFmpeg with
6728 @code{--enable-libfreetype}.
6729 To enable default font fallback and the @var{font} option you need to
6730 configure FFmpeg with @code{--enable-libfontconfig}.
6731 To enable the @var{text_shaping} option, you need to configure FFmpeg with
6732 @code{--enable-libfribidi}.
6736 It accepts the following parameters:
6741 Used to draw a box around text using the background color.
6742 The value must be either 1 (enable) or 0 (disable).
6743 The default value of @var{box} is 0.
6746 Set the width of the border to be drawn around the box using @var{boxcolor}.
6747 The default value of @var{boxborderw} is 0.
6750 The color to be used for drawing box around text. For the syntax of this
6751 option, check the "Color" section in the ffmpeg-utils manual.
6753 The default value of @var{boxcolor} is "white".
6756 Set the line spacing in pixels of the border to be drawn around the box using @var{box}.
6757 The default value of @var{line_spacing} is 0.
6760 Set the width of the border to be drawn around the text using @var{bordercolor}.
6761 The default value of @var{borderw} is 0.
6764 Set the color to be used for drawing border around text. For the syntax of this
6765 option, check the "Color" section in the ffmpeg-utils manual.
6767 The default value of @var{bordercolor} is "black".
6770 Select how the @var{text} is expanded. Can be either @code{none},
6771 @code{strftime} (deprecated) or
6772 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
6776 Set a start time for the count. Value is in microseconds. Only applied
6777 in the deprecated strftime expansion mode. To emulate in normal expansion
6778 mode use the @code{pts} function, supplying the start time (in seconds)
6779 as the second argument.
6782 If true, check and fix text coords to avoid clipping.
6785 The color to be used for drawing fonts. For the syntax of this option, check
6786 the "Color" section in the ffmpeg-utils manual.
6788 The default value of @var{fontcolor} is "black".
6790 @item fontcolor_expr
6791 String which is expanded the same way as @var{text} to obtain dynamic
6792 @var{fontcolor} value. By default this option has empty value and is not
6793 processed. When this option is set, it overrides @var{fontcolor} option.
6796 The font family to be used for drawing text. By default Sans.
6799 The font file to be used for drawing text. The path must be included.
6800 This parameter is mandatory if the fontconfig support is disabled.
6803 Draw the text applying alpha blending. The value can
6804 be a number between 0.0 and 1.0.
6805 The expression accepts the same variables @var{x, y} as well.
6806 The default value is 1.
6807 Please see @var{fontcolor_expr}.
6810 The font size to be used for drawing text.
6811 The default value of @var{fontsize} is 16.
6814 If set to 1, attempt to shape the text (for example, reverse the order of
6815 right-to-left text and join Arabic characters) before drawing it.
6816 Otherwise, just draw the text exactly as given.
6817 By default 1 (if supported).
6820 The flags to be used for loading the fonts.
6822 The flags map the corresponding flags supported by libfreetype, and are
6823 a combination of the following values:
6830 @item vertical_layout
6831 @item force_autohint
6834 @item ignore_global_advance_width
6836 @item ignore_transform
6842 Default value is "default".
6844 For more information consult the documentation for the FT_LOAD_*
6848 The color to be used for drawing a shadow behind the drawn text. For the
6849 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
6851 The default value of @var{shadowcolor} is "black".
6855 The x and y offsets for the text shadow position with respect to the
6856 position of the text. They can be either positive or negative
6857 values. The default value for both is "0".
6860 The starting frame number for the n/frame_num variable. The default value
6864 The size in number of spaces to use for rendering the tab.
6868 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
6869 format. It can be used with or without text parameter. @var{timecode_rate}
6870 option must be specified.
6872 @item timecode_rate, rate, r
6873 Set the timecode frame rate (timecode only).
6876 If set to 1, the output of the timecode option will wrap around at 24 hours.
6877 Default is 0 (disabled).
6880 The text string to be drawn. The text must be a sequence of UTF-8
6882 This parameter is mandatory if no file is specified with the parameter
6886 A text file containing text to be drawn. The text must be a sequence
6887 of UTF-8 encoded characters.
6889 This parameter is mandatory if no text string is specified with the
6890 parameter @var{text}.
6892 If both @var{text} and @var{textfile} are specified, an error is thrown.
6895 If set to 1, the @var{textfile} will be reloaded before each frame.
6896 Be sure to update it atomically, or it may be read partially, or even fail.
6900 The expressions which specify the offsets where text will be drawn
6901 within the video frame. They are relative to the top/left border of the
6904 The default value of @var{x} and @var{y} is "0".
6906 See below for the list of accepted constants and functions.
6909 The parameters for @var{x} and @var{y} are expressions containing the
6910 following constants and functions:
6914 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
6918 horizontal and vertical chroma subsample values. For example for the
6919 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6922 the height of each text line
6930 @item max_glyph_a, ascent
6931 the maximum distance from the baseline to the highest/upper grid
6932 coordinate used to place a glyph outline point, for all the rendered
6934 It is a positive value, due to the grid's orientation with the Y axis
6937 @item max_glyph_d, descent
6938 the maximum distance from the baseline to the lowest grid coordinate
6939 used to place a glyph outline point, for all the rendered glyphs.
6940 This is a negative value, due to the grid's orientation, with the Y axis
6944 maximum glyph height, that is the maximum height for all the glyphs
6945 contained in the rendered text, it is equivalent to @var{ascent} -
6949 maximum glyph width, that is the maximum width for all the glyphs
6950 contained in the rendered text
6953 the number of input frame, starting from 0
6955 @item rand(min, max)
6956 return a random number included between @var{min} and @var{max}
6959 The input sample aspect ratio.
6962 timestamp expressed in seconds, NAN if the input timestamp is unknown
6965 the height of the rendered text
6968 the width of the rendered text
6972 the x and y offset coordinates where the text is drawn.
6974 These parameters allow the @var{x} and @var{y} expressions to refer
6975 each other, so you can for example specify @code{y=x/dar}.
6978 @anchor{drawtext_expansion}
6979 @subsection Text expansion
6981 If @option{expansion} is set to @code{strftime},
6982 the filter recognizes strftime() sequences in the provided text and
6983 expands them accordingly. Check the documentation of strftime(). This
6984 feature is deprecated.
6986 If @option{expansion} is set to @code{none}, the text is printed verbatim.
6988 If @option{expansion} is set to @code{normal} (which is the default),
6989 the following expansion mechanism is used.
6991 The backslash character @samp{\}, followed by any character, always expands to
6992 the second character.
6994 Sequences of the form @code{%@{...@}} are expanded. The text between the
6995 braces is a function name, possibly followed by arguments separated by ':'.
6996 If the arguments contain special characters or delimiters (':' or '@}'),
6997 they should be escaped.
6999 Note that they probably must also be escaped as the value for the
7000 @option{text} option in the filter argument string and as the filter
7001 argument in the filtergraph description, and possibly also for the shell,
7002 that makes up to four levels of escaping; using a text file avoids these
7005 The following functions are available:
7010 The expression evaluation result.
7012 It must take one argument specifying the expression to be evaluated,
7013 which accepts the same constants and functions as the @var{x} and
7014 @var{y} values. Note that not all constants should be used, for
7015 example the text size is not known when evaluating the expression, so
7016 the constants @var{text_w} and @var{text_h} will have an undefined
7019 @item expr_int_format, eif
7020 Evaluate the expression's value and output as formatted integer.
7022 The first argument is the expression to be evaluated, just as for the @var{expr} function.
7023 The second argument specifies the output format. Allowed values are @samp{x},
7024 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
7025 @code{printf} function.
7026 The third parameter is optional and sets the number of positions taken by the output.
7027 It can be used to add padding with zeros from the left.
7030 The time at which the filter is running, expressed in UTC.
7031 It can accept an argument: a strftime() format string.
7034 The time at which the filter is running, expressed in the local time zone.
7035 It can accept an argument: a strftime() format string.
7038 Frame metadata. Takes one or two arguments.
7040 The first argument is mandatory and specifies the metadata key.
7042 The second argument is optional and specifies a default value, used when the
7043 metadata key is not found or empty.
7046 The frame number, starting from 0.
7049 A 1 character description of the current picture type.
7052 The timestamp of the current frame.
7053 It can take up to three arguments.
7055 The first argument is the format of the timestamp; it defaults to @code{flt}
7056 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
7057 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
7058 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
7059 @code{localtime} stands for the timestamp of the frame formatted as
7060 local time zone time.
7062 The second argument is an offset added to the timestamp.
7064 If the format is set to @code{localtime} or @code{gmtime},
7065 a third argument may be supplied: a strftime() format string.
7066 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
7069 @subsection Examples
7073 Draw "Test Text" with font FreeSerif, using the default values for the
7074 optional parameters.
7077 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
7081 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
7082 and y=50 (counting from the top-left corner of the screen), text is
7083 yellow with a red box around it. Both the text and the box have an
7087 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
7088 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
7091 Note that the double quotes are not necessary if spaces are not used
7092 within the parameter list.
7095 Show the text at the center of the video frame:
7097 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
7101 Show the text at a random position, switching to a new position every 30 seconds:
7103 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)"
7107 Show a text line sliding from right to left in the last row of the video
7108 frame. The file @file{LONG_LINE} is assumed to contain a single line
7111 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
7115 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
7117 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
7121 Draw a single green letter "g", at the center of the input video.
7122 The glyph baseline is placed at half screen height.
7124 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
7128 Show text for 1 second every 3 seconds:
7130 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
7134 Use fontconfig to set the font. Note that the colons need to be escaped.
7136 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
7140 Print the date of a real-time encoding (see strftime(3)):
7142 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
7146 Show text fading in and out (appearing/disappearing):
7149 DS=1.0 # display start
7150 DE=10.0 # display end
7151 FID=1.5 # fade in duration
7152 FOD=5 # fade out duration
7153 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 @}"
7157 Horizontally align multiple separate texts. Note that @option{max_glyph_a}
7158 and the @option{fontsize} value are included in the @option{y} offset.
7160 drawtext=fontfile=FreeSans.ttf:text=DOG:fontsize=24:x=10:y=20+24-max_glyph_a,
7161 drawtext=fontfile=FreeSans.ttf:text=cow:fontsize=24:x=80:y=20+24-max_glyph_a
7166 For more information about libfreetype, check:
7167 @url{http://www.freetype.org/}.
7169 For more information about fontconfig, check:
7170 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
7172 For more information about libfribidi, check:
7173 @url{http://fribidi.org/}.
7177 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
7179 The filter accepts the following options:
7184 Set low and high threshold values used by the Canny thresholding
7187 The high threshold selects the "strong" edge pixels, which are then
7188 connected through 8-connectivity with the "weak" edge pixels selected
7189 by the low threshold.
7191 @var{low} and @var{high} threshold values must be chosen in the range
7192 [0,1], and @var{low} should be lesser or equal to @var{high}.
7194 Default value for @var{low} is @code{20/255}, and default value for @var{high}
7198 Define the drawing mode.
7202 Draw white/gray wires on black background.
7205 Mix the colors to create a paint/cartoon effect.
7208 Default value is @var{wires}.
7211 @subsection Examples
7215 Standard edge detection with custom values for the hysteresis thresholding:
7217 edgedetect=low=0.1:high=0.4
7221 Painting effect without thresholding:
7223 edgedetect=mode=colormix:high=0
7228 Set brightness, contrast, saturation and approximate gamma adjustment.
7230 The filter accepts the following options:
7234 Set the contrast expression. The value must be a float value in range
7235 @code{-2.0} to @code{2.0}. The default value is "1".
7238 Set the brightness expression. The value must be a float value in
7239 range @code{-1.0} to @code{1.0}. The default value is "0".
7242 Set the saturation expression. The value must be a float in
7243 range @code{0.0} to @code{3.0}. The default value is "1".
7246 Set the gamma expression. The value must be a float in range
7247 @code{0.1} to @code{10.0}. The default value is "1".
7250 Set the gamma expression for red. The value must be a float in
7251 range @code{0.1} to @code{10.0}. The default value is "1".
7254 Set the gamma expression for green. The value must be a float in range
7255 @code{0.1} to @code{10.0}. The default value is "1".
7258 Set the gamma expression for blue. The value must be a float in range
7259 @code{0.1} to @code{10.0}. The default value is "1".
7262 Set the gamma weight expression. It can be used to reduce the effect
7263 of a high gamma value on bright image areas, e.g. keep them from
7264 getting overamplified and just plain white. The value must be a float
7265 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
7266 gamma correction all the way down while @code{1.0} leaves it at its
7267 full strength. Default is "1".
7270 Set when the expressions for brightness, contrast, saturation and
7271 gamma expressions are evaluated.
7273 It accepts the following values:
7276 only evaluate expressions once during the filter initialization or
7277 when a command is processed
7280 evaluate expressions for each incoming frame
7283 Default value is @samp{init}.
7286 The expressions accept the following parameters:
7289 frame count of the input frame starting from 0
7292 byte position of the corresponding packet in the input file, NAN if
7296 frame rate of the input video, NAN if the input frame rate is unknown
7299 timestamp expressed in seconds, NAN if the input timestamp is unknown
7302 @subsection Commands
7303 The filter supports the following commands:
7307 Set the contrast expression.
7310 Set the brightness expression.
7313 Set the saturation expression.
7316 Set the gamma expression.
7319 Set the gamma_r expression.
7322 Set gamma_g expression.
7325 Set gamma_b expression.
7328 Set gamma_weight expression.
7330 The command accepts the same syntax of the corresponding option.
7332 If the specified expression is not valid, it is kept at its current
7339 Apply erosion effect to the video.
7341 This filter replaces the pixel by the local(3x3) minimum.
7343 It accepts the following options:
7350 Limit the maximum change for each plane, default is 65535.
7351 If 0, plane will remain unchanged.
7354 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
7357 Flags to local 3x3 coordinates maps like this:
7364 @section extractplanes
7366 Extract color channel components from input video stream into
7367 separate grayscale video streams.
7369 The filter accepts the following option:
7373 Set plane(s) to extract.
7375 Available values for planes are:
7386 Choosing planes not available in the input will result in an error.
7387 That means you cannot select @code{r}, @code{g}, @code{b} planes
7388 with @code{y}, @code{u}, @code{v} planes at same time.
7391 @subsection Examples
7395 Extract luma, u and v color channel component from input video frame
7396 into 3 grayscale outputs:
7398 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
7404 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
7406 For each input image, the filter will compute the optimal mapping from
7407 the input to the output given the codebook length, that is the number
7408 of distinct output colors.
7410 This filter accepts the following options.
7413 @item codebook_length, l
7414 Set codebook length. The value must be a positive integer, and
7415 represents the number of distinct output colors. Default value is 256.
7418 Set the maximum number of iterations to apply for computing the optimal
7419 mapping. The higher the value the better the result and the higher the
7420 computation time. Default value is 1.
7423 Set a random seed, must be an integer included between 0 and
7424 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
7425 will try to use a good random seed on a best effort basis.
7428 Set pal8 output pixel format. This option does not work with codebook
7429 length greater than 256.
7434 Apply a fade-in/out effect to the input video.
7436 It accepts the following parameters:
7440 The effect type can be either "in" for a fade-in, or "out" for a fade-out
7442 Default is @code{in}.
7444 @item start_frame, s
7445 Specify the number of the frame to start applying the fade
7446 effect at. Default is 0.
7449 The number of frames that the fade effect lasts. At the end of the
7450 fade-in effect, the output video will have the same intensity as the input video.
7451 At the end of the fade-out transition, the output video will be filled with the
7452 selected @option{color}.
7456 If set to 1, fade only alpha channel, if one exists on the input.
7459 @item start_time, st
7460 Specify the timestamp (in seconds) of the frame to start to apply the fade
7461 effect. If both start_frame and start_time are specified, the fade will start at
7462 whichever comes last. Default is 0.
7465 The number of seconds for which the fade effect has to last. At the end of the
7466 fade-in effect the output video will have the same intensity as the input video,
7467 at the end of the fade-out transition the output video will be filled with the
7468 selected @option{color}.
7469 If both duration and nb_frames are specified, duration is used. Default is 0
7470 (nb_frames is used by default).
7473 Specify the color of the fade. Default is "black".
7476 @subsection Examples
7480 Fade in the first 30 frames of video:
7485 The command above is equivalent to:
7491 Fade out the last 45 frames of a 200-frame video:
7494 fade=type=out:start_frame=155:nb_frames=45
7498 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
7500 fade=in:0:25, fade=out:975:25
7504 Make the first 5 frames yellow, then fade in from frame 5-24:
7506 fade=in:5:20:color=yellow
7510 Fade in alpha over first 25 frames of video:
7512 fade=in:0:25:alpha=1
7516 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
7518 fade=t=in:st=5.5:d=0.5
7524 Apply arbitrary expressions to samples in frequency domain
7528 Adjust the dc value (gain) of the luma plane of the image. The filter
7529 accepts an integer value in range @code{0} to @code{1000}. The default
7530 value is set to @code{0}.
7533 Adjust the dc value (gain) of the 1st chroma plane of the image. The
7534 filter accepts an integer value in range @code{0} to @code{1000}. The
7535 default value is set to @code{0}.
7538 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
7539 filter accepts an integer value in range @code{0} to @code{1000}. The
7540 default value is set to @code{0}.
7543 Set the frequency domain weight expression for the luma plane.
7546 Set the frequency domain weight expression for the 1st chroma plane.
7549 Set the frequency domain weight expression for the 2nd chroma plane.
7551 The filter accepts the following variables:
7554 The coordinates of the current sample.
7558 The width and height of the image.
7561 @subsection Examples
7567 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
7573 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
7579 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
7585 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
7592 Extract a single field from an interlaced image using stride
7593 arithmetic to avoid wasting CPU time. The output frames are marked as
7596 The filter accepts the following options:
7600 Specify whether to extract the top (if the value is @code{0} or
7601 @code{top}) or the bottom field (if the value is @code{1} or
7607 Create new frames by copying the top and bottom fields from surrounding frames
7608 supplied as numbers by the hint file.
7612 Set file containing hints: absolute/relative frame numbers.
7614 There must be one line for each frame in a clip. Each line must contain two
7615 numbers separated by the comma, optionally followed by @code{-} or @code{+}.
7616 Numbers supplied on each line of file can not be out of [N-1,N+1] where N
7617 is current frame number for @code{absolute} mode or out of [-1, 1] range
7618 for @code{relative} mode. First number tells from which frame to pick up top
7619 field and second number tells from which frame to pick up bottom field.
7621 If optionally followed by @code{+} output frame will be marked as interlaced,
7622 else if followed by @code{-} output frame will be marked as progressive, else
7623 it will be marked same as input frame.
7624 If line starts with @code{#} or @code{;} that line is skipped.
7627 Can be item @code{absolute} or @code{relative}. Default is @code{absolute}.
7630 Example of first several lines of @code{hint} file for @code{relative} mode:
7633 1,0 - # second frame, use third's frame top field and second's frame bottom field
7634 1,0 - # third frame, use fourth's frame top field and third's frame bottom field
7651 Field matching filter for inverse telecine. It is meant to reconstruct the
7652 progressive frames from a telecined stream. The filter does not drop duplicated
7653 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
7654 followed by a decimation filter such as @ref{decimate} in the filtergraph.
7656 The separation of the field matching and the decimation is notably motivated by
7657 the possibility of inserting a de-interlacing filter fallback between the two.
7658 If the source has mixed telecined and real interlaced content,
7659 @code{fieldmatch} will not be able to match fields for the interlaced parts.
7660 But these remaining combed frames will be marked as interlaced, and thus can be
7661 de-interlaced by a later filter such as @ref{yadif} before decimation.
7663 In addition to the various configuration options, @code{fieldmatch} can take an
7664 optional second stream, activated through the @option{ppsrc} option. If
7665 enabled, the frames reconstruction will be based on the fields and frames from
7666 this second stream. This allows the first input to be pre-processed in order to
7667 help the various algorithms of the filter, while keeping the output lossless
7668 (assuming the fields are matched properly). Typically, a field-aware denoiser,
7669 or brightness/contrast adjustments can help.
7671 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
7672 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
7673 which @code{fieldmatch} is based on. While the semantic and usage are very
7674 close, some behaviour and options names can differ.
7676 The @ref{decimate} filter currently only works for constant frame rate input.
7677 If your input has mixed telecined (30fps) and progressive content with a lower
7678 framerate like 24fps use the following filterchain to produce the necessary cfr
7679 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
7681 The filter accepts the following options:
7685 Specify the assumed field order of the input stream. Available values are:
7689 Auto detect parity (use FFmpeg's internal parity value).
7691 Assume bottom field first.
7693 Assume top field first.
7696 Note that it is sometimes recommended not to trust the parity announced by the
7699 Default value is @var{auto}.
7702 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
7703 sense that it won't risk creating jerkiness due to duplicate frames when
7704 possible, but if there are bad edits or blended fields it will end up
7705 outputting combed frames when a good match might actually exist. On the other
7706 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
7707 but will almost always find a good frame if there is one. The other values are
7708 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
7709 jerkiness and creating duplicate frames versus finding good matches in sections
7710 with bad edits, orphaned fields, blended fields, etc.
7712 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
7714 Available values are:
7718 2-way matching (p/c)
7720 2-way matching, and trying 3rd match if still combed (p/c + n)
7722 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
7724 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
7725 still combed (p/c + n + u/b)
7727 3-way matching (p/c/n)
7729 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
7730 detected as combed (p/c/n + u/b)
7733 The parenthesis at the end indicate the matches that would be used for that
7734 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
7737 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
7740 Default value is @var{pc_n}.
7743 Mark the main input stream as a pre-processed input, and enable the secondary
7744 input stream as the clean source to pick the fields from. See the filter
7745 introduction for more details. It is similar to the @option{clip2} feature from
7748 Default value is @code{0} (disabled).
7751 Set the field to match from. It is recommended to set this to the same value as
7752 @option{order} unless you experience matching failures with that setting. In
7753 certain circumstances changing the field that is used to match from can have a
7754 large impact on matching performance. Available values are:
7758 Automatic (same value as @option{order}).
7760 Match from the bottom field.
7762 Match from the top field.
7765 Default value is @var{auto}.
7768 Set whether or not chroma is included during the match comparisons. In most
7769 cases it is recommended to leave this enabled. You should set this to @code{0}
7770 only if your clip has bad chroma problems such as heavy rainbowing or other
7771 artifacts. Setting this to @code{0} could also be used to speed things up at
7772 the cost of some accuracy.
7774 Default value is @code{1}.
7778 These define an exclusion band which excludes the lines between @option{y0} and
7779 @option{y1} from being included in the field matching decision. An exclusion
7780 band can be used to ignore subtitles, a logo, or other things that may
7781 interfere with the matching. @option{y0} sets the starting scan line and
7782 @option{y1} sets the ending line; all lines in between @option{y0} and
7783 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
7784 @option{y0} and @option{y1} to the same value will disable the feature.
7785 @option{y0} and @option{y1} defaults to @code{0}.
7788 Set the scene change detection threshold as a percentage of maximum change on
7789 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
7790 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
7791 @option{scthresh} is @code{[0.0, 100.0]}.
7793 Default value is @code{12.0}.
7796 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
7797 account the combed scores of matches when deciding what match to use as the
7798 final match. Available values are:
7802 No final matching based on combed scores.
7804 Combed scores are only used when a scene change is detected.
7806 Use combed scores all the time.
7809 Default is @var{sc}.
7812 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
7813 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
7814 Available values are:
7818 No forced calculation.
7820 Force p/c/n calculations.
7822 Force p/c/n/u/b calculations.
7825 Default value is @var{none}.
7828 This is the area combing threshold used for combed frame detection. This
7829 essentially controls how "strong" or "visible" combing must be to be detected.
7830 Larger values mean combing must be more visible and smaller values mean combing
7831 can be less visible or strong and still be detected. Valid settings are from
7832 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
7833 be detected as combed). This is basically a pixel difference value. A good
7834 range is @code{[8, 12]}.
7836 Default value is @code{9}.
7839 Sets whether or not chroma is considered in the combed frame decision. Only
7840 disable this if your source has chroma problems (rainbowing, etc.) that are
7841 causing problems for the combed frame detection with chroma enabled. Actually,
7842 using @option{chroma}=@var{0} is usually more reliable, except for the case
7843 where there is chroma only combing in the source.
7845 Default value is @code{0}.
7849 Respectively set the x-axis and y-axis size of the window used during combed
7850 frame detection. This has to do with the size of the area in which
7851 @option{combpel} pixels are required to be detected as combed for a frame to be
7852 declared combed. See the @option{combpel} parameter description for more info.
7853 Possible values are any number that is a power of 2 starting at 4 and going up
7856 Default value is @code{16}.
7859 The number of combed pixels inside any of the @option{blocky} by
7860 @option{blockx} size blocks on the frame for the frame to be detected as
7861 combed. While @option{cthresh} controls how "visible" the combing must be, this
7862 setting controls "how much" combing there must be in any localized area (a
7863 window defined by the @option{blockx} and @option{blocky} settings) on the
7864 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
7865 which point no frames will ever be detected as combed). This setting is known
7866 as @option{MI} in TFM/VFM vocabulary.
7868 Default value is @code{80}.
7871 @anchor{p/c/n/u/b meaning}
7872 @subsection p/c/n/u/b meaning
7874 @subsubsection p/c/n
7876 We assume the following telecined stream:
7879 Top fields: 1 2 2 3 4
7880 Bottom fields: 1 2 3 4 4
7883 The numbers correspond to the progressive frame the fields relate to. Here, the
7884 first two frames are progressive, the 3rd and 4th are combed, and so on.
7886 When @code{fieldmatch} is configured to run a matching from bottom
7887 (@option{field}=@var{bottom}) this is how this input stream get transformed:
7892 B 1 2 3 4 4 <-- matching reference
7901 As a result of the field matching, we can see that some frames get duplicated.
7902 To perform a complete inverse telecine, you need to rely on a decimation filter
7903 after this operation. See for instance the @ref{decimate} filter.
7905 The same operation now matching from top fields (@option{field}=@var{top})
7910 T 1 2 2 3 4 <-- matching reference
7920 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
7921 basically, they refer to the frame and field of the opposite parity:
7924 @item @var{p} matches the field of the opposite parity in the previous frame
7925 @item @var{c} matches the field of the opposite parity in the current frame
7926 @item @var{n} matches the field of the opposite parity in the next frame
7931 The @var{u} and @var{b} matching are a bit special in the sense that they match
7932 from the opposite parity flag. In the following examples, we assume that we are
7933 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
7934 'x' is placed above and below each matched fields.
7936 With bottom matching (@option{field}=@var{bottom}):
7941 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7942 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7950 With top matching (@option{field}=@var{top}):
7955 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7956 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7964 @subsection Examples
7966 Simple IVTC of a top field first telecined stream:
7968 fieldmatch=order=tff:combmatch=none, decimate
7971 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
7973 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
7978 Transform the field order of the input video.
7980 It accepts the following parameters:
7985 The output field order. Valid values are @var{tff} for top field first or @var{bff}
7986 for bottom field first.
7989 The default value is @samp{tff}.
7991 The transformation is done by shifting the picture content up or down
7992 by one line, and filling the remaining line with appropriate picture content.
7993 This method is consistent with most broadcast field order converters.
7995 If the input video is not flagged as being interlaced, or it is already
7996 flagged as being of the required output field order, then this filter does
7997 not alter the incoming video.
7999 It is very useful when converting to or from PAL DV material,
8000 which is bottom field first.
8004 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
8007 @section fifo, afifo
8009 Buffer input images and send them when they are requested.
8011 It is mainly useful when auto-inserted by the libavfilter
8014 It does not take parameters.
8018 Find a rectangular object
8020 It accepts the following options:
8024 Filepath of the object image, needs to be in gray8.
8027 Detection threshold, default is 0.5.
8030 Number of mipmaps, default is 3.
8032 @item xmin, ymin, xmax, ymax
8033 Specifies the rectangle in which to search.
8036 @subsection Examples
8040 Generate a representative palette of a given video using @command{ffmpeg}:
8042 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8048 Cover a rectangular object
8050 It accepts the following options:
8054 Filepath of the optional cover image, needs to be in yuv420.
8059 It accepts the following values:
8062 cover it by the supplied image
8064 cover it by interpolating the surrounding pixels
8067 Default value is @var{blur}.
8070 @subsection Examples
8074 Generate a representative palette of a given video using @command{ffmpeg}:
8076 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8083 Convert the input video to one of the specified pixel formats.
8084 Libavfilter will try to pick one that is suitable as input to
8087 It accepts the following parameters:
8091 A '|'-separated list of pixel format names, such as
8092 "pix_fmts=yuv420p|monow|rgb24".
8096 @subsection Examples
8100 Convert the input video to the @var{yuv420p} format
8102 format=pix_fmts=yuv420p
8105 Convert the input video to any of the formats in the list
8107 format=pix_fmts=yuv420p|yuv444p|yuv410p
8114 Convert the video to specified constant frame rate by duplicating or dropping
8115 frames as necessary.
8117 It accepts the following parameters:
8121 The desired output frame rate. The default is @code{25}.
8126 Possible values are:
8129 zero round towards 0
8133 round towards -infinity
8135 round towards +infinity
8139 The default is @code{near}.
8142 Assume the first PTS should be the given value, in seconds. This allows for
8143 padding/trimming at the start of stream. By default, no assumption is made
8144 about the first frame's expected PTS, so no padding or trimming is done.
8145 For example, this could be set to 0 to pad the beginning with duplicates of
8146 the first frame if a video stream starts after the audio stream or to trim any
8147 frames with a negative PTS.
8151 Alternatively, the options can be specified as a flat string:
8152 @var{fps}[:@var{round}].
8154 See also the @ref{setpts} filter.
8156 @subsection Examples
8160 A typical usage in order to set the fps to 25:
8166 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
8168 fps=fps=film:round=near
8174 Pack two different video streams into a stereoscopic video, setting proper
8175 metadata on supported codecs. The two views should have the same size and
8176 framerate and processing will stop when the shorter video ends. Please note
8177 that you may conveniently adjust view properties with the @ref{scale} and
8180 It accepts the following parameters:
8184 The desired packing format. Supported values are:
8189 The views are next to each other (default).
8192 The views are on top of each other.
8195 The views are packed by line.
8198 The views are packed by column.
8201 The views are temporally interleaved.
8210 # Convert left and right views into a frame-sequential video
8211 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
8213 # Convert views into a side-by-side video with the same output resolution as the input
8214 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
8219 Change the frame rate by interpolating new video output frames from the source
8222 This filter is not designed to function correctly with interlaced media. If
8223 you wish to change the frame rate of interlaced media then you are required
8224 to deinterlace before this filter and re-interlace after this filter.
8226 A description of the accepted options follows.
8230 Specify the output frames per second. This option can also be specified
8231 as a value alone. The default is @code{50}.
8234 Specify the start of a range where the output frame will be created as a
8235 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8236 the default is @code{15}.
8239 Specify the end of a range where the output frame will be created as a
8240 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8241 the default is @code{240}.
8244 Specify the level at which a scene change is detected as a value between
8245 0 and 100 to indicate a new scene; a low value reflects a low
8246 probability for the current frame to introduce a new scene, while a higher
8247 value means the current frame is more likely to be one.
8248 The default is @code{7}.
8251 Specify flags influencing the filter process.
8253 Available value for @var{flags} is:
8256 @item scene_change_detect, scd
8257 Enable scene change detection using the value of the option @var{scene}.
8258 This flag is enabled by default.
8264 Select one frame every N-th frame.
8266 This filter accepts the following option:
8269 Select frame after every @code{step} frames.
8270 Allowed values are positive integers higher than 0. Default value is @code{1}.
8276 Apply a frei0r effect to the input video.
8278 To enable the compilation of this filter, you need to install the frei0r
8279 header and configure FFmpeg with @code{--enable-frei0r}.
8281 It accepts the following parameters:
8286 The name of the frei0r effect to load. If the environment variable
8287 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
8288 directories specified by the colon-separated list in @env{FREIOR_PATH}.
8289 Otherwise, the standard frei0r paths are searched, in this order:
8290 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
8291 @file{/usr/lib/frei0r-1/}.
8294 A '|'-separated list of parameters to pass to the frei0r effect.
8298 A frei0r effect parameter can be a boolean (its value is either
8299 "y" or "n"), a double, a color (specified as
8300 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
8301 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
8302 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
8303 @var{X} and @var{Y} are floating point numbers) and/or a string.
8305 The number and types of parameters depend on the loaded effect. If an
8306 effect parameter is not specified, the default value is set.
8308 @subsection Examples
8312 Apply the distort0r effect, setting the first two double parameters:
8314 frei0r=filter_name=distort0r:filter_params=0.5|0.01
8318 Apply the colordistance effect, taking a color as the first parameter:
8320 frei0r=colordistance:0.2/0.3/0.4
8321 frei0r=colordistance:violet
8322 frei0r=colordistance:0x112233
8326 Apply the perspective effect, specifying the top left and top right image
8329 frei0r=perspective:0.2/0.2|0.8/0.2
8333 For more information, see
8334 @url{http://frei0r.dyne.org}
8338 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
8340 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
8341 processing filter, one of them is performed once per block, not per pixel.
8342 This allows for much higher speed.
8344 The filter accepts the following options:
8348 Set quality. This option defines the number of levels for averaging. It accepts
8349 an integer in the range 4-5. Default value is @code{4}.
8352 Force a constant quantization parameter. It accepts an integer in range 0-63.
8353 If not set, the filter will use the QP from the video stream (if available).
8356 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
8357 more details but also more artifacts, while higher values make the image smoother
8358 but also blurrier. Default value is @code{0} − PSNR optimal.
8361 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8362 option may cause flicker since the B-Frames have often larger QP. Default is
8363 @code{0} (not enabled).
8369 Apply Gaussian blur filter.
8371 The filter accepts the following options:
8375 Set horizontal sigma, standard deviation of Gaussian blur. Default is @code{0.5}.
8378 Set number of steps for Gaussian approximation. Defauls is @code{1}.
8381 Set which planes to filter. By default all planes are filtered.
8384 Set vertical sigma, if negative it will be same as @code{sigma}.
8385 Default is @code{-1}.
8390 The filter accepts the following options:
8394 Set the luminance expression.
8396 Set the chrominance blue expression.
8398 Set the chrominance red expression.
8400 Set the alpha expression.
8402 Set the red expression.
8404 Set the green expression.
8406 Set the blue expression.
8409 The colorspace is selected according to the specified options. If one
8410 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
8411 options is specified, the filter will automatically select a YCbCr
8412 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
8413 @option{blue_expr} options is specified, it will select an RGB
8416 If one of the chrominance expression is not defined, it falls back on the other
8417 one. If no alpha expression is specified it will evaluate to opaque value.
8418 If none of chrominance expressions are specified, they will evaluate
8419 to the luminance expression.
8421 The expressions can use the following variables and functions:
8425 The sequential number of the filtered frame, starting from @code{0}.
8429 The coordinates of the current sample.
8433 The width and height of the image.
8437 Width and height scale depending on the currently filtered plane. It is the
8438 ratio between the corresponding luma plane number of pixels and the current
8439 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
8440 @code{0.5,0.5} for chroma planes.
8443 Time of the current frame, expressed in seconds.
8446 Return the value of the pixel at location (@var{x},@var{y}) of the current
8450 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
8454 Return the value of the pixel at location (@var{x},@var{y}) of the
8455 blue-difference chroma plane. Return 0 if there is no such plane.
8458 Return the value of the pixel at location (@var{x},@var{y}) of the
8459 red-difference chroma plane. Return 0 if there is no such plane.
8464 Return the value of the pixel at location (@var{x},@var{y}) of the
8465 red/green/blue component. Return 0 if there is no such component.
8468 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
8469 plane. Return 0 if there is no such plane.
8472 For functions, if @var{x} and @var{y} are outside the area, the value will be
8473 automatically clipped to the closer edge.
8475 @subsection Examples
8479 Flip the image horizontally:
8485 Generate a bidimensional sine wave, with angle @code{PI/3} and a
8486 wavelength of 100 pixels:
8488 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
8492 Generate a fancy enigmatic moving light:
8494 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
8498 Generate a quick emboss effect:
8500 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
8504 Modify RGB components depending on pixel position:
8506 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
8510 Create a radial gradient that is the same size as the input (also see
8511 the @ref{vignette} filter):
8513 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
8519 Fix the banding artifacts that are sometimes introduced into nearly flat
8520 regions by truncation to 8-bit color depth.
8521 Interpolate the gradients that should go where the bands are, and
8524 It is designed for playback only. Do not use it prior to
8525 lossy compression, because compression tends to lose the dither and
8526 bring back the bands.
8528 It accepts the following parameters:
8533 The maximum amount by which the filter will change any one pixel. This is also
8534 the threshold for detecting nearly flat regions. Acceptable values range from
8535 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
8539 The neighborhood to fit the gradient to. A larger radius makes for smoother
8540 gradients, but also prevents the filter from modifying the pixels near detailed
8541 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
8542 values will be clipped to the valid range.
8546 Alternatively, the options can be specified as a flat string:
8547 @var{strength}[:@var{radius}]
8549 @subsection Examples
8553 Apply the filter with a @code{3.5} strength and radius of @code{8}:
8559 Specify radius, omitting the strength (which will fall-back to the default
8570 Apply a Hald CLUT to a video stream.
8572 First input is the video stream to process, and second one is the Hald CLUT.
8573 The Hald CLUT input can be a simple picture or a complete video stream.
8575 The filter accepts the following options:
8579 Force termination when the shortest input terminates. Default is @code{0}.
8581 Continue applying the last CLUT after the end of the stream. A value of
8582 @code{0} disable the filter after the last frame of the CLUT is reached.
8583 Default is @code{1}.
8586 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
8587 filters share the same internals).
8589 More information about the Hald CLUT can be found on Eskil Steenberg's website
8590 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
8592 @subsection Workflow examples
8594 @subsubsection Hald CLUT video stream
8596 Generate an identity Hald CLUT stream altered with various effects:
8598 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "hue=H=2*PI*t:s=sin(2*PI*t)+1, curves=cross_process" -t 10 -c:v ffv1 clut.nut
8601 Note: make sure you use a lossless codec.
8603 Then use it with @code{haldclut} to apply it on some random stream:
8605 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
8608 The Hald CLUT will be applied to the 10 first seconds (duration of
8609 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
8610 to the remaining frames of the @code{mandelbrot} stream.
8612 @subsubsection Hald CLUT with preview
8614 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
8615 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
8616 biggest possible square starting at the top left of the picture. The remaining
8617 padding pixels (bottom or right) will be ignored. This area can be used to add
8618 a preview of the Hald CLUT.
8620 Typically, the following generated Hald CLUT will be supported by the
8621 @code{haldclut} filter:
8624 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
8625 pad=iw+320 [padded_clut];
8626 smptebars=s=320x256, split [a][b];
8627 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
8628 [main][b] overlay=W-320" -frames:v 1 clut.png
8631 It contains the original and a preview of the effect of the CLUT: SMPTE color
8632 bars are displayed on the right-top, and below the same color bars processed by
8635 Then, the effect of this Hald CLUT can be visualized with:
8637 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
8642 Flip the input video horizontally.
8644 For example, to horizontally flip the input video with @command{ffmpeg}:
8646 ffmpeg -i in.avi -vf "hflip" out.avi
8650 This filter applies a global color histogram equalization on a
8653 It can be used to correct video that has a compressed range of pixel
8654 intensities. The filter redistributes the pixel intensities to
8655 equalize their distribution across the intensity range. It may be
8656 viewed as an "automatically adjusting contrast filter". This filter is
8657 useful only for correcting degraded or poorly captured source
8660 The filter accepts the following options:
8664 Determine the amount of equalization to be applied. As the strength
8665 is reduced, the distribution of pixel intensities more-and-more
8666 approaches that of the input frame. The value must be a float number
8667 in the range [0,1] and defaults to 0.200.
8670 Set the maximum intensity that can generated and scale the output
8671 values appropriately. The strength should be set as desired and then
8672 the intensity can be limited if needed to avoid washing-out. The value
8673 must be a float number in the range [0,1] and defaults to 0.210.
8676 Set the antibanding level. If enabled the filter will randomly vary
8677 the luminance of output pixels by a small amount to avoid banding of
8678 the histogram. Possible values are @code{none}, @code{weak} or
8679 @code{strong}. It defaults to @code{none}.
8684 Compute and draw a color distribution histogram for the input video.
8686 The computed histogram is a representation of the color component
8687 distribution in an image.
8689 Standard histogram displays the color components distribution in an image.
8690 Displays color graph for each color component. Shows distribution of
8691 the Y, U, V, A or R, G, B components, depending on input format, in the
8692 current frame. Below each graph a color component scale meter is shown.
8694 The filter accepts the following options:
8698 Set height of level. Default value is @code{200}.
8699 Allowed range is [50, 2048].
8702 Set height of color scale. Default value is @code{12}.
8703 Allowed range is [0, 40].
8707 It accepts the following values:
8710 Per color component graphs are placed below each other.
8713 Presents information identical to that in the @code{parade}, except
8714 that the graphs representing color components are superimposed directly
8717 Default is @code{parade}.
8720 Set mode. Can be either @code{linear}, or @code{logarithmic}.
8721 Default is @code{linear}.
8724 Set what color components to display.
8725 Default is @code{7}.
8728 Set foreground opacity. Default is @code{0.7}.
8731 Set background opacity. Default is @code{0.5}.
8734 @subsection Examples
8739 Calculate and draw histogram:
8741 ffplay -i input -vf histogram
8749 This is a high precision/quality 3d denoise filter. It aims to reduce
8750 image noise, producing smooth images and making still images really
8751 still. It should enhance compressibility.
8753 It accepts the following optional parameters:
8757 A non-negative floating point number which specifies spatial luma strength.
8760 @item chroma_spatial
8761 A non-negative floating point number which specifies spatial chroma strength.
8762 It defaults to 3.0*@var{luma_spatial}/4.0.
8765 A floating point number which specifies luma temporal strength. It defaults to
8766 6.0*@var{luma_spatial}/4.0.
8769 A floating point number which specifies chroma temporal strength. It defaults to
8770 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
8773 @anchor{hwupload_cuda}
8774 @section hwupload_cuda
8776 Upload system memory frames to a CUDA device.
8778 It accepts the following optional parameters:
8782 The number of the CUDA device to use
8787 Apply a high-quality magnification filter designed for pixel art. This filter
8788 was originally created by Maxim Stepin.
8790 It accepts the following option:
8794 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
8795 @code{hq3x} and @code{4} for @code{hq4x}.
8796 Default is @code{3}.
8800 Stack input videos horizontally.
8802 All streams must be of same pixel format and of same height.
8804 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
8805 to create same output.
8807 The filter accept the following option:
8811 Set number of input streams. Default is 2.
8814 If set to 1, force the output to terminate when the shortest input
8815 terminates. Default value is 0.
8820 Modify the hue and/or the saturation of the input.
8822 It accepts the following parameters:
8826 Specify the hue angle as a number of degrees. It accepts an expression,
8827 and defaults to "0".
8830 Specify the saturation in the [-10,10] range. It accepts an expression and
8834 Specify the hue angle as a number of radians. It accepts an
8835 expression, and defaults to "0".
8838 Specify the brightness in the [-10,10] range. It accepts an expression and
8842 @option{h} and @option{H} are mutually exclusive, and can't be
8843 specified at the same time.
8845 The @option{b}, @option{h}, @option{H} and @option{s} option values are
8846 expressions containing the following constants:
8850 frame count of the input frame starting from 0
8853 presentation timestamp of the input frame expressed in time base units
8856 frame rate of the input video, NAN if the input frame rate is unknown
8859 timestamp expressed in seconds, NAN if the input timestamp is unknown
8862 time base of the input video
8865 @subsection Examples
8869 Set the hue to 90 degrees and the saturation to 1.0:
8875 Same command but expressing the hue in radians:
8881 Rotate hue and make the saturation swing between 0
8882 and 2 over a period of 1 second:
8884 hue="H=2*PI*t: s=sin(2*PI*t)+1"
8888 Apply a 3 seconds saturation fade-in effect starting at 0:
8893 The general fade-in expression can be written as:
8895 hue="s=min(0\, max((t-START)/DURATION\, 1))"
8899 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
8901 hue="s=max(0\, min(1\, (8-t)/3))"
8904 The general fade-out expression can be written as:
8906 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
8911 @subsection Commands
8913 This filter supports the following commands:
8919 Modify the hue and/or the saturation and/or brightness of the input video.
8920 The command accepts the same syntax of the corresponding option.
8922 If the specified expression is not valid, it is kept at its current
8928 Grow first stream into second stream by connecting components.
8929 This makes it possible to build more robust edge masks.
8931 This filter accepts the following options:
8935 Set which planes will be processed as bitmap, unprocessed planes will be
8936 copied from first stream.
8937 By default value 0xf, all planes will be processed.
8940 Set threshold which is used in filtering. If pixel component value is higher than
8941 this value filter algorithm for connecting components is activated.
8942 By default value is 0.
8947 Detect video interlacing type.
8949 This filter tries to detect if the input frames are interlaced, progressive,
8950 top or bottom field first. It will also try to detect fields that are
8951 repeated between adjacent frames (a sign of telecine).
8953 Single frame detection considers only immediately adjacent frames when classifying each frame.
8954 Multiple frame detection incorporates the classification history of previous frames.
8956 The filter will log these metadata values:
8959 @item single.current_frame
8960 Detected type of current frame using single-frame detection. One of:
8961 ``tff'' (top field first), ``bff'' (bottom field first),
8962 ``progressive'', or ``undetermined''
8965 Cumulative number of frames detected as top field first using single-frame detection.
8968 Cumulative number of frames detected as top field first using multiple-frame detection.
8971 Cumulative number of frames detected as bottom field first using single-frame detection.
8973 @item multiple.current_frame
8974 Detected type of current frame using multiple-frame detection. One of:
8975 ``tff'' (top field first), ``bff'' (bottom field first),
8976 ``progressive'', or ``undetermined''
8979 Cumulative number of frames detected as bottom field first using multiple-frame detection.
8981 @item single.progressive
8982 Cumulative number of frames detected as progressive using single-frame detection.
8984 @item multiple.progressive
8985 Cumulative number of frames detected as progressive using multiple-frame detection.
8987 @item single.undetermined
8988 Cumulative number of frames that could not be classified using single-frame detection.
8990 @item multiple.undetermined
8991 Cumulative number of frames that could not be classified using multiple-frame detection.
8993 @item repeated.current_frame
8994 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
8996 @item repeated.neither
8997 Cumulative number of frames with no repeated field.
9000 Cumulative number of frames with the top field repeated from the previous frame's top field.
9002 @item repeated.bottom
9003 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
9006 The filter accepts the following options:
9010 Set interlacing threshold.
9012 Set progressive threshold.
9014 Threshold for repeated field detection.
9016 Number of frames after which a given frame's contribution to the
9017 statistics is halved (i.e., it contributes only 0.5 to its
9018 classification). The default of 0 means that all frames seen are given
9019 full weight of 1.0 forever.
9020 @item analyze_interlaced_flag
9021 When this is not 0 then idet will use the specified number of frames to determine
9022 if the interlaced flag is accurate, it will not count undetermined frames.
9023 If the flag is found to be accurate it will be used without any further
9024 computations, if it is found to be inaccurate it will be cleared without any
9025 further computations. This allows inserting the idet filter as a low computational
9026 method to clean up the interlaced flag
9031 Deinterleave or interleave fields.
9033 This filter allows one to process interlaced images fields without
9034 deinterlacing them. Deinterleaving splits the input frame into 2
9035 fields (so called half pictures). Odd lines are moved to the top
9036 half of the output image, even lines to the bottom half.
9037 You can process (filter) them independently and then re-interleave them.
9039 The filter accepts the following options:
9043 @item chroma_mode, c
9045 Available values for @var{luma_mode}, @var{chroma_mode} and
9046 @var{alpha_mode} are:
9052 @item deinterleave, d
9053 Deinterleave fields, placing one above the other.
9056 Interleave fields. Reverse the effect of deinterleaving.
9058 Default value is @code{none}.
9061 @item chroma_swap, cs
9062 @item alpha_swap, as
9063 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
9068 Apply inflate effect to the video.
9070 This filter replaces the pixel by the local(3x3) average by taking into account
9071 only values higher than the pixel.
9073 It accepts the following options:
9080 Limit the maximum change for each plane, default is 65535.
9081 If 0, plane will remain unchanged.
9086 Simple interlacing filter from progressive contents. This interleaves upper (or
9087 lower) lines from odd frames with lower (or upper) lines from even frames,
9088 halving the frame rate and preserving image height.
9091 Original Original New Frame
9092 Frame 'j' Frame 'j+1' (tff)
9093 ========== =========== ==================
9094 Line 0 --------------------> Frame 'j' Line 0
9095 Line 1 Line 1 ----> Frame 'j+1' Line 1
9096 Line 2 ---------------------> Frame 'j' Line 2
9097 Line 3 Line 3 ----> Frame 'j+1' Line 3
9099 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
9102 It accepts the following optional parameters:
9106 This determines whether the interlaced frame is taken from the even
9107 (tff - default) or odd (bff) lines of the progressive frame.
9110 Enable (default) or disable the vertical lowpass filter to avoid twitter
9111 interlacing and reduce moire patterns.
9116 Deinterlace input video by applying Donald Graft's adaptive kernel
9117 deinterling. Work on interlaced parts of a video to produce
9120 The description of the accepted parameters follows.
9124 Set the threshold which affects the filter's tolerance when
9125 determining if a pixel line must be processed. It must be an integer
9126 in the range [0,255] and defaults to 10. A value of 0 will result in
9127 applying the process on every pixels.
9130 Paint pixels exceeding the threshold value to white if set to 1.
9134 Set the fields order. Swap fields if set to 1, leave fields alone if
9138 Enable additional sharpening if set to 1. Default is 0.
9141 Enable twoway sharpening if set to 1. Default is 0.
9144 @subsection Examples
9148 Apply default values:
9150 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
9154 Enable additional sharpening:
9160 Paint processed pixels in white:
9166 @section lenscorrection
9168 Correct radial lens distortion
9170 This filter can be used to correct for radial distortion as can result from the use
9171 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
9172 one can use tools available for example as part of opencv or simply trial-and-error.
9173 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
9174 and extract the k1 and k2 coefficients from the resulting matrix.
9176 Note that effectively the same filter is available in the open-source tools Krita and
9177 Digikam from the KDE project.
9179 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
9180 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
9181 brightness distribution, so you may want to use both filters together in certain
9182 cases, though you will have to take care of ordering, i.e. whether vignetting should
9183 be applied before or after lens correction.
9187 The filter accepts the following options:
9191 Relative x-coordinate of the focal point of the image, and thereby the center of the
9192 distortion. This value has a range [0,1] and is expressed as fractions of the image
9195 Relative y-coordinate of the focal point of the image, and thereby the center of the
9196 distortion. This value has a range [0,1] and is expressed as fractions of the image
9199 Coefficient of the quadratic correction term. 0.5 means no correction.
9201 Coefficient of the double quadratic correction term. 0.5 means no correction.
9204 The formula that generates the correction is:
9206 @var{r_src} = @var{r_tgt} * (1 + @var{k1} * (@var{r_tgt} / @var{r_0})^2 + @var{k2} * (@var{r_tgt} / @var{r_0})^4)
9208 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
9209 distances from the focal point in the source and target images, respectively.
9215 The filter accepts the following options:
9219 Set the number of loops.
9222 Set maximal size in number of frames.
9225 Set first frame of loop.
9231 Apply a 3D LUT to an input video.
9233 The filter accepts the following options:
9237 Set the 3D LUT file name.
9239 Currently supported formats:
9251 Select interpolation mode.
9253 Available values are:
9257 Use values from the nearest defined point.
9259 Interpolate values using the 8 points defining a cube.
9261 Interpolate values using a tetrahedron.
9265 @section lut, lutrgb, lutyuv
9267 Compute a look-up table for binding each pixel component input value
9268 to an output value, and apply it to the input video.
9270 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
9271 to an RGB input video.
9273 These filters accept the following parameters:
9276 set first pixel component expression
9278 set second pixel component expression
9280 set third pixel component expression
9282 set fourth pixel component expression, corresponds to the alpha component
9285 set red component expression
9287 set green component expression
9289 set blue component expression
9291 alpha component expression
9294 set Y/luminance component expression
9296 set U/Cb component expression
9298 set V/Cr component expression
9301 Each of them specifies the expression to use for computing the lookup table for
9302 the corresponding pixel component values.
9304 The exact component associated to each of the @var{c*} options depends on the
9307 The @var{lut} filter requires either YUV or RGB pixel formats in input,
9308 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
9310 The expressions can contain the following constants and functions:
9315 The input width and height.
9318 The input value for the pixel component.
9321 The input value, clipped to the @var{minval}-@var{maxval} range.
9324 The maximum value for the pixel component.
9327 The minimum value for the pixel component.
9330 The negated value for the pixel component value, clipped to the
9331 @var{minval}-@var{maxval} range; it corresponds to the expression
9332 "maxval-clipval+minval".
9335 The computed value in @var{val}, clipped to the
9336 @var{minval}-@var{maxval} range.
9338 @item gammaval(gamma)
9339 The computed gamma correction value of the pixel component value,
9340 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
9342 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
9346 All expressions default to "val".
9348 @subsection Examples
9354 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
9355 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
9358 The above is the same as:
9360 lutrgb="r=negval:g=negval:b=negval"
9361 lutyuv="y=negval:u=negval:v=negval"
9371 Remove chroma components, turning the video into a graytone image:
9373 lutyuv="u=128:v=128"
9377 Apply a luma burning effect:
9383 Remove green and blue components:
9389 Set a constant alpha channel value on input:
9391 format=rgba,lutrgb=a="maxval-minval/2"
9395 Correct luminance gamma by a factor of 0.5:
9397 lutyuv=y=gammaval(0.5)
9401 Discard least significant bits of luma:
9403 lutyuv=y='bitand(val, 128+64+32)'
9407 Technicolor like effect:
9409 lutyuv=u='(val-maxval/2)*2+maxval/2':v='(val-maxval/2)*2+maxval/2'
9415 Compute and apply a lookup table from two video inputs.
9417 This filter accepts the following parameters:
9420 set first pixel component expression
9422 set second pixel component expression
9424 set third pixel component expression
9426 set fourth pixel component expression, corresponds to the alpha component
9429 Each of them specifies the expression to use for computing the lookup table for
9430 the corresponding pixel component values.
9432 The exact component associated to each of the @var{c*} options depends on the
9435 The expressions can contain the following constants:
9440 The input width and height.
9443 The first input value for the pixel component.
9446 The second input value for the pixel component.
9449 The first input video bit depth.
9452 The second input video bit depth.
9455 All expressions default to "x".
9457 @subsection Examples
9461 Highlight differences between two RGB video streams:
9463 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)'
9467 Highlight differences between two YUV video streams:
9469 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)'
9473 @section maskedclamp
9475 Clamp the first input stream with the second input and third input stream.
9477 Returns the value of first stream to be between second input
9478 stream - @code{undershoot} and third input stream + @code{overshoot}.
9480 This filter accepts the following options:
9483 Default value is @code{0}.
9486 Default value is @code{0}.
9489 Set which planes will be processed as bitmap, unprocessed planes will be
9490 copied from first stream.
9491 By default value 0xf, all planes will be processed.
9494 @section maskedmerge
9496 Merge the first input stream with the second input stream using per pixel
9497 weights in the third input stream.
9499 A value of 0 in the third stream pixel component means that pixel component
9500 from first stream is returned unchanged, while maximum value (eg. 255 for
9501 8-bit videos) means that pixel component from second stream is returned
9502 unchanged. Intermediate values define the amount of merging between both
9503 input stream's pixel components.
9505 This filter accepts the following options:
9508 Set which planes will be processed as bitmap, unprocessed planes will be
9509 copied from first stream.
9510 By default value 0xf, all planes will be processed.
9515 Apply motion-compensation deinterlacing.
9517 It needs one field per frame as input and must thus be used together
9518 with yadif=1/3 or equivalent.
9520 This filter accepts the following options:
9523 Set the deinterlacing mode.
9525 It accepts one of the following values:
9530 use iterative motion estimation
9532 like @samp{slow}, but use multiple reference frames.
9534 Default value is @samp{fast}.
9537 Set the picture field parity assumed for the input video. It must be
9538 one of the following values:
9542 assume top field first
9544 assume bottom field first
9547 Default value is @samp{bff}.
9550 Set per-block quantization parameter (QP) used by the internal
9553 Higher values should result in a smoother motion vector field but less
9554 optimal individual vectors. Default value is 1.
9557 @section mergeplanes
9559 Merge color channel components from several video streams.
9561 The filter accepts up to 4 input streams, and merge selected input
9562 planes to the output video.
9564 This filter accepts the following options:
9567 Set input to output plane mapping. Default is @code{0}.
9569 The mappings is specified as a bitmap. It should be specified as a
9570 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
9571 mapping for the first plane of the output stream. 'A' sets the number of
9572 the input stream to use (from 0 to 3), and 'a' the plane number of the
9573 corresponding input to use (from 0 to 3). The rest of the mappings is
9574 similar, 'Bb' describes the mapping for the output stream second
9575 plane, 'Cc' describes the mapping for the output stream third plane and
9576 'Dd' describes the mapping for the output stream fourth plane.
9579 Set output pixel format. Default is @code{yuva444p}.
9582 @subsection Examples
9586 Merge three gray video streams of same width and height into single video stream:
9588 [a0][a1][a2]mergeplanes=0x001020:yuv444p
9592 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
9594 [a0][a1]mergeplanes=0x00010210:yuva444p
9598 Swap Y and A plane in yuva444p stream:
9600 format=yuva444p,mergeplanes=0x03010200:yuva444p
9604 Swap U and V plane in yuv420p stream:
9606 format=yuv420p,mergeplanes=0x000201:yuv420p
9610 Cast a rgb24 clip to yuv444p:
9612 format=rgb24,mergeplanes=0x000102:yuv444p
9618 Estimate and export motion vectors using block matching algorithms.
9619 Motion vectors are stored in frame side data to be used by other filters.
9621 This filter accepts the following options:
9624 Specify the motion estimation method. Accepts one of the following values:
9628 Exhaustive search algorithm.
9630 Three step search algorithm.
9632 Two dimensional logarithmic search algorithm.
9634 New three step search algorithm.
9636 Four step search algorithm.
9638 Diamond search algorithm.
9640 Hexagon-based search algorithm.
9642 Enhanced predictive zonal search algorithm.
9644 Uneven multi-hexagon search algorithm.
9646 Default value is @samp{esa}.
9649 Macroblock size. Default @code{16}.
9652 Search parameter. Default @code{7}.
9655 @section midequalizer
9657 Apply Midway Image Equalization effect using two video streams.
9659 Midway Image Equalization adjusts a pair of images to have the same
9660 histogram, while maintaining their dynamics as much as possible. It's
9661 useful for e.g. matching exposures from a pair of stereo cameras.
9663 This filter has two inputs and one output, which must be of same pixel format, but
9664 may be of different sizes. The output of filter is first input adjusted with
9665 midway histogram of both inputs.
9667 This filter accepts the following option:
9671 Set which planes to process. Default is @code{15}, which is all available planes.
9674 @section minterpolate
9676 Convert the video to specified frame rate using motion interpolation.
9678 This filter accepts the following options:
9681 Specify the output frame rate. This can be rational e.g. @code{60000/1001}. Frames are dropped if @var{fps} is lower than source fps. Default @code{60}.
9684 Motion interpolation mode. Following values are accepted:
9687 Duplicate previous or next frame for interpolating new ones.
9689 Blend source frames. Interpolated frame is mean of previous and next frames.
9691 Motion compensated interpolation. Following options are effective when this mode is selected:
9695 Motion compensation mode. Following values are accepted:
9698 Overlapped block motion compensation.
9700 Adaptive overlapped block motion compensation. Window weighting coefficients are controlled adaptively according to the reliabilities of the neighboring motion vectors to reduce oversmoothing.
9702 Default mode is @samp{obmc}.
9705 Motion estimation mode. Following values are accepted:
9708 Bidirectional motion estimation. Motion vectors are estimated for each source frame in both forward and backward directions.
9710 Bilateral motion estimation. Motion vectors are estimated directly for interpolated frame.
9712 Default mode is @samp{bilat}.
9715 The algorithm to be used for motion estimation. Following values are accepted:
9718 Exhaustive search algorithm.
9720 Three step search algorithm.
9722 Two dimensional logarithmic search algorithm.
9724 New three step search algorithm.
9726 Four step search algorithm.
9728 Diamond search algorithm.
9730 Hexagon-based search algorithm.
9732 Enhanced predictive zonal search algorithm.
9734 Uneven multi-hexagon search algorithm.
9736 Default algorithm is @samp{epzs}.
9739 Macroblock size. Default @code{16}.
9742 Motion estimation search parameter. Default @code{32}.
9745 Enable variable-size block motion compensation. Motion estimation is applied with smaller block sizes at object boundaries in order to make the them less blur. Default is @code{0} (disabled).
9750 Scene change detection method. Scene change leads motion vectors to be in random direction. Scene change detection replace interpolated frames by duplicate ones. May not be needed for other modes. Following values are accepted:
9753 Disable scene change detection.
9755 Frame difference. Corresponding pixel values are compared and if it satisfies @var{scd_threshold} scene change is detected.
9757 Default method is @samp{fdiff}.
9760 Scene change detection threshold. Default is @code{5.0}.
9765 Drop frames that do not differ greatly from the previous frame in
9766 order to reduce frame rate.
9768 The main use of this filter is for very-low-bitrate encoding
9769 (e.g. streaming over dialup modem), but it could in theory be used for
9770 fixing movies that were inverse-telecined incorrectly.
9772 A description of the accepted options follows.
9776 Set the maximum number of consecutive frames which can be dropped (if
9777 positive), or the minimum interval between dropped frames (if
9778 negative). If the value is 0, the frame is dropped unregarding the
9779 number of previous sequentially dropped frames.
9786 Set the dropping threshold values.
9788 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
9789 represent actual pixel value differences, so a threshold of 64
9790 corresponds to 1 unit of difference for each pixel, or the same spread
9791 out differently over the block.
9793 A frame is a candidate for dropping if no 8x8 blocks differ by more
9794 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
9795 meaning the whole image) differ by more than a threshold of @option{lo}.
9797 Default value for @option{hi} is 64*12, default value for @option{lo} is
9798 64*5, and default value for @option{frac} is 0.33.
9806 It accepts an integer in input; if non-zero it negates the
9807 alpha component (if available). The default value in input is 0.
9811 Denoise frames using Non-Local Means algorithm.
9813 Each pixel is adjusted by looking for other pixels with similar contexts. This
9814 context similarity is defined by comparing their surrounding patches of size
9815 @option{p}x@option{p}. Patches are searched in an area of @option{r}x@option{r}
9818 Note that the research area defines centers for patches, which means some
9819 patches will be made of pixels outside that research area.
9821 The filter accepts the following options.
9825 Set denoising strength.
9831 Same as @option{p} but for chroma planes.
9833 The default value is @var{0} and means automatic.
9839 Same as @option{r} but for chroma planes.
9841 The default value is @var{0} and means automatic.
9846 Deinterlace video using neural network edge directed interpolation.
9848 This filter accepts the following options:
9852 Mandatory option, without binary file filter can not work.
9853 Currently file can be found here:
9854 https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
9857 Set which frames to deinterlace, by default it is @code{all}.
9858 Can be @code{all} or @code{interlaced}.
9861 Set mode of operation.
9863 Can be one of the following:
9867 Use frame flags, both fields.
9869 Use frame flags, single field.
9873 Use bottom field only.
9875 Use both fields, top first.
9877 Use both fields, bottom first.
9881 Set which planes to process, by default filter process all frames.
9884 Set size of local neighborhood around each pixel, used by the predictor neural
9887 Can be one of the following:
9900 Set the number of neurons in predicctor neural network.
9901 Can be one of the following:
9912 Controls the number of different neural network predictions that are blended
9913 together to compute the final output value. Can be @code{fast}, default or
9917 Set which set of weights to use in the predictor.
9918 Can be one of the following:
9922 weights trained to minimize absolute error
9924 weights trained to minimize squared error
9928 Controls whether or not the prescreener neural network is used to decide
9929 which pixels should be processed by the predictor neural network and which
9930 can be handled by simple cubic interpolation.
9931 The prescreener is trained to know whether cubic interpolation will be
9932 sufficient for a pixel or whether it should be predicted by the predictor nn.
9933 The computational complexity of the prescreener nn is much less than that of
9934 the predictor nn. Since most pixels can be handled by cubic interpolation,
9935 using the prescreener generally results in much faster processing.
9936 The prescreener is pretty accurate, so the difference between using it and not
9937 using it is almost always unnoticeable.
9939 Can be one of the following:
9947 Default is @code{new}.
9950 Set various debugging flags.
9955 Force libavfilter not to use any of the specified pixel formats for the
9956 input to the next filter.
9958 It accepts the following parameters:
9962 A '|'-separated list of pixel format names, such as
9963 apix_fmts=yuv420p|monow|rgb24".
9967 @subsection Examples
9971 Force libavfilter to use a format different from @var{yuv420p} for the
9972 input to the vflip filter:
9974 noformat=pix_fmts=yuv420p,vflip
9978 Convert the input video to any of the formats not contained in the list:
9980 noformat=yuv420p|yuv444p|yuv410p
9986 Add noise on video input frame.
9988 The filter accepts the following options:
9996 Set noise seed for specific pixel component or all pixel components in case
9997 of @var{all_seed}. Default value is @code{123457}.
9999 @item all_strength, alls
10000 @item c0_strength, c0s
10001 @item c1_strength, c1s
10002 @item c2_strength, c2s
10003 @item c3_strength, c3s
10004 Set noise strength for specific pixel component or all pixel components in case
10005 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
10007 @item all_flags, allf
10008 @item c0_flags, c0f
10009 @item c1_flags, c1f
10010 @item c2_flags, c2f
10011 @item c3_flags, c3f
10012 Set pixel component flags or set flags for all components if @var{all_flags}.
10013 Available values for component flags are:
10016 averaged temporal noise (smoother)
10018 mix random noise with a (semi)regular pattern
10020 temporal noise (noise pattern changes between frames)
10022 uniform noise (gaussian otherwise)
10026 @subsection Examples
10028 Add temporal and uniform noise to input video:
10030 noise=alls=20:allf=t+u
10035 Pass the video source unchanged to the output.
10038 Optical Character Recognition
10040 This filter uses Tesseract for optical character recognition.
10042 It accepts the following options:
10046 Set datapath to tesseract data. Default is to use whatever was
10047 set at installation.
10050 Set language, default is "eng".
10053 Set character whitelist.
10056 Set character blacklist.
10059 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
10063 Apply a video transform using libopencv.
10065 To enable this filter, install the libopencv library and headers and
10066 configure FFmpeg with @code{--enable-libopencv}.
10068 It accepts the following parameters:
10073 The name of the libopencv filter to apply.
10075 @item filter_params
10076 The parameters to pass to the libopencv filter. If not specified, the default
10077 values are assumed.
10081 Refer to the official libopencv documentation for more precise
10083 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
10085 Several libopencv filters are supported; see the following subsections.
10090 Dilate an image by using a specific structuring element.
10091 It corresponds to the libopencv function @code{cvDilate}.
10093 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
10095 @var{struct_el} represents a structuring element, and has the syntax:
10096 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
10098 @var{cols} and @var{rows} represent the number of columns and rows of
10099 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
10100 point, and @var{shape} the shape for the structuring element. @var{shape}
10101 must be "rect", "cross", "ellipse", or "custom".
10103 If the value for @var{shape} is "custom", it must be followed by a
10104 string of the form "=@var{filename}". The file with name
10105 @var{filename} is assumed to represent a binary image, with each
10106 printable character corresponding to a bright pixel. When a custom
10107 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
10108 or columns and rows of the read file are assumed instead.
10110 The default value for @var{struct_el} is "3x3+0x0/rect".
10112 @var{nb_iterations} specifies the number of times the transform is
10113 applied to the image, and defaults to 1.
10117 # Use the default values
10120 # Dilate using a structuring element with a 5x5 cross, iterating two times
10121 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
10123 # Read the shape from the file diamond.shape, iterating two times.
10124 # The file diamond.shape may contain a pattern of characters like this
10130 # The specified columns and rows are ignored
10131 # but the anchor point coordinates are not
10132 ocv=dilate:0x0+2x2/custom=diamond.shape|2
10137 Erode an image by using a specific structuring element.
10138 It corresponds to the libopencv function @code{cvErode}.
10140 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
10141 with the same syntax and semantics as the @ref{dilate} filter.
10145 Smooth the input video.
10147 The filter takes the following parameters:
10148 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
10150 @var{type} is the type of smooth filter to apply, and must be one of
10151 the following values: "blur", "blur_no_scale", "median", "gaussian",
10152 or "bilateral". The default value is "gaussian".
10154 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
10155 depend on the smooth type. @var{param1} and
10156 @var{param2} accept integer positive values or 0. @var{param3} and
10157 @var{param4} accept floating point values.
10159 The default value for @var{param1} is 3. The default value for the
10160 other parameters is 0.
10162 These parameters correspond to the parameters assigned to the
10163 libopencv function @code{cvSmooth}.
10168 Overlay one video on top of another.
10170 It takes two inputs and has one output. The first input is the "main"
10171 video on which the second input is overlaid.
10173 It accepts the following parameters:
10175 A description of the accepted options follows.
10180 Set the expression for the x and y coordinates of the overlaid video
10181 on the main video. Default value is "0" for both expressions. In case
10182 the expression is invalid, it is set to a huge value (meaning that the
10183 overlay will not be displayed within the output visible area).
10186 The action to take when EOF is encountered on the secondary input; it accepts
10187 one of the following values:
10191 Repeat the last frame (the default).
10195 Pass the main input through.
10199 Set when the expressions for @option{x}, and @option{y} are evaluated.
10201 It accepts the following values:
10204 only evaluate expressions once during the filter initialization or
10205 when a command is processed
10208 evaluate expressions for each incoming frame
10211 Default value is @samp{frame}.
10214 If set to 1, force the output to terminate when the shortest input
10215 terminates. Default value is 0.
10218 Set the format for the output video.
10220 It accepts the following values:
10223 force YUV420 output
10226 force YUV422 output
10229 force YUV444 output
10232 force packed RGB output
10235 force planar RGB output
10238 Default value is @samp{yuv420}.
10240 @item rgb @emph{(deprecated)}
10241 If set to 1, force the filter to accept inputs in the RGB
10242 color space. Default value is 0. This option is deprecated, use
10243 @option{format} instead.
10246 If set to 1, force the filter to draw the last overlay frame over the
10247 main input until the end of the stream. A value of 0 disables this
10248 behavior. Default value is 1.
10251 The @option{x}, and @option{y} expressions can contain the following
10257 The main input width and height.
10261 The overlay input width and height.
10265 The computed values for @var{x} and @var{y}. They are evaluated for
10270 horizontal and vertical chroma subsample values of the output
10271 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
10275 the number of input frame, starting from 0
10278 the position in the file of the input frame, NAN if unknown
10281 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
10285 Note that the @var{n}, @var{pos}, @var{t} variables are available only
10286 when evaluation is done @emph{per frame}, and will evaluate to NAN
10287 when @option{eval} is set to @samp{init}.
10289 Be aware that frames are taken from each input video in timestamp
10290 order, hence, if their initial timestamps differ, it is a good idea
10291 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
10292 have them begin in the same zero timestamp, as the example for
10293 the @var{movie} filter does.
10295 You can chain together more overlays but you should test the
10296 efficiency of such approach.
10298 @subsection Commands
10300 This filter supports the following commands:
10304 Modify the x and y of the overlay input.
10305 The command accepts the same syntax of the corresponding option.
10307 If the specified expression is not valid, it is kept at its current
10311 @subsection Examples
10315 Draw the overlay at 10 pixels from the bottom right corner of the main
10318 overlay=main_w-overlay_w-10:main_h-overlay_h-10
10321 Using named options the example above becomes:
10323 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
10327 Insert a transparent PNG logo in the bottom left corner of the input,
10328 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
10330 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
10334 Insert 2 different transparent PNG logos (second logo on bottom
10335 right corner) using the @command{ffmpeg} tool:
10337 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
10341 Add a transparent color layer on top of the main video; @code{WxH}
10342 must specify the size of the main input to the overlay filter:
10344 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
10348 Play an original video and a filtered version (here with the deshake
10349 filter) side by side using the @command{ffplay} tool:
10351 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
10354 The above command is the same as:
10356 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
10360 Make a sliding overlay appearing from the left to the right top part of the
10361 screen starting since time 2:
10363 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
10367 Compose output by putting two input videos side to side:
10369 ffmpeg -i left.avi -i right.avi -filter_complex "
10370 nullsrc=size=200x100 [background];
10371 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
10372 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
10373 [background][left] overlay=shortest=1 [background+left];
10374 [background+left][right] overlay=shortest=1:x=100 [left+right]
10379 Mask 10-20 seconds of a video by applying the delogo filter to a section
10381 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
10382 -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]'
10387 Chain several overlays in cascade:
10389 nullsrc=s=200x200 [bg];
10390 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
10391 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
10392 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
10393 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
10394 [in3] null, [mid2] overlay=100:100 [out0]
10401 Apply Overcomplete Wavelet denoiser.
10403 The filter accepts the following options:
10409 Larger depth values will denoise lower frequency components more, but
10410 slow down filtering.
10412 Must be an int in the range 8-16, default is @code{8}.
10414 @item luma_strength, ls
10417 Must be a double value in the range 0-1000, default is @code{1.0}.
10419 @item chroma_strength, cs
10420 Set chroma strength.
10422 Must be a double value in the range 0-1000, default is @code{1.0}.
10428 Add paddings to the input image, and place the original input at the
10429 provided @var{x}, @var{y} coordinates.
10431 It accepts the following parameters:
10436 Specify an expression for the size of the output image with the
10437 paddings added. If the value for @var{width} or @var{height} is 0, the
10438 corresponding input size is used for the output.
10440 The @var{width} expression can reference the value set by the
10441 @var{height} expression, and vice versa.
10443 The default value of @var{width} and @var{height} is 0.
10447 Specify the offsets to place the input image at within the padded area,
10448 with respect to the top/left border of the output image.
10450 The @var{x} expression can reference the value set by the @var{y}
10451 expression, and vice versa.
10453 The default value of @var{x} and @var{y} is 0.
10456 Specify the color of the padded area. For the syntax of this option,
10457 check the "Color" section in the ffmpeg-utils manual.
10459 The default value of @var{color} is "black".
10462 Specify when to evaluate @var{width}, @var{height}, @var{x} and @var{y} expression.
10464 It accepts the following values:
10468 Only evaluate expressions once during the filter initialization or when
10469 a command is processed.
10472 Evaluate expressions for each incoming frame.
10476 Default value is @samp{init}.
10480 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
10481 options are expressions containing the following constants:
10486 The input video width and height.
10490 These are the same as @var{in_w} and @var{in_h}.
10494 The output width and height (the size of the padded area), as
10495 specified by the @var{width} and @var{height} expressions.
10499 These are the same as @var{out_w} and @var{out_h}.
10503 The x and y offsets as specified by the @var{x} and @var{y}
10504 expressions, or NAN if not yet specified.
10507 same as @var{iw} / @var{ih}
10510 input sample aspect ratio
10513 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
10517 The horizontal and vertical chroma subsample values. For example for the
10518 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10521 @subsection Examples
10525 Add paddings with the color "violet" to the input video. The output video
10526 size is 640x480, and the top-left corner of the input video is placed at
10529 pad=640:480:0:40:violet
10532 The example above is equivalent to the following command:
10534 pad=width=640:height=480:x=0:y=40:color=violet
10538 Pad the input to get an output with dimensions increased by 3/2,
10539 and put the input video at the center of the padded area:
10541 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
10545 Pad the input to get a squared output with size equal to the maximum
10546 value between the input width and height, and put the input video at
10547 the center of the padded area:
10549 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
10553 Pad the input to get a final w/h ratio of 16:9:
10555 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
10559 In case of anamorphic video, in order to set the output display aspect
10560 correctly, it is necessary to use @var{sar} in the expression,
10561 according to the relation:
10563 (ih * X / ih) * sar = output_dar
10564 X = output_dar / sar
10567 Thus the previous example needs to be modified to:
10569 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
10573 Double the output size and put the input video in the bottom-right
10574 corner of the output padded area:
10576 pad="2*iw:2*ih:ow-iw:oh-ih"
10580 @anchor{palettegen}
10581 @section palettegen
10583 Generate one palette for a whole video stream.
10585 It accepts the following options:
10589 Set the maximum number of colors to quantize in the palette.
10590 Note: the palette will still contain 256 colors; the unused palette entries
10593 @item reserve_transparent
10594 Create a palette of 255 colors maximum and reserve the last one for
10595 transparency. Reserving the transparency color is useful for GIF optimization.
10596 If not set, the maximum of colors in the palette will be 256. You probably want
10597 to disable this option for a standalone image.
10601 Set statistics mode.
10603 It accepts the following values:
10606 Compute full frame histograms.
10608 Compute histograms only for the part that differs from previous frame. This
10609 might be relevant to give more importance to the moving part of your input if
10610 the background is static.
10612 Compute new histogram for each frame.
10615 Default value is @var{full}.
10618 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10619 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10620 color quantization of the palette. This information is also visible at
10621 @var{info} logging level.
10623 @subsection Examples
10627 Generate a representative palette of a given video using @command{ffmpeg}:
10629 ffmpeg -i input.mkv -vf palettegen palette.png
10633 @section paletteuse
10635 Use a palette to downsample an input video stream.
10637 The filter takes two inputs: one video stream and a palette. The palette must
10638 be a 256 pixels image.
10640 It accepts the following options:
10644 Select dithering mode. Available algorithms are:
10647 Ordered 8x8 bayer dithering (deterministic)
10649 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10650 Note: this dithering is sometimes considered "wrong" and is included as a
10652 @item floyd_steinberg
10653 Floyd and Steingberg dithering (error diffusion)
10655 Frankie Sierra dithering v2 (error diffusion)
10657 Frankie Sierra dithering v2 "Lite" (error diffusion)
10660 Default is @var{sierra2_4a}.
10663 When @var{bayer} dithering is selected, this option defines the scale of the
10664 pattern (how much the crosshatch pattern is visible). A low value means more
10665 visible pattern for less banding, and higher value means less visible pattern
10666 at the cost of more banding.
10668 The option must be an integer value in the range [0,5]. Default is @var{2}.
10671 If set, define the zone to process
10675 Only the changing rectangle will be reprocessed. This is similar to GIF
10676 cropping/offsetting compression mechanism. This option can be useful for speed
10677 if only a part of the image is changing, and has use cases such as limiting the
10678 scope of the error diffusal @option{dither} to the rectangle that bounds the
10679 moving scene (it leads to more deterministic output if the scene doesn't change
10680 much, and as a result less moving noise and better GIF compression).
10683 Default is @var{none}.
10686 Take new palette for each output frame.
10689 @subsection Examples
10693 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10694 using @command{ffmpeg}:
10696 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10700 @section perspective
10702 Correct perspective of video not recorded perpendicular to the screen.
10704 A description of the accepted parameters follows.
10715 Set coordinates expression for top left, top right, bottom left and bottom right corners.
10716 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10717 If the @code{sense} option is set to @code{source}, then the specified points will be sent
10718 to the corners of the destination. If the @code{sense} option is set to @code{destination},
10719 then the corners of the source will be sent to the specified coordinates.
10721 The expressions can use the following variables:
10726 the width and height of video frame.
10730 Output frame count.
10733 @item interpolation
10734 Set interpolation for perspective correction.
10736 It accepts the following values:
10742 Default value is @samp{linear}.
10745 Set interpretation of coordinate options.
10747 It accepts the following values:
10751 Send point in the source specified by the given coordinates to
10752 the corners of the destination.
10754 @item 1, destination
10756 Send the corners of the source to the point in the destination specified
10757 by the given coordinates.
10759 Default value is @samp{source}.
10763 Set when the expressions for coordinates @option{x0,y0,...x3,y3} are evaluated.
10765 It accepts the following values:
10768 only evaluate expressions once during the filter initialization or
10769 when a command is processed
10772 evaluate expressions for each incoming frame
10775 Default value is @samp{init}.
10780 Delay interlaced video by one field time so that the field order changes.
10782 The intended use is to fix PAL movies that have been captured with the
10783 opposite field order to the film-to-video transfer.
10785 A description of the accepted parameters follows.
10791 It accepts the following values:
10794 Capture field order top-first, transfer bottom-first.
10795 Filter will delay the bottom field.
10798 Capture field order bottom-first, transfer top-first.
10799 Filter will delay the top field.
10802 Capture and transfer with the same field order. This mode only exists
10803 for the documentation of the other options to refer to, but if you
10804 actually select it, the filter will faithfully do nothing.
10807 Capture field order determined automatically by field flags, transfer
10809 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10810 basis using field flags. If no field information is available,
10811 then this works just like @samp{u}.
10814 Capture unknown or varying, transfer opposite.
10815 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10816 analyzing the images and selecting the alternative that produces best
10817 match between the fields.
10820 Capture top-first, transfer unknown or varying.
10821 Filter selects among @samp{t} and @samp{p} using image analysis.
10824 Capture bottom-first, transfer unknown or varying.
10825 Filter selects among @samp{b} and @samp{p} using image analysis.
10828 Capture determined by field flags, transfer unknown or varying.
10829 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10830 image analysis. If no field information is available, then this works just
10831 like @samp{U}. This is the default mode.
10834 Both capture and transfer unknown or varying.
10835 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10839 @section pixdesctest
10841 Pixel format descriptor test filter, mainly useful for internal
10842 testing. The output video should be equal to the input video.
10846 format=monow, pixdesctest
10849 can be used to test the monowhite pixel format descriptor definition.
10853 Enable the specified chain of postprocessing subfilters using libpostproc. This
10854 library should be automatically selected with a GPL build (@code{--enable-gpl}).
10855 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
10856 Each subfilter and some options have a short and a long name that can be used
10857 interchangeably, i.e. dr/dering are the same.
10859 The filters accept the following options:
10863 Set postprocessing subfilters string.
10866 All subfilters share common options to determine their scope:
10870 Honor the quality commands for this subfilter.
10873 Do chrominance filtering, too (default).
10876 Do luminance filtering only (no chrominance).
10879 Do chrominance filtering only (no luminance).
10882 These options can be appended after the subfilter name, separated by a '|'.
10884 Available subfilters are:
10887 @item hb/hdeblock[|difference[|flatness]]
10888 Horizontal deblocking filter
10891 Difference factor where higher values mean more deblocking (default: @code{32}).
10893 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10896 @item vb/vdeblock[|difference[|flatness]]
10897 Vertical deblocking filter
10900 Difference factor where higher values mean more deblocking (default: @code{32}).
10902 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10905 @item ha/hadeblock[|difference[|flatness]]
10906 Accurate horizontal deblocking filter
10909 Difference factor where higher values mean more deblocking (default: @code{32}).
10911 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10914 @item va/vadeblock[|difference[|flatness]]
10915 Accurate vertical deblocking filter
10918 Difference factor where higher values mean more deblocking (default: @code{32}).
10920 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10924 The horizontal and vertical deblocking filters share the difference and
10925 flatness values so you cannot set different horizontal and vertical
10929 @item h1/x1hdeblock
10930 Experimental horizontal deblocking filter
10932 @item v1/x1vdeblock
10933 Experimental vertical deblocking filter
10938 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
10941 larger -> stronger filtering
10943 larger -> stronger filtering
10945 larger -> stronger filtering
10948 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
10951 Stretch luminance to @code{0-255}.
10954 @item lb/linblenddeint
10955 Linear blend deinterlacing filter that deinterlaces the given block by
10956 filtering all lines with a @code{(1 2 1)} filter.
10958 @item li/linipoldeint
10959 Linear interpolating deinterlacing filter that deinterlaces the given block by
10960 linearly interpolating every second line.
10962 @item ci/cubicipoldeint
10963 Cubic interpolating deinterlacing filter deinterlaces the given block by
10964 cubically interpolating every second line.
10966 @item md/mediandeint
10967 Median deinterlacing filter that deinterlaces the given block by applying a
10968 median filter to every second line.
10970 @item fd/ffmpegdeint
10971 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
10972 second line with a @code{(-1 4 2 4 -1)} filter.
10975 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
10976 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
10978 @item fq/forceQuant[|quantizer]
10979 Overrides the quantizer table from the input with the constant quantizer you
10987 Default pp filter combination (@code{hb|a,vb|a,dr|a})
10990 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
10993 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
10996 @subsection Examples
11000 Apply horizontal and vertical deblocking, deringing and automatic
11001 brightness/contrast:
11007 Apply default filters without brightness/contrast correction:
11013 Apply default filters and temporal denoiser:
11015 pp=default/tmpnoise|1|2|3
11019 Apply deblocking on luminance only, and switch vertical deblocking on or off
11020 automatically depending on available CPU time:
11027 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
11028 similar to spp = 6 with 7 point DCT, where only the center sample is
11031 The filter accepts the following options:
11035 Force a constant quantization parameter. It accepts an integer in range
11036 0 to 63. If not set, the filter will use the QP from the video stream
11040 Set thresholding mode. Available modes are:
11044 Set hard thresholding.
11046 Set soft thresholding (better de-ringing effect, but likely blurrier).
11048 Set medium thresholding (good results, default).
11052 @section premultiply
11053 Apply alpha premultiply effect to input video stream using first plane
11054 of second stream as alpha.
11056 Both streams must have same dimensions and same pixel format.
11059 Apply prewitt operator to input video stream.
11061 The filter accepts the following option:
11065 Set which planes will be processed, unprocessed planes will be copied.
11066 By default value 0xf, all planes will be processed.
11069 Set value which will be multiplied with filtered result.
11072 Set value which will be added to filtered result.
11077 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
11078 Ratio) between two input videos.
11080 This filter takes in input two input videos, the first input is
11081 considered the "main" source and is passed unchanged to the
11082 output. The second input is used as a "reference" video for computing
11085 Both video inputs must have the same resolution and pixel format for
11086 this filter to work correctly. Also it assumes that both inputs
11087 have the same number of frames, which are compared one by one.
11089 The obtained average PSNR is printed through the logging system.
11091 The filter stores the accumulated MSE (mean squared error) of each
11092 frame, and at the end of the processing it is averaged across all frames
11093 equally, and the following formula is applied to obtain the PSNR:
11096 PSNR = 10*log10(MAX^2/MSE)
11099 Where MAX is the average of the maximum values of each component of the
11102 The description of the accepted parameters follows.
11105 @item stats_file, f
11106 If specified the filter will use the named file to save the PSNR of
11107 each individual frame. When filename equals "-" the data is sent to
11110 @item stats_version
11111 Specifies which version of the stats file format to use. Details of
11112 each format are written below.
11113 Default value is 1.
11115 @item stats_add_max
11116 Determines whether the max value is output to the stats log.
11117 Default value is 0.
11118 Requires stats_version >= 2. If this is set and stats_version < 2,
11119 the filter will return an error.
11122 The file printed if @var{stats_file} is selected, contains a sequence of
11123 key/value pairs of the form @var{key}:@var{value} for each compared
11126 If a @var{stats_version} greater than 1 is specified, a header line precedes
11127 the list of per-frame-pair stats, with key value pairs following the frame
11128 format with the following parameters:
11131 @item psnr_log_version
11132 The version of the log file format. Will match @var{stats_version}.
11135 A comma separated list of the per-frame-pair parameters included in
11139 A description of each shown per-frame-pair parameter follows:
11143 sequential number of the input frame, starting from 1
11146 Mean Square Error pixel-by-pixel average difference of the compared
11147 frames, averaged over all the image components.
11149 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
11150 Mean Square Error pixel-by-pixel average difference of the compared
11151 frames for the component specified by the suffix.
11153 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
11154 Peak Signal to Noise ratio of the compared frames for the component
11155 specified by the suffix.
11157 @item max_avg, max_y, max_u, max_v
11158 Maximum allowed value for each channel, and average over all
11164 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
11165 [main][ref] psnr="stats_file=stats.log" [out]
11168 On this example the input file being processed is compared with the
11169 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
11170 is stored in @file{stats.log}.
11175 Pulldown reversal (inverse telecine) filter, capable of handling mixed
11176 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
11179 The pullup filter is designed to take advantage of future context in making
11180 its decisions. This filter is stateless in the sense that it does not lock
11181 onto a pattern to follow, but it instead looks forward to the following
11182 fields in order to identify matches and rebuild progressive frames.
11184 To produce content with an even framerate, insert the fps filter after
11185 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
11186 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
11188 The filter accepts the following options:
11195 These options set the amount of "junk" to ignore at the left, right, top, and
11196 bottom of the image, respectively. Left and right are in units of 8 pixels,
11197 while top and bottom are in units of 2 lines.
11198 The default is 8 pixels on each side.
11201 Set the strict breaks. Setting this option to 1 will reduce the chances of
11202 filter generating an occasional mismatched frame, but it may also cause an
11203 excessive number of frames to be dropped during high motion sequences.
11204 Conversely, setting it to -1 will make filter match fields more easily.
11205 This may help processing of video where there is slight blurring between
11206 the fields, but may also cause there to be interlaced frames in the output.
11207 Default value is @code{0}.
11210 Set the metric plane to use. It accepts the following values:
11216 Use chroma blue plane.
11219 Use chroma red plane.
11222 This option may be set to use chroma plane instead of the default luma plane
11223 for doing filter's computations. This may improve accuracy on very clean
11224 source material, but more likely will decrease accuracy, especially if there
11225 is chroma noise (rainbow effect) or any grayscale video.
11226 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
11227 load and make pullup usable in realtime on slow machines.
11230 For best results (without duplicated frames in the output file) it is
11231 necessary to change the output frame rate. For example, to inverse
11232 telecine NTSC input:
11234 ffmpeg -i input -vf pullup -r 24000/1001 ...
11239 Change video quantization parameters (QP).
11241 The filter accepts the following option:
11245 Set expression for quantization parameter.
11248 The expression is evaluated through the eval API and can contain, among others,
11249 the following constants:
11253 1 if index is not 129, 0 otherwise.
11256 Sequentional index starting from -129 to 128.
11259 @subsection Examples
11263 Some equation like:
11271 Flush video frames from internal cache of frames into a random order.
11272 No frame is discarded.
11273 Inspired by @ref{frei0r} nervous filter.
11277 Set size in number of frames of internal cache, in range from @code{2} to
11278 @code{512}. Default is @code{30}.
11281 Set seed for random number generator, must be an integer included between
11282 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11283 less than @code{0}, the filter will try to use a good random seed on a
11287 @section readeia608
11289 Read closed captioning (EIA-608) information from the top lines of a video frame.
11291 This filter adds frame metadata for @code{lavfi.readeia608.X.cc} and
11292 @code{lavfi.readeia608.X.line}, where @code{X} is the number of the identified line
11293 with EIA-608 data (starting from 0). A description of each metadata value follows:
11296 @item lavfi.readeia608.X.cc
11297 The two bytes stored as EIA-608 data (printed in hexadecimal).
11299 @item lavfi.readeia608.X.line
11300 The number of the line on which the EIA-608 data was identified and read.
11303 This filter accepts the following options:
11307 Set the line to start scanning for EIA-608 data. Default is @code{0}.
11310 Set the line to end scanning for EIA-608 data. Default is @code{29}.
11313 Set minimal acceptable amplitude change for sync codes detection.
11314 Default is @code{0.2}. Allowed range is @code{[0.001 - 1]}.
11317 Set the ratio of width reserved for sync code detection.
11318 Default is @code{0.27}. Allowed range is @code{[0.01 - 0.7]}.
11321 Set the max peaks height difference for sync code detection.
11322 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11325 Set max peaks period difference for sync code detection.
11326 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11329 Set the first two max start code bits differences.
11330 Default is @code{0.02}. Allowed range is @code{[0.0 - 0.5]}.
11333 Set the minimum ratio of bits height compared to 3rd start code bit.
11334 Default is @code{0.75}. Allowed range is @code{[0.01 - 1]}.
11337 Set the white color threshold. Default is @code{0.35}. Allowed range is @code{[0.1 - 1]}.
11340 Set the black color threshold. Default is @code{0.15}. Allowed range is @code{[0.0 - 0.5]}.
11343 Enable checking the parity bit. In the event of a parity error, the filter will output
11344 @code{0x00} for that character. Default is false.
11347 @subsection Examples
11351 Output a csv with presentation time and the first two lines of identified EIA-608 captioning data.
11353 ffprobe -f lavfi -i movie=captioned_video.mov,readeia608 -show_entries frame=pkt_pts_time:frame_tags=lavfi.readeia608.0.cc,lavfi.readeia608.1.cc -of csv
11359 Read vertical interval timecode (VITC) information from the top lines of a
11362 The filter adds frame metadata key @code{lavfi.readvitc.tc_str} with the
11363 timecode value, if a valid timecode has been detected. Further metadata key
11364 @code{lavfi.readvitc.found} is set to 0/1 depending on whether
11365 timecode data has been found or not.
11367 This filter accepts the following options:
11371 Set the maximum number of lines to scan for VITC data. If the value is set to
11372 @code{-1} the full video frame is scanned. Default is @code{45}.
11375 Set the luma threshold for black. Accepts float numbers in the range [0.0,1.0],
11376 default value is @code{0.2}. The value must be equal or less than @code{thr_w}.
11379 Set the luma threshold for white. Accepts float numbers in the range [0.0,1.0],
11380 default value is @code{0.6}. The value must be equal or greater than @code{thr_b}.
11383 @subsection Examples
11387 Detect and draw VITC data onto the video frame; if no valid VITC is detected,
11388 draw @code{--:--:--:--} as a placeholder:
11390 ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%@{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--@}:x=(w-tw)/2:y=400-ascent'
11396 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
11398 Destination pixel at position (X, Y) will be picked from source (x, y) position
11399 where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
11400 value for pixel will be used for destination pixel.
11402 Xmap and Ymap input video streams must be of same dimensions. Output video stream
11403 will have Xmap/Ymap video stream dimensions.
11404 Xmap and Ymap input video streams are 16bit depth, single channel.
11406 @section removegrain
11408 The removegrain filter is a spatial denoiser for progressive video.
11412 Set mode for the first plane.
11415 Set mode for the second plane.
11418 Set mode for the third plane.
11421 Set mode for the fourth plane.
11424 Range of mode is from 0 to 24. Description of each mode follows:
11428 Leave input plane unchanged. Default.
11431 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
11434 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
11437 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
11440 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
11441 This is equivalent to a median filter.
11444 Line-sensitive clipping giving the minimal change.
11447 Line-sensitive clipping, intermediate.
11450 Line-sensitive clipping, intermediate.
11453 Line-sensitive clipping, intermediate.
11456 Line-sensitive clipping on a line where the neighbours pixels are the closest.
11459 Replaces the target pixel with the closest neighbour.
11462 [1 2 1] horizontal and vertical kernel blur.
11468 Bob mode, interpolates top field from the line where the neighbours
11469 pixels are the closest.
11472 Bob mode, interpolates bottom field from the line where the neighbours
11473 pixels are the closest.
11476 Bob mode, interpolates top field. Same as 13 but with a more complicated
11477 interpolation formula.
11480 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
11481 interpolation formula.
11484 Clips the pixel with the minimum and maximum of respectively the maximum and
11485 minimum of each pair of opposite neighbour pixels.
11488 Line-sensitive clipping using opposite neighbours whose greatest distance from
11489 the current pixel is minimal.
11492 Replaces the pixel with the average of its 8 neighbours.
11495 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
11498 Clips pixels using the averages of opposite neighbour.
11501 Same as mode 21 but simpler and faster.
11504 Small edge and halo removal, but reputed useless.
11510 @section removelogo
11512 Suppress a TV station logo, using an image file to determine which
11513 pixels comprise the logo. It works by filling in the pixels that
11514 comprise the logo with neighboring pixels.
11516 The filter accepts the following options:
11520 Set the filter bitmap file, which can be any image format supported by
11521 libavformat. The width and height of the image file must match those of the
11522 video stream being processed.
11525 Pixels in the provided bitmap image with a value of zero are not
11526 considered part of the logo, non-zero pixels are considered part of
11527 the logo. If you use white (255) for the logo and black (0) for the
11528 rest, you will be safe. For making the filter bitmap, it is
11529 recommended to take a screen capture of a black frame with the logo
11530 visible, and then using a threshold filter followed by the erode
11531 filter once or twice.
11533 If needed, little splotches can be fixed manually. Remember that if
11534 logo pixels are not covered, the filter quality will be much
11535 reduced. Marking too many pixels as part of the logo does not hurt as
11536 much, but it will increase the amount of blurring needed to cover over
11537 the image and will destroy more information than necessary, and extra
11538 pixels will slow things down on a large logo.
11540 @section repeatfields
11542 This filter uses the repeat_field flag from the Video ES headers and hard repeats
11543 fields based on its value.
11547 Reverse a video clip.
11549 Warning: This filter requires memory to buffer the entire clip, so trimming
11552 @subsection Examples
11556 Take the first 5 seconds of a clip, and reverse it.
11564 Rotate video by an arbitrary angle expressed in radians.
11566 The filter accepts the following options:
11568 A description of the optional parameters follows.
11571 Set an expression for the angle by which to rotate the input video
11572 clockwise, expressed as a number of radians. A negative value will
11573 result in a counter-clockwise rotation. By default it is set to "0".
11575 This expression is evaluated for each frame.
11578 Set the output width expression, default value is "iw".
11579 This expression is evaluated just once during configuration.
11582 Set the output height expression, default value is "ih".
11583 This expression is evaluated just once during configuration.
11586 Enable bilinear interpolation if set to 1, a value of 0 disables
11587 it. Default value is 1.
11590 Set the color used to fill the output area not covered by the rotated
11591 image. For the general syntax of this option, check the "Color" section in the
11592 ffmpeg-utils manual. If the special value "none" is selected then no
11593 background is printed (useful for example if the background is never shown).
11595 Default value is "black".
11598 The expressions for the angle and the output size can contain the
11599 following constants and functions:
11603 sequential number of the input frame, starting from 0. It is always NAN
11604 before the first frame is filtered.
11607 time in seconds of the input frame, it is set to 0 when the filter is
11608 configured. It is always NAN before the first frame is filtered.
11612 horizontal and vertical chroma subsample values. For example for the
11613 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11617 the input video width and height
11621 the output width and height, that is the size of the padded area as
11622 specified by the @var{width} and @var{height} expressions
11626 the minimal width/height required for completely containing the input
11627 video rotated by @var{a} radians.
11629 These are only available when computing the @option{out_w} and
11630 @option{out_h} expressions.
11633 @subsection Examples
11637 Rotate the input by PI/6 radians clockwise:
11643 Rotate the input by PI/6 radians counter-clockwise:
11649 Rotate the input by 45 degrees clockwise:
11655 Apply a constant rotation with period T, starting from an angle of PI/3:
11657 rotate=PI/3+2*PI*t/T
11661 Make the input video rotation oscillating with a period of T
11662 seconds and an amplitude of A radians:
11664 rotate=A*sin(2*PI/T*t)
11668 Rotate the video, output size is chosen so that the whole rotating
11669 input video is always completely contained in the output:
11671 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
11675 Rotate the video, reduce the output size so that no background is ever
11678 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
11682 @subsection Commands
11684 The filter supports the following commands:
11688 Set the angle expression.
11689 The command accepts the same syntax of the corresponding option.
11691 If the specified expression is not valid, it is kept at its current
11697 Apply Shape Adaptive Blur.
11699 The filter accepts the following options:
11702 @item luma_radius, lr
11703 Set luma blur filter strength, must be a value in range 0.1-4.0, default
11704 value is 1.0. A greater value will result in a more blurred image, and
11705 in slower processing.
11707 @item luma_pre_filter_radius, lpfr
11708 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
11711 @item luma_strength, ls
11712 Set luma maximum difference between pixels to still be considered, must
11713 be a value in the 0.1-100.0 range, default value is 1.0.
11715 @item chroma_radius, cr
11716 Set chroma blur filter strength, must be a value in range -0.9-4.0. A
11717 greater value will result in a more blurred image, and in slower
11720 @item chroma_pre_filter_radius, cpfr
11721 Set chroma pre-filter radius, must be a value in the -0.9-2.0 range.
11723 @item chroma_strength, cs
11724 Set chroma maximum difference between pixels to still be considered,
11725 must be a value in the -0.9-100.0 range.
11728 Each chroma option value, if not explicitly specified, is set to the
11729 corresponding luma option value.
11734 Scale (resize) the input video, using the libswscale library.
11736 The scale filter forces the output display aspect ratio to be the same
11737 of the input, by changing the output sample aspect ratio.
11739 If the input image format is different from the format requested by
11740 the next filter, the scale filter will convert the input to the
11743 @subsection Options
11744 The filter accepts the following options, or any of the options
11745 supported by the libswscale scaler.
11747 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
11748 the complete list of scaler options.
11753 Set the output video dimension expression. Default value is the input
11756 If the value is 0, the input width is used for the output.
11758 If one of the values is -1, the scale filter will use a value that
11759 maintains the aspect ratio of the input image, calculated from the
11760 other specified dimension. If both of them are -1, the input size is
11763 If one of the values is -n with n > 1, the scale filter will also use a value
11764 that maintains the aspect ratio of the input image, calculated from the other
11765 specified dimension. After that it will, however, make sure that the calculated
11766 dimension is divisible by n and adjust the value if necessary.
11768 See below for the list of accepted constants for use in the dimension
11772 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
11776 Only evaluate expressions once during the filter initialization or when a command is processed.
11779 Evaluate expressions for each incoming frame.
11783 Default value is @samp{init}.
11787 Set the interlacing mode. It accepts the following values:
11791 Force interlaced aware scaling.
11794 Do not apply interlaced scaling.
11797 Select interlaced aware scaling depending on whether the source frames
11798 are flagged as interlaced or not.
11801 Default value is @samp{0}.
11804 Set libswscale scaling flags. See
11805 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11806 complete list of values. If not explicitly specified the filter applies
11810 @item param0, param1
11811 Set libswscale input parameters for scaling algorithms that need them. See
11812 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11813 complete documentation. If not explicitly specified the filter applies
11819 Set the video size. For the syntax of this option, check the
11820 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11822 @item in_color_matrix
11823 @item out_color_matrix
11824 Set in/output YCbCr color space type.
11826 This allows the autodetected value to be overridden as well as allows forcing
11827 a specific value used for the output and encoder.
11829 If not specified, the color space type depends on the pixel format.
11835 Choose automatically.
11838 Format conforming to International Telecommunication Union (ITU)
11839 Recommendation BT.709.
11842 Set color space conforming to the United States Federal Communications
11843 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
11846 Set color space conforming to:
11850 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
11853 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
11856 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
11861 Set color space conforming to SMPTE ST 240:1999.
11866 Set in/output YCbCr sample range.
11868 This allows the autodetected value to be overridden as well as allows forcing
11869 a specific value used for the output and encoder. If not specified, the
11870 range depends on the pixel format. Possible values:
11874 Choose automatically.
11877 Set full range (0-255 in case of 8-bit luma).
11880 Set "MPEG" range (16-235 in case of 8-bit luma).
11883 @item force_original_aspect_ratio
11884 Enable decreasing or increasing output video width or height if necessary to
11885 keep the original aspect ratio. Possible values:
11889 Scale the video as specified and disable this feature.
11892 The output video dimensions will automatically be decreased if needed.
11895 The output video dimensions will automatically be increased if needed.
11899 One useful instance of this option is that when you know a specific device's
11900 maximum allowed resolution, you can use this to limit the output video to
11901 that, while retaining the aspect ratio. For example, device A allows
11902 1280x720 playback, and your video is 1920x800. Using this option (set it to
11903 decrease) and specifying 1280x720 to the command line makes the output
11906 Please note that this is a different thing than specifying -1 for @option{w}
11907 or @option{h}, you still need to specify the output resolution for this option
11912 The values of the @option{w} and @option{h} options are expressions
11913 containing the following constants:
11918 The input width and height
11922 These are the same as @var{in_w} and @var{in_h}.
11926 The output (scaled) width and height
11930 These are the same as @var{out_w} and @var{out_h}
11933 The same as @var{iw} / @var{ih}
11936 input sample aspect ratio
11939 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
11943 horizontal and vertical input chroma subsample values. For example for the
11944 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11948 horizontal and vertical output chroma subsample values. For example for the
11949 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11952 @subsection Examples
11956 Scale the input video to a size of 200x100
11961 This is equivalent to:
11972 Specify a size abbreviation for the output size:
11977 which can also be written as:
11983 Scale the input to 2x:
11985 scale=w=2*iw:h=2*ih
11989 The above is the same as:
11991 scale=2*in_w:2*in_h
11995 Scale the input to 2x with forced interlaced scaling:
11997 scale=2*iw:2*ih:interl=1
12001 Scale the input to half size:
12003 scale=w=iw/2:h=ih/2
12007 Increase the width, and set the height to the same size:
12013 Seek Greek harmony:
12020 Increase the height, and set the width to 3/2 of the height:
12022 scale=w=3/2*oh:h=3/5*ih
12026 Increase the size, making the size a multiple of the chroma
12029 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
12033 Increase the width to a maximum of 500 pixels,
12034 keeping the same aspect ratio as the input:
12036 scale=w='min(500\, iw*3/2):h=-1'
12040 @subsection Commands
12042 This filter supports the following commands:
12046 Set the output video dimension expression.
12047 The command accepts the same syntax of the corresponding option.
12049 If the specified expression is not valid, it is kept at its current
12055 Use the NVIDIA Performance Primitives (libnpp) to perform scaling and/or pixel
12056 format conversion on CUDA video frames. Setting the output width and height
12057 works in the same way as for the @var{scale} filter.
12059 The following additional options are accepted:
12062 The pixel format of the output CUDA frames. If set to the string "same" (the
12063 default), the input format will be kept. Note that automatic format negotiation
12064 and conversion is not yet supported for hardware frames
12067 The interpolation algorithm used for resizing. One of the following:
12074 @item cubic2p_bspline
12075 2-parameter cubic (B=1, C=0)
12077 @item cubic2p_catmullrom
12078 2-parameter cubic (B=0, C=1/2)
12080 @item cubic2p_b05c03
12081 2-parameter cubic (B=1/2, C=3/10)
12093 Scale (resize) the input video, based on a reference video.
12095 See the scale filter for available options, scale2ref supports the same but
12096 uses the reference video instead of the main input as basis.
12098 @subsection Examples
12102 Scale a subtitle stream to match the main video in size before overlaying
12104 'scale2ref[b][a];[a][b]overlay'
12108 @anchor{selectivecolor}
12109 @section selectivecolor
12111 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
12112 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
12113 by the "purity" of the color (that is, how saturated it already is).
12115 This filter is similar to the Adobe Photoshop Selective Color tool.
12117 The filter accepts the following options:
12120 @item correction_method
12121 Select color correction method.
12123 Available values are:
12126 Specified adjustments are applied "as-is" (added/subtracted to original pixel
12129 Specified adjustments are relative to the original component value.
12131 Default is @code{absolute}.
12133 Adjustments for red pixels (pixels where the red component is the maximum)
12135 Adjustments for yellow pixels (pixels where the blue component is the minimum)
12137 Adjustments for green pixels (pixels where the green component is the maximum)
12139 Adjustments for cyan pixels (pixels where the red component is the minimum)
12141 Adjustments for blue pixels (pixels where the blue component is the maximum)
12143 Adjustments for magenta pixels (pixels where the green component is the minimum)
12145 Adjustments for white pixels (pixels where all components are greater than 128)
12147 Adjustments for all pixels except pure black and pure white
12149 Adjustments for black pixels (pixels where all components are lesser than 128)
12151 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
12154 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
12155 4 space separated floating point adjustment values in the [-1,1] range,
12156 respectively to adjust the amount of cyan, magenta, yellow and black for the
12157 pixels of its range.
12159 @subsection Examples
12163 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
12164 increase magenta by 27% in blue areas:
12166 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
12170 Use a Photoshop selective color preset:
12172 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
12176 @anchor{separatefields}
12177 @section separatefields
12179 The @code{separatefields} takes a frame-based video input and splits
12180 each frame into its components fields, producing a new half height clip
12181 with twice the frame rate and twice the frame count.
12183 This filter use field-dominance information in frame to decide which
12184 of each pair of fields to place first in the output.
12185 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
12187 @section setdar, setsar
12189 The @code{setdar} filter sets the Display Aspect Ratio for the filter
12192 This is done by changing the specified Sample (aka Pixel) Aspect
12193 Ratio, according to the following equation:
12195 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
12198 Keep in mind that the @code{setdar} filter does not modify the pixel
12199 dimensions of the video frame. Also, the display aspect ratio set by
12200 this filter may be changed by later filters in the filterchain,
12201 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
12204 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
12205 the filter output video.
12207 Note that as a consequence of the application of this filter, the
12208 output display aspect ratio will change according to the equation
12211 Keep in mind that the sample aspect ratio set by the @code{setsar}
12212 filter may be changed by later filters in the filterchain, e.g. if
12213 another "setsar" or a "setdar" filter is applied.
12215 It accepts the following parameters:
12218 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
12219 Set the aspect ratio used by the filter.
12221 The parameter can be a floating point number string, an expression, or
12222 a string of the form @var{num}:@var{den}, where @var{num} and
12223 @var{den} are the numerator and denominator of the aspect ratio. If
12224 the parameter is not specified, it is assumed the value "0".
12225 In case the form "@var{num}:@var{den}" is used, the @code{:} character
12229 Set the maximum integer value to use for expressing numerator and
12230 denominator when reducing the expressed aspect ratio to a rational.
12231 Default value is @code{100}.
12235 The parameter @var{sar} is an expression containing
12236 the following constants:
12240 These are approximated values for the mathematical constants e
12241 (Euler's number), pi (Greek pi), and phi (the golden ratio).
12244 The input width and height.
12247 These are the same as @var{w} / @var{h}.
12250 The input sample aspect ratio.
12253 The input display aspect ratio. It is the same as
12254 (@var{w} / @var{h}) * @var{sar}.
12257 Horizontal and vertical chroma subsample values. For example, for the
12258 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12261 @subsection Examples
12266 To change the display aspect ratio to 16:9, specify one of the following:
12273 To change the sample aspect ratio to 10:11, specify:
12279 To set a display aspect ratio of 16:9, and specify a maximum integer value of
12280 1000 in the aspect ratio reduction, use the command:
12282 setdar=ratio=16/9:max=1000
12290 Force field for the output video frame.
12292 The @code{setfield} filter marks the interlace type field for the
12293 output frames. It does not change the input frame, but only sets the
12294 corresponding property, which affects how the frame is treated by
12295 following filters (e.g. @code{fieldorder} or @code{yadif}).
12297 The filter accepts the following options:
12302 Available values are:
12306 Keep the same field property.
12309 Mark the frame as bottom-field-first.
12312 Mark the frame as top-field-first.
12315 Mark the frame as progressive.
12321 Show a line containing various information for each input video frame.
12322 The input video is not modified.
12324 The shown line contains a sequence of key/value pairs of the form
12325 @var{key}:@var{value}.
12327 The following values are shown in the output:
12331 The (sequential) number of the input frame, starting from 0.
12334 The Presentation TimeStamp of the input frame, expressed as a number of
12335 time base units. The time base unit depends on the filter input pad.
12338 The Presentation TimeStamp of the input frame, expressed as a number of
12342 The position of the frame in the input stream, or -1 if this information is
12343 unavailable and/or meaningless (for example in case of synthetic video).
12346 The pixel format name.
12349 The sample aspect ratio of the input frame, expressed in the form
12350 @var{num}/@var{den}.
12353 The size of the input frame. For the syntax of this option, check the
12354 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12357 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
12358 for bottom field first).
12361 This is 1 if the frame is a key frame, 0 otherwise.
12364 The picture type of the input frame ("I" for an I-frame, "P" for a
12365 P-frame, "B" for a B-frame, or "?" for an unknown type).
12366 Also refer to the documentation of the @code{AVPictureType} enum and of
12367 the @code{av_get_picture_type_char} function defined in
12368 @file{libavutil/avutil.h}.
12371 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
12373 @item plane_checksum
12374 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
12375 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
12378 @section showpalette
12380 Displays the 256 colors palette of each frame. This filter is only relevant for
12381 @var{pal8} pixel format frames.
12383 It accepts the following option:
12387 Set the size of the box used to represent one palette color entry. Default is
12388 @code{30} (for a @code{30x30} pixel box).
12391 @section shuffleframes
12393 Reorder and/or duplicate and/or drop video frames.
12395 It accepts the following parameters:
12399 Set the destination indexes of input frames.
12400 This is space or '|' separated list of indexes that maps input frames to output
12401 frames. Number of indexes also sets maximal value that each index may have.
12402 '-1' index have special meaning and that is to drop frame.
12405 The first frame has the index 0. The default is to keep the input unchanged.
12407 @subsection Examples
12411 Swap second and third frame of every three frames of the input:
12413 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
12417 Swap 10th and 1st frame of every ten frames of the input:
12419 ffmpeg -i INPUT -vf "shuffleframes=9 1 2 3 4 5 6 7 8 0" OUTPUT
12423 @section shuffleplanes
12425 Reorder and/or duplicate video planes.
12427 It accepts the following parameters:
12432 The index of the input plane to be used as the first output plane.
12435 The index of the input plane to be used as the second output plane.
12438 The index of the input plane to be used as the third output plane.
12441 The index of the input plane to be used as the fourth output plane.
12445 The first plane has the index 0. The default is to keep the input unchanged.
12447 @subsection Examples
12451 Swap the second and third planes of the input:
12453 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
12457 @anchor{signalstats}
12458 @section signalstats
12459 Evaluate various visual metrics that assist in determining issues associated
12460 with the digitization of analog video media.
12462 By default the filter will log these metadata values:
12466 Display the minimal Y value contained within the input frame. Expressed in
12470 Display the Y value at the 10% percentile within the input frame. Expressed in
12474 Display the average Y value within the input frame. Expressed in range of
12478 Display the Y value at the 90% percentile within the input frame. Expressed in
12482 Display the maximum Y value contained within the input frame. Expressed in
12486 Display the minimal U value contained within the input frame. Expressed in
12490 Display the U value at the 10% percentile within the input frame. Expressed in
12494 Display the average U value within the input frame. Expressed in range of
12498 Display the U value at the 90% percentile within the input frame. Expressed in
12502 Display the maximum U value contained within the input frame. Expressed in
12506 Display the minimal V value contained within the input frame. Expressed in
12510 Display the V value at the 10% percentile within the input frame. Expressed in
12514 Display the average V value within the input frame. Expressed in range of
12518 Display the V value at the 90% percentile within the input frame. Expressed in
12522 Display the maximum V value contained within the input frame. Expressed in
12526 Display the minimal saturation value contained within the input frame.
12527 Expressed in range of [0-~181.02].
12530 Display the saturation value at the 10% percentile within the input frame.
12531 Expressed in range of [0-~181.02].
12534 Display the average saturation value within the input frame. Expressed in range
12538 Display the saturation value at the 90% percentile within the input frame.
12539 Expressed in range of [0-~181.02].
12542 Display the maximum saturation value contained within the input frame.
12543 Expressed in range of [0-~181.02].
12546 Display the median value for hue within the input frame. Expressed in range of
12550 Display the average value for hue within the input frame. Expressed in range of
12554 Display the average of sample value difference between all values of the Y
12555 plane in the current frame and corresponding values of the previous input frame.
12556 Expressed in range of [0-255].
12559 Display the average of sample value difference between all values of the U
12560 plane in the current frame and corresponding values of the previous input frame.
12561 Expressed in range of [0-255].
12564 Display the average of sample value difference between all values of the V
12565 plane in the current frame and corresponding values of the previous input frame.
12566 Expressed in range of [0-255].
12569 Display bit depth of Y plane in current frame.
12570 Expressed in range of [0-16].
12573 Display bit depth of U plane in current frame.
12574 Expressed in range of [0-16].
12577 Display bit depth of V plane in current frame.
12578 Expressed in range of [0-16].
12581 The filter accepts the following options:
12587 @option{stat} specify an additional form of image analysis.
12588 @option{out} output video with the specified type of pixel highlighted.
12590 Both options accept the following values:
12594 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
12595 unlike the neighboring pixels of the same field. Examples of temporal outliers
12596 include the results of video dropouts, head clogs, or tape tracking issues.
12599 Identify @var{vertical line repetition}. Vertical line repetition includes
12600 similar rows of pixels within a frame. In born-digital video vertical line
12601 repetition is common, but this pattern is uncommon in video digitized from an
12602 analog source. When it occurs in video that results from the digitization of an
12603 analog source it can indicate concealment from a dropout compensator.
12606 Identify pixels that fall outside of legal broadcast range.
12610 Set the highlight color for the @option{out} option. The default color is
12614 @subsection Examples
12618 Output data of various video metrics:
12620 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
12624 Output specific data about the minimum and maximum values of the Y plane per frame:
12626 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
12630 Playback video while highlighting pixels that are outside of broadcast range in red.
12632 ffplay example.mov -vf signalstats="out=brng:color=red"
12636 Playback video with signalstats metadata drawn over the frame.
12638 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
12641 The contents of signalstat_drawtext.txt used in the command are:
12644 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
12645 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
12646 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
12647 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
12655 Blur the input video without impacting the outlines.
12657 It accepts the following options:
12660 @item luma_radius, lr
12661 Set the luma radius. The option value must be a float number in
12662 the range [0.1,5.0] that specifies the variance of the gaussian filter
12663 used to blur the image (slower if larger). Default value is 1.0.
12665 @item luma_strength, ls
12666 Set the luma strength. The option value must be a float number
12667 in the range [-1.0,1.0] that configures the blurring. A value included
12668 in [0.0,1.0] will blur the image whereas a value included in
12669 [-1.0,0.0] will sharpen the image. Default value is 1.0.
12671 @item luma_threshold, lt
12672 Set the luma threshold used as a coefficient to determine
12673 whether a pixel should be blurred or not. The option value must be an
12674 integer in the range [-30,30]. A value of 0 will filter all the image,
12675 a value included in [0,30] will filter flat areas and a value included
12676 in [-30,0] will filter edges. Default value is 0.
12678 @item chroma_radius, cr
12679 Set the chroma radius. The option value must be a float number in
12680 the range [0.1,5.0] that specifies the variance of the gaussian filter
12681 used to blur the image (slower if larger). Default value is @option{luma_radius}.
12683 @item chroma_strength, cs
12684 Set the chroma strength. The option value must be a float number
12685 in the range [-1.0,1.0] that configures the blurring. A value included
12686 in [0.0,1.0] will blur the image whereas a value included in
12687 [-1.0,0.0] will sharpen the image. Default value is @option{luma_strength}.
12689 @item chroma_threshold, ct
12690 Set the chroma threshold used as a coefficient to determine
12691 whether a pixel should be blurred or not. The option value must be an
12692 integer in the range [-30,30]. A value of 0 will filter all the image,
12693 a value included in [0,30] will filter flat areas and a value included
12694 in [-30,0] will filter edges. Default value is @option{luma_threshold}.
12697 If a chroma option is not explicitly set, the corresponding luma value
12702 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
12704 This filter takes in input two input videos, the first input is
12705 considered the "main" source and is passed unchanged to the
12706 output. The second input is used as a "reference" video for computing
12709 Both video inputs must have the same resolution and pixel format for
12710 this filter to work correctly. Also it assumes that both inputs
12711 have the same number of frames, which are compared one by one.
12713 The filter stores the calculated SSIM of each frame.
12715 The description of the accepted parameters follows.
12718 @item stats_file, f
12719 If specified the filter will use the named file to save the SSIM of
12720 each individual frame. When filename equals "-" the data is sent to
12724 The file printed if @var{stats_file} is selected, contains a sequence of
12725 key/value pairs of the form @var{key}:@var{value} for each compared
12728 A description of each shown parameter follows:
12732 sequential number of the input frame, starting from 1
12734 @item Y, U, V, R, G, B
12735 SSIM of the compared frames for the component specified by the suffix.
12738 SSIM of the compared frames for the whole frame.
12741 Same as above but in dB representation.
12746 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
12747 [main][ref] ssim="stats_file=stats.log" [out]
12750 On this example the input file being processed is compared with the
12751 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
12752 is stored in @file{stats.log}.
12754 Another example with both psnr and ssim at same time:
12756 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
12761 Convert between different stereoscopic image formats.
12763 The filters accept the following options:
12767 Set stereoscopic image format of input.
12769 Available values for input image formats are:
12772 side by side parallel (left eye left, right eye right)
12775 side by side crosseye (right eye left, left eye right)
12778 side by side parallel with half width resolution
12779 (left eye left, right eye right)
12782 side by side crosseye with half width resolution
12783 (right eye left, left eye right)
12786 above-below (left eye above, right eye below)
12789 above-below (right eye above, left eye below)
12792 above-below with half height resolution
12793 (left eye above, right eye below)
12796 above-below with half height resolution
12797 (right eye above, left eye below)
12800 alternating frames (left eye first, right eye second)
12803 alternating frames (right eye first, left eye second)
12806 interleaved rows (left eye has top row, right eye starts on next row)
12809 interleaved rows (right eye has top row, left eye starts on next row)
12812 interleaved columns, left eye first
12815 interleaved columns, right eye first
12817 Default value is @samp{sbsl}.
12821 Set stereoscopic image format of output.
12825 side by side parallel (left eye left, right eye right)
12828 side by side crosseye (right eye left, left eye right)
12831 side by side parallel with half width resolution
12832 (left eye left, right eye right)
12835 side by side crosseye with half width resolution
12836 (right eye left, left eye right)
12839 above-below (left eye above, right eye below)
12842 above-below (right eye above, left eye below)
12845 above-below with half height resolution
12846 (left eye above, right eye below)
12849 above-below with half height resolution
12850 (right eye above, left eye below)
12853 alternating frames (left eye first, right eye second)
12856 alternating frames (right eye first, left eye second)
12859 interleaved rows (left eye has top row, right eye starts on next row)
12862 interleaved rows (right eye has top row, left eye starts on next row)
12865 anaglyph red/blue gray
12866 (red filter on left eye, blue filter on right eye)
12869 anaglyph red/green gray
12870 (red filter on left eye, green filter on right eye)
12873 anaglyph red/cyan gray
12874 (red filter on left eye, cyan filter on right eye)
12877 anaglyph red/cyan half colored
12878 (red filter on left eye, cyan filter on right eye)
12881 anaglyph red/cyan color
12882 (red filter on left eye, cyan filter on right eye)
12885 anaglyph red/cyan color optimized with the least squares projection of dubois
12886 (red filter on left eye, cyan filter on right eye)
12889 anaglyph green/magenta gray
12890 (green filter on left eye, magenta filter on right eye)
12893 anaglyph green/magenta half colored
12894 (green filter on left eye, magenta filter on right eye)
12897 anaglyph green/magenta colored
12898 (green filter on left eye, magenta filter on right eye)
12901 anaglyph green/magenta color optimized with the least squares projection of dubois
12902 (green filter on left eye, magenta filter on right eye)
12905 anaglyph yellow/blue gray
12906 (yellow filter on left eye, blue filter on right eye)
12909 anaglyph yellow/blue half colored
12910 (yellow filter on left eye, blue filter on right eye)
12913 anaglyph yellow/blue colored
12914 (yellow filter on left eye, blue filter on right eye)
12917 anaglyph yellow/blue color optimized with the least squares projection of dubois
12918 (yellow filter on left eye, blue filter on right eye)
12921 mono output (left eye only)
12924 mono output (right eye only)
12927 checkerboard, left eye first
12930 checkerboard, right eye first
12933 interleaved columns, left eye first
12936 interleaved columns, right eye first
12942 Default value is @samp{arcd}.
12945 @subsection Examples
12949 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
12955 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
12961 @section streamselect, astreamselect
12962 Select video or audio streams.
12964 The filter accepts the following options:
12968 Set number of inputs. Default is 2.
12971 Set input indexes to remap to outputs.
12974 @subsection Commands
12976 The @code{streamselect} and @code{astreamselect} filter supports the following
12981 Set input indexes to remap to outputs.
12984 @subsection Examples
12988 Select first 5 seconds 1st stream and rest of time 2nd stream:
12990 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
12994 Same as above, but for audio:
12996 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
13001 Apply sobel operator to input video stream.
13003 The filter accepts the following option:
13007 Set which planes will be processed, unprocessed planes will be copied.
13008 By default value 0xf, all planes will be processed.
13011 Set value which will be multiplied with filtered result.
13014 Set value which will be added to filtered result.
13020 Apply a simple postprocessing filter that compresses and decompresses the image
13021 at several (or - in the case of @option{quality} level @code{6} - all) shifts
13022 and average the results.
13024 The filter accepts the following options:
13028 Set quality. This option defines the number of levels for averaging. It accepts
13029 an integer in the range 0-6. If set to @code{0}, the filter will have no
13030 effect. A value of @code{6} means the higher quality. For each increment of
13031 that value the speed drops by a factor of approximately 2. Default value is
13035 Force a constant quantization parameter. If not set, the filter will use the QP
13036 from the video stream (if available).
13039 Set thresholding mode. Available modes are:
13043 Set hard thresholding (default).
13045 Set soft thresholding (better de-ringing effect, but likely blurrier).
13048 @item use_bframe_qp
13049 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
13050 option may cause flicker since the B-Frames have often larger QP. Default is
13051 @code{0} (not enabled).
13057 Draw subtitles on top of input video using the libass library.
13059 To enable compilation of this filter you need to configure FFmpeg with
13060 @code{--enable-libass}. This filter also requires a build with libavcodec and
13061 libavformat to convert the passed subtitles file to ASS (Advanced Substation
13062 Alpha) subtitles format.
13064 The filter accepts the following options:
13068 Set the filename of the subtitle file to read. It must be specified.
13070 @item original_size
13071 Specify the size of the original video, the video for which the ASS file
13072 was composed. For the syntax of this option, check the
13073 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13074 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
13075 correctly scale the fonts if the aspect ratio has been changed.
13078 Set a directory path containing fonts that can be used by the filter.
13079 These fonts will be used in addition to whatever the font provider uses.
13082 Set subtitles input character encoding. @code{subtitles} filter only. Only
13083 useful if not UTF-8.
13085 @item stream_index, si
13086 Set subtitles stream index. @code{subtitles} filter only.
13089 Override default style or script info parameters of the subtitles. It accepts a
13090 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
13093 If the first key is not specified, it is assumed that the first value
13094 specifies the @option{filename}.
13096 For example, to render the file @file{sub.srt} on top of the input
13097 video, use the command:
13102 which is equivalent to:
13104 subtitles=filename=sub.srt
13107 To render the default subtitles stream from file @file{video.mkv}, use:
13109 subtitles=video.mkv
13112 To render the second subtitles stream from that file, use:
13114 subtitles=video.mkv:si=1
13117 To make the subtitles stream from @file{sub.srt} appear in transparent green
13118 @code{DejaVu Serif}, use:
13120 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
13123 @section super2xsai
13125 Scale the input by 2x and smooth using the Super2xSaI (Scale and
13126 Interpolate) pixel art scaling algorithm.
13128 Useful for enlarging pixel art images without reducing sharpness.
13132 Swap two rectangular objects in video.
13134 This filter accepts the following options:
13144 Set 1st rect x coordinate.
13147 Set 1st rect y coordinate.
13150 Set 2nd rect x coordinate.
13153 Set 2nd rect y coordinate.
13155 All expressions are evaluated once for each frame.
13158 The all options are expressions containing the following constants:
13163 The input width and height.
13166 same as @var{w} / @var{h}
13169 input sample aspect ratio
13172 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
13175 The number of the input frame, starting from 0.
13178 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
13181 the position in the file of the input frame, NAN if unknown
13189 Apply telecine process to the video.
13191 This filter accepts the following options:
13200 The default value is @code{top}.
13204 A string of numbers representing the pulldown pattern you wish to apply.
13205 The default value is @code{23}.
13209 Some typical patterns:
13214 24p: 2332 (preferred)
13221 24p: 222222222223 ("Euro pulldown")
13228 Apply threshold effect to video stream.
13230 This filter needs four video streams to perform thresholding.
13231 First stream is stream we are filtering.
13232 Second stream is holding threshold values, third stream is holding min values,
13233 and last, fourth stream is holding max values.
13235 The filter accepts the following option:
13239 Set which planes will be processed, unprocessed planes will be copied.
13240 By default value 0xf, all planes will be processed.
13243 For example if first stream pixel's component value is less then threshold value
13244 of pixel component from 2nd threshold stream, third stream value will picked,
13245 otherwise fourth stream pixel component value will be picked.
13247 Using color source filter one can perform various types of thresholding:
13249 @subsection Examples
13253 Binary threshold, using gray color as threshold:
13255 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=black -f lavfi -i color=white -lavfi threshold output.avi
13259 Inverted binary threshold, using gray color as threshold:
13261 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -f lavfi -i color=black -lavfi threshold output.avi
13265 Truncate binary threshold, using gray color as threshold:
13267 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=gray -lavfi threshold output.avi
13271 Threshold to zero, using gray color as threshold:
13273 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -i 320x240.avi -lavfi threshold output.avi
13277 Inverted threshold to zero, using gray color as threshold:
13279 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=white -lavfi threshold output.avi
13284 Select the most representative frame in a given sequence of consecutive frames.
13286 The filter accepts the following options:
13290 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
13291 will pick one of them, and then handle the next batch of @var{n} frames until
13292 the end. Default is @code{100}.
13295 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
13296 value will result in a higher memory usage, so a high value is not recommended.
13298 @subsection Examples
13302 Extract one picture each 50 frames:
13308 Complete example of a thumbnail creation with @command{ffmpeg}:
13310 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
13316 Tile several successive frames together.
13318 The filter accepts the following options:
13323 Set the grid size (i.e. the number of lines and columns). For the syntax of
13324 this option, check the
13325 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13328 Set the maximum number of frames to render in the given area. It must be less
13329 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
13330 the area will be used.
13333 Set the outer border margin in pixels.
13336 Set the inner border thickness (i.e. the number of pixels between frames). For
13337 more advanced padding options (such as having different values for the edges),
13338 refer to the pad video filter.
13341 Specify the color of the unused area. For the syntax of this option, check the
13342 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
13346 @subsection Examples
13350 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
13352 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
13354 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
13355 duplicating each output frame to accommodate the originally detected frame
13359 Display @code{5} pictures in an area of @code{3x2} frames,
13360 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
13361 mixed flat and named options:
13363 tile=3x2:nb_frames=5:padding=7:margin=2
13367 @section tinterlace
13369 Perform various types of temporal field interlacing.
13371 Frames are counted starting from 1, so the first input frame is
13374 The filter accepts the following options:
13379 Specify the mode of the interlacing. This option can also be specified
13380 as a value alone. See below for a list of values for this option.
13382 Available values are:
13386 Move odd frames into the upper field, even into the lower field,
13387 generating a double height frame at half frame rate.
13391 Frame 1 Frame 2 Frame 3 Frame 4
13393 11111 22222 33333 44444
13394 11111 22222 33333 44444
13395 11111 22222 33333 44444
13396 11111 22222 33333 44444
13410 Only output odd frames, even frames are dropped, generating a frame with
13411 unchanged height at half frame rate.
13416 Frame 1 Frame 2 Frame 3 Frame 4
13418 11111 22222 33333 44444
13419 11111 22222 33333 44444
13420 11111 22222 33333 44444
13421 11111 22222 33333 44444
13431 Only output even frames, odd frames are dropped, generating a frame with
13432 unchanged height at half frame rate.
13437 Frame 1 Frame 2 Frame 3 Frame 4
13439 11111 22222 33333 44444
13440 11111 22222 33333 44444
13441 11111 22222 33333 44444
13442 11111 22222 33333 44444
13452 Expand each frame to full height, but pad alternate lines with black,
13453 generating a frame with double height at the same input frame rate.
13458 Frame 1 Frame 2 Frame 3 Frame 4
13460 11111 22222 33333 44444
13461 11111 22222 33333 44444
13462 11111 22222 33333 44444
13463 11111 22222 33333 44444
13466 11111 ..... 33333 .....
13467 ..... 22222 ..... 44444
13468 11111 ..... 33333 .....
13469 ..... 22222 ..... 44444
13470 11111 ..... 33333 .....
13471 ..... 22222 ..... 44444
13472 11111 ..... 33333 .....
13473 ..... 22222 ..... 44444
13477 @item interleave_top, 4
13478 Interleave the upper field from odd frames with the lower field from
13479 even frames, generating a frame with unchanged height at half frame rate.
13484 Frame 1 Frame 2 Frame 3 Frame 4
13486 11111<- 22222 33333<- 44444
13487 11111 22222<- 33333 44444<-
13488 11111<- 22222 33333<- 44444
13489 11111 22222<- 33333 44444<-
13499 @item interleave_bottom, 5
13500 Interleave the lower field from odd frames with the upper field from
13501 even frames, generating a frame with unchanged height at half frame rate.
13506 Frame 1 Frame 2 Frame 3 Frame 4
13508 11111 22222<- 33333 44444<-
13509 11111<- 22222 33333<- 44444
13510 11111 22222<- 33333 44444<-
13511 11111<- 22222 33333<- 44444
13521 @item interlacex2, 6
13522 Double frame rate with unchanged height. Frames are inserted each
13523 containing the second temporal field from the previous input frame and
13524 the first temporal field from the next input frame. This mode relies on
13525 the top_field_first flag. Useful for interlaced video displays with no
13526 field synchronisation.
13531 Frame 1 Frame 2 Frame 3 Frame 4
13533 11111 22222 33333 44444
13534 11111 22222 33333 44444
13535 11111 22222 33333 44444
13536 11111 22222 33333 44444
13539 11111 22222 22222 33333 33333 44444 44444
13540 11111 11111 22222 22222 33333 33333 44444
13541 11111 22222 22222 33333 33333 44444 44444
13542 11111 11111 22222 22222 33333 33333 44444
13547 Move odd frames into the upper field, even into the lower field,
13548 generating a double height frame at same frame rate.
13553 Frame 1 Frame 2 Frame 3 Frame 4
13555 11111 22222 33333 44444
13556 11111 22222 33333 44444
13557 11111 22222 33333 44444
13558 11111 22222 33333 44444
13561 11111 33333 33333 55555
13562 22222 22222 44444 44444
13563 11111 33333 33333 55555
13564 22222 22222 44444 44444
13565 11111 33333 33333 55555
13566 22222 22222 44444 44444
13567 11111 33333 33333 55555
13568 22222 22222 44444 44444
13573 Numeric values are deprecated but are accepted for backward
13574 compatibility reasons.
13576 Default mode is @code{merge}.
13579 Specify flags influencing the filter process.
13581 Available value for @var{flags} is:
13584 @item low_pass_filter, vlfp
13585 Enable vertical low-pass filtering in the filter.
13586 Vertical low-pass filtering is required when creating an interlaced
13587 destination from a progressive source which contains high-frequency
13588 vertical detail. Filtering will reduce interlace 'twitter' and Moire
13591 Vertical low-pass filtering can only be enabled for @option{mode}
13592 @var{interleave_top} and @var{interleave_bottom}.
13599 Transpose rows with columns in the input video and optionally flip it.
13601 It accepts the following parameters:
13606 Specify the transposition direction.
13608 Can assume the following values:
13610 @item 0, 4, cclock_flip
13611 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
13619 Rotate by 90 degrees clockwise, that is:
13627 Rotate by 90 degrees counterclockwise, that is:
13634 @item 3, 7, clock_flip
13635 Rotate by 90 degrees clockwise and vertically flip, that is:
13643 For values between 4-7, the transposition is only done if the input
13644 video geometry is portrait and not landscape. These values are
13645 deprecated, the @code{passthrough} option should be used instead.
13647 Numerical values are deprecated, and should be dropped in favor of
13648 symbolic constants.
13651 Do not apply the transposition if the input geometry matches the one
13652 specified by the specified value. It accepts the following values:
13655 Always apply transposition.
13657 Preserve portrait geometry (when @var{height} >= @var{width}).
13659 Preserve landscape geometry (when @var{width} >= @var{height}).
13662 Default value is @code{none}.
13665 For example to rotate by 90 degrees clockwise and preserve portrait
13668 transpose=dir=1:passthrough=portrait
13671 The command above can also be specified as:
13673 transpose=1:portrait
13677 Trim the input so that the output contains one continuous subpart of the input.
13679 It accepts the following parameters:
13682 Specify the time of the start of the kept section, i.e. the frame with the
13683 timestamp @var{start} will be the first frame in the output.
13686 Specify the time of the first frame that will be dropped, i.e. the frame
13687 immediately preceding the one with the timestamp @var{end} will be the last
13688 frame in the output.
13691 This is the same as @var{start}, except this option sets the start timestamp
13692 in timebase units instead of seconds.
13695 This is the same as @var{end}, except this option sets the end timestamp
13696 in timebase units instead of seconds.
13699 The maximum duration of the output in seconds.
13702 The number of the first frame that should be passed to the output.
13705 The number of the first frame that should be dropped.
13708 @option{start}, @option{end}, and @option{duration} are expressed as time
13709 duration specifications; see
13710 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
13711 for the accepted syntax.
13713 Note that the first two sets of the start/end options and the @option{duration}
13714 option look at the frame timestamp, while the _frame variants simply count the
13715 frames that pass through the filter. Also note that this filter does not modify
13716 the timestamps. If you wish for the output timestamps to start at zero, insert a
13717 setpts filter after the trim filter.
13719 If multiple start or end options are set, this filter tries to be greedy and
13720 keep all the frames that match at least one of the specified constraints. To keep
13721 only the part that matches all the constraints at once, chain multiple trim
13724 The defaults are such that all the input is kept. So it is possible to set e.g.
13725 just the end values to keep everything before the specified time.
13730 Drop everything except the second minute of input:
13732 ffmpeg -i INPUT -vf trim=60:120
13736 Keep only the first second:
13738 ffmpeg -i INPUT -vf trim=duration=1
13747 Sharpen or blur the input video.
13749 It accepts the following parameters:
13752 @item luma_msize_x, lx
13753 Set the luma matrix horizontal size. It must be an odd integer between
13754 3 and 23. The default value is 5.
13756 @item luma_msize_y, ly
13757 Set the luma matrix vertical size. It must be an odd integer between 3
13758 and 23. The default value is 5.
13760 @item luma_amount, la
13761 Set the luma effect strength. It must be a floating point number, reasonable
13762 values lay between -1.5 and 1.5.
13764 Negative values will blur the input video, while positive values will
13765 sharpen it, a value of zero will disable the effect.
13767 Default value is 1.0.
13769 @item chroma_msize_x, cx
13770 Set the chroma matrix horizontal size. It must be an odd integer
13771 between 3 and 23. The default value is 5.
13773 @item chroma_msize_y, cy
13774 Set the chroma matrix vertical size. It must be an odd integer
13775 between 3 and 23. The default value is 5.
13777 @item chroma_amount, ca
13778 Set the chroma effect strength. It must be a floating point number, reasonable
13779 values lay between -1.5 and 1.5.
13781 Negative values will blur the input video, while positive values will
13782 sharpen it, a value of zero will disable the effect.
13784 Default value is 0.0.
13787 If set to 1, specify using OpenCL capabilities, only available if
13788 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
13792 All parameters are optional and default to the equivalent of the
13793 string '5:5:1.0:5:5:0.0'.
13795 @subsection Examples
13799 Apply strong luma sharpen effect:
13801 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
13805 Apply a strong blur of both luma and chroma parameters:
13807 unsharp=7:7:-2:7:7:-2
13813 Apply ultra slow/simple postprocessing filter that compresses and decompresses
13814 the image at several (or - in the case of @option{quality} level @code{8} - all)
13815 shifts and average the results.
13817 The way this differs from the behavior of spp is that uspp actually encodes &
13818 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
13819 DCT similar to MJPEG.
13821 The filter accepts the following options:
13825 Set quality. This option defines the number of levels for averaging. It accepts
13826 an integer in the range 0-8. If set to @code{0}, the filter will have no
13827 effect. A value of @code{8} means the higher quality. For each increment of
13828 that value the speed drops by a factor of approximately 2. Default value is
13832 Force a constant quantization parameter. If not set, the filter will use the QP
13833 from the video stream (if available).
13836 @section vaguedenoiser
13838 Apply a wavelet based denoiser.
13840 It transforms each frame from the video input into the wavelet domain,
13841 using Cohen-Daubechies-Feauveau 9/7. Then it applies some filtering to
13842 the obtained coefficients. It does an inverse wavelet transform after.
13843 Due to wavelet properties, it should give a nice smoothed result, and
13844 reduced noise, without blurring picture features.
13846 This filter accepts the following options:
13850 The filtering strength. The higher, the more filtered the video will be.
13851 Hard thresholding can use a higher threshold than soft thresholding
13852 before the video looks overfiltered.
13855 The filtering method the filter will use.
13857 It accepts the following values:
13860 All values under the threshold will be zeroed.
13863 All values under the threshold will be zeroed. All values above will be
13864 reduced by the threshold.
13867 Scales or nullifies coefficients - intermediary between (more) soft and
13868 (less) hard thresholding.
13872 Number of times, the wavelet will decompose the picture. Picture can't
13873 be decomposed beyond a particular point (typically, 8 for a 640x480
13874 frame - as 2^9 = 512 > 480)
13877 Partial of full denoising (limited coefficients shrinking), from 0 to 100.
13880 A list of the planes to process. By default all planes are processed.
13883 @section vectorscope
13885 Display 2 color component values in the two dimensional graph (which is called
13888 This filter accepts the following options:
13892 Set vectorscope mode.
13894 It accepts the following values:
13897 Gray values are displayed on graph, higher brightness means more pixels have
13898 same component color value on location in graph. This is the default mode.
13901 Gray values are displayed on graph. Surrounding pixels values which are not
13902 present in video frame are drawn in gradient of 2 color components which are
13903 set by option @code{x} and @code{y}. The 3rd color component is static.
13906 Actual color components values present in video frame are displayed on graph.
13909 Similar as color2 but higher frequency of same values @code{x} and @code{y}
13910 on graph increases value of another color component, which is luminance by
13911 default values of @code{x} and @code{y}.
13914 Actual colors present in video frame are displayed on graph. If two different
13915 colors map to same position on graph then color with higher value of component
13916 not present in graph is picked.
13919 Gray values are displayed on graph. Similar to @code{color} but with 3rd color
13920 component picked from radial gradient.
13924 Set which color component will be represented on X-axis. Default is @code{1}.
13927 Set which color component will be represented on Y-axis. Default is @code{2}.
13930 Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
13931 of color component which represents frequency of (X, Y) location in graph.
13936 No envelope, this is default.
13939 Instant envelope, even darkest single pixel will be clearly highlighted.
13942 Hold maximum and minimum values presented in graph over time. This way you
13943 can still spot out of range values without constantly looking at vectorscope.
13946 Peak and instant envelope combined together.
13950 Set what kind of graticule to draw.
13958 Set graticule opacity.
13961 Set graticule flags.
13965 Draw graticule for white point.
13968 Draw graticule for black point.
13971 Draw color points short names.
13975 Set background opacity.
13977 @item lthreshold, l
13978 Set low threshold for color component not represented on X or Y axis.
13979 Values lower than this value will be ignored. Default is 0.
13980 Note this value is multiplied with actual max possible value one pixel component
13981 can have. So for 8-bit input and low threshold value of 0.1 actual threshold
13984 @item hthreshold, h
13985 Set high threshold for color component not represented on X or Y axis.
13986 Values higher than this value will be ignored. Default is 1.
13987 Note this value is multiplied with actual max possible value one pixel component
13988 can have. So for 8-bit input and high threshold value of 0.9 actual threshold
13989 is 0.9 * 255 = 230.
13991 @item colorspace, c
13992 Set what kind of colorspace to use when drawing graticule.
14001 @anchor{vidstabdetect}
14002 @section vidstabdetect
14004 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
14005 @ref{vidstabtransform} for pass 2.
14007 This filter generates a file with relative translation and rotation
14008 transform information about subsequent frames, which is then used by
14009 the @ref{vidstabtransform} filter.
14011 To enable compilation of this filter you need to configure FFmpeg with
14012 @code{--enable-libvidstab}.
14014 This filter accepts the following options:
14018 Set the path to the file used to write the transforms information.
14019 Default value is @file{transforms.trf}.
14022 Set how shaky the video is and how quick the camera is. It accepts an
14023 integer in the range 1-10, a value of 1 means little shakiness, a
14024 value of 10 means strong shakiness. Default value is 5.
14027 Set the accuracy of the detection process. It must be a value in the
14028 range 1-15. A value of 1 means low accuracy, a value of 15 means high
14029 accuracy. Default value is 15.
14032 Set stepsize of the search process. The region around minimum is
14033 scanned with 1 pixel resolution. Default value is 6.
14036 Set minimum contrast. Below this value a local measurement field is
14037 discarded. Must be a floating point value in the range 0-1. Default
14041 Set reference frame number for tripod mode.
14043 If enabled, the motion of the frames is compared to a reference frame
14044 in the filtered stream, identified by the specified number. The idea
14045 is to compensate all movements in a more-or-less static scene and keep
14046 the camera view absolutely still.
14048 If set to 0, it is disabled. The frames are counted starting from 1.
14051 Show fields and transforms in the resulting frames. It accepts an
14052 integer in the range 0-2. Default value is 0, which disables any
14056 @subsection Examples
14060 Use default values:
14066 Analyze strongly shaky movie and put the results in file
14067 @file{mytransforms.trf}:
14069 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
14073 Visualize the result of internal transformations in the resulting
14076 vidstabdetect=show=1
14080 Analyze a video with medium shakiness using @command{ffmpeg}:
14082 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
14086 @anchor{vidstabtransform}
14087 @section vidstabtransform
14089 Video stabilization/deshaking: pass 2 of 2,
14090 see @ref{vidstabdetect} for pass 1.
14092 Read a file with transform information for each frame and
14093 apply/compensate them. Together with the @ref{vidstabdetect}
14094 filter this can be used to deshake videos. See also
14095 @url{http://public.hronopik.de/vid.stab}. It is important to also use
14096 the @ref{unsharp} filter, see below.
14098 To enable compilation of this filter you need to configure FFmpeg with
14099 @code{--enable-libvidstab}.
14101 @subsection Options
14105 Set path to the file used to read the transforms. Default value is
14106 @file{transforms.trf}.
14109 Set the number of frames (value*2 + 1) used for lowpass filtering the
14110 camera movements. Default value is 10.
14112 For example a number of 10 means that 21 frames are used (10 in the
14113 past and 10 in the future) to smoothen the motion in the video. A
14114 larger value leads to a smoother video, but limits the acceleration of
14115 the camera (pan/tilt movements). 0 is a special case where a static
14116 camera is simulated.
14119 Set the camera path optimization algorithm.
14121 Accepted values are:
14124 gaussian kernel low-pass filter on camera motion (default)
14126 averaging on transformations
14130 Set maximal number of pixels to translate frames. Default value is -1,
14134 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
14135 value is -1, meaning no limit.
14138 Specify how to deal with borders that may be visible due to movement
14141 Available values are:
14144 keep image information from previous frame (default)
14146 fill the border black
14150 Invert transforms if set to 1. Default value is 0.
14153 Consider transforms as relative to previous frame if set to 1,
14154 absolute if set to 0. Default value is 0.
14157 Set percentage to zoom. A positive value will result in a zoom-in
14158 effect, a negative value in a zoom-out effect. Default value is 0 (no
14162 Set optimal zooming to avoid borders.
14164 Accepted values are:
14169 optimal static zoom value is determined (only very strong movements
14170 will lead to visible borders) (default)
14172 optimal adaptive zoom value is determined (no borders will be
14173 visible), see @option{zoomspeed}
14176 Note that the value given at zoom is added to the one calculated here.
14179 Set percent to zoom maximally each frame (enabled when
14180 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
14184 Specify type of interpolation.
14186 Available values are:
14191 linear only horizontal
14193 linear in both directions (default)
14195 cubic in both directions (slow)
14199 Enable virtual tripod mode if set to 1, which is equivalent to
14200 @code{relative=0:smoothing=0}. Default value is 0.
14202 Use also @code{tripod} option of @ref{vidstabdetect}.
14205 Increase log verbosity if set to 1. Also the detected global motions
14206 are written to the temporary file @file{global_motions.trf}. Default
14210 @subsection Examples
14214 Use @command{ffmpeg} for a typical stabilization with default values:
14216 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
14219 Note the use of the @ref{unsharp} filter which is always recommended.
14222 Zoom in a bit more and load transform data from a given file:
14224 vidstabtransform=zoom=5:input="mytransforms.trf"
14228 Smoothen the video even more:
14230 vidstabtransform=smoothing=30
14236 Flip the input video vertically.
14238 For example, to vertically flip a video with @command{ffmpeg}:
14240 ffmpeg -i in.avi -vf "vflip" out.avi
14246 Make or reverse a natural vignetting effect.
14248 The filter accepts the following options:
14252 Set lens angle expression as a number of radians.
14254 The value is clipped in the @code{[0,PI/2]} range.
14256 Default value: @code{"PI/5"}
14260 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
14264 Set forward/backward mode.
14266 Available modes are:
14269 The larger the distance from the central point, the darker the image becomes.
14272 The larger the distance from the central point, the brighter the image becomes.
14273 This can be used to reverse a vignette effect, though there is no automatic
14274 detection to extract the lens @option{angle} and other settings (yet). It can
14275 also be used to create a burning effect.
14278 Default value is @samp{forward}.
14281 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
14283 It accepts the following values:
14286 Evaluate expressions only once during the filter initialization.
14289 Evaluate expressions for each incoming frame. This is way slower than the
14290 @samp{init} mode since it requires all the scalers to be re-computed, but it
14291 allows advanced dynamic expressions.
14294 Default value is @samp{init}.
14297 Set dithering to reduce the circular banding effects. Default is @code{1}
14301 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
14302 Setting this value to the SAR of the input will make a rectangular vignetting
14303 following the dimensions of the video.
14305 Default is @code{1/1}.
14308 @subsection Expressions
14310 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
14311 following parameters.
14316 input width and height
14319 the number of input frame, starting from 0
14322 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
14323 @var{TB} units, NAN if undefined
14326 frame rate of the input video, NAN if the input frame rate is unknown
14329 the PTS (Presentation TimeStamp) of the filtered video frame,
14330 expressed in seconds, NAN if undefined
14333 time base of the input video
14337 @subsection Examples
14341 Apply simple strong vignetting effect:
14347 Make a flickering vignetting:
14349 vignette='PI/4+random(1)*PI/50':eval=frame
14355 Stack input videos vertically.
14357 All streams must be of same pixel format and of same width.
14359 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
14360 to create same output.
14362 The filter accept the following option:
14366 Set number of input streams. Default is 2.
14369 If set to 1, force the output to terminate when the shortest input
14370 terminates. Default value is 0.
14375 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
14376 Deinterlacing Filter").
14378 Based on the process described by Martin Weston for BBC R&D, and
14379 implemented based on the de-interlace algorithm written by Jim
14380 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
14381 uses filter coefficients calculated by BBC R&D.
14383 There are two sets of filter coefficients, so called "simple":
14384 and "complex". Which set of filter coefficients is used can
14385 be set by passing an optional parameter:
14389 Set the interlacing filter coefficients. Accepts one of the following values:
14393 Simple filter coefficient set.
14395 More-complex filter coefficient set.
14397 Default value is @samp{complex}.
14400 Specify which frames to deinterlace. Accept one of the following values:
14404 Deinterlace all frames,
14406 Only deinterlace frames marked as interlaced.
14409 Default value is @samp{all}.
14413 Video waveform monitor.
14415 The waveform monitor plots color component intensity. By default luminance
14416 only. Each column of the waveform corresponds to a column of pixels in the
14419 It accepts the following options:
14423 Can be either @code{row}, or @code{column}. Default is @code{column}.
14424 In row mode, the graph on the left side represents color component value 0 and
14425 the right side represents value = 255. In column mode, the top side represents
14426 color component value = 0 and bottom side represents value = 255.
14429 Set intensity. Smaller values are useful to find out how many values of the same
14430 luminance are distributed across input rows/columns.
14431 Default value is @code{0.04}. Allowed range is [0, 1].
14434 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
14435 In mirrored mode, higher values will be represented on the left
14436 side for @code{row} mode and at the top for @code{column} mode. Default is
14437 @code{1} (mirrored).
14441 It accepts the following values:
14444 Presents information identical to that in the @code{parade}, except
14445 that the graphs representing color components are superimposed directly
14448 This display mode makes it easier to spot relative differences or similarities
14449 in overlapping areas of the color components that are supposed to be identical,
14450 such as neutral whites, grays, or blacks.
14453 Display separate graph for the color components side by side in
14454 @code{row} mode or one below the other in @code{column} mode.
14457 Display separate graph for the color components side by side in
14458 @code{column} mode or one below the other in @code{row} mode.
14460 Using this display mode makes it easy to spot color casts in the highlights
14461 and shadows of an image, by comparing the contours of the top and the bottom
14462 graphs of each waveform. Since whites, grays, and blacks are characterized
14463 by exactly equal amounts of red, green, and blue, neutral areas of the picture
14464 should display three waveforms of roughly equal width/height. If not, the
14465 correction is easy to perform by making level adjustments the three waveforms.
14467 Default is @code{stack}.
14469 @item components, c
14470 Set which color components to display. Default is 1, which means only luminance
14471 or red color component if input is in RGB colorspace. If is set for example to
14472 7 it will display all 3 (if) available color components.
14477 No envelope, this is default.
14480 Instant envelope, minimum and maximum values presented in graph will be easily
14481 visible even with small @code{step} value.
14484 Hold minimum and maximum values presented in graph across time. This way you
14485 can still spot out of range values without constantly looking at waveforms.
14488 Peak and instant envelope combined together.
14494 No filtering, this is default.
14497 Luma and chroma combined together.
14500 Similar as above, but shows difference between blue and red chroma.
14503 Displays only chroma.
14506 Displays actual color value on waveform.
14509 Similar as above, but with luma showing frequency of chroma values.
14513 Set which graticule to display.
14517 Do not display graticule.
14520 Display green graticule showing legal broadcast ranges.
14524 Set graticule opacity.
14527 Set graticule flags.
14531 Draw numbers above lines. By default enabled.
14534 Draw dots instead of lines.
14538 Set scale used for displaying graticule.
14545 Default is digital.
14548 Set background opacity.
14553 The @code{weave} takes a field-based video input and join
14554 each two sequential fields into single frame, producing a new double
14555 height clip with half the frame rate and half the frame count.
14557 It accepts the following option:
14561 Set first field. Available values are:
14565 Set the frame as top-field-first.
14568 Set the frame as bottom-field-first.
14572 @subsection Examples
14576 Interlace video using @ref{select} and @ref{separatefields} filter:
14578 separatefields,select=eq(mod(n,4),0)+eq(mod(n,4),3),weave
14583 Apply the xBR high-quality magnification filter which is designed for pixel
14584 art. It follows a set of edge-detection rules, see
14585 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
14587 It accepts the following option:
14591 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
14592 @code{3xBR} and @code{4} for @code{4xBR}.
14593 Default is @code{3}.
14599 Deinterlace the input video ("yadif" means "yet another deinterlacing
14602 It accepts the following parameters:
14608 The interlacing mode to adopt. It accepts one of the following values:
14611 @item 0, send_frame
14612 Output one frame for each frame.
14613 @item 1, send_field
14614 Output one frame for each field.
14615 @item 2, send_frame_nospatial
14616 Like @code{send_frame}, but it skips the spatial interlacing check.
14617 @item 3, send_field_nospatial
14618 Like @code{send_field}, but it skips the spatial interlacing check.
14621 The default value is @code{send_frame}.
14624 The picture field parity assumed for the input interlaced video. It accepts one
14625 of the following values:
14629 Assume the top field is first.
14631 Assume the bottom field is first.
14633 Enable automatic detection of field parity.
14636 The default value is @code{auto}.
14637 If the interlacing is unknown or the decoder does not export this information,
14638 top field first will be assumed.
14641 Specify which frames to deinterlace. Accept one of the following
14646 Deinterlace all frames.
14647 @item 1, interlaced
14648 Only deinterlace frames marked as interlaced.
14651 The default value is @code{all}.
14656 Apply Zoom & Pan effect.
14658 This filter accepts the following options:
14662 Set the zoom expression. Default is 1.
14666 Set the x and y expression. Default is 0.
14669 Set the duration expression in number of frames.
14670 This sets for how many number of frames effect will last for
14671 single input image.
14674 Set the output image size, default is 'hd720'.
14677 Set the output frame rate, default is '25'.
14680 Each expression can contain the following constants:
14699 Output frame count.
14703 Last calculated 'x' and 'y' position from 'x' and 'y' expression
14704 for current input frame.
14708 'x' and 'y' of last output frame of previous input frame or 0 when there was
14709 not yet such frame (first input frame).
14712 Last calculated zoom from 'z' expression for current input frame.
14715 Last calculated zoom of last output frame of previous input frame.
14718 Number of output frames for current input frame. Calculated from 'd' expression
14719 for each input frame.
14722 number of output frames created for previous input frame
14725 Rational number: input width / input height
14728 sample aspect ratio
14731 display aspect ratio
14735 @subsection Examples
14739 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
14741 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
14745 Zoom-in up to 1.5 and pan always at center of picture:
14747 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14751 Same as above but without pausing:
14753 zoompan=z='min(max(zoom,pzoom)+0.0015,1.5)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14758 Scale (resize) the input video, using the z.lib library:
14759 https://github.com/sekrit-twc/zimg.
14761 The zscale filter forces the output display aspect ratio to be the same
14762 as the input, by changing the output sample aspect ratio.
14764 If the input image format is different from the format requested by
14765 the next filter, the zscale filter will convert the input to the
14768 @subsection Options
14769 The filter accepts the following options.
14774 Set the output video dimension expression. Default value is the input
14777 If the @var{width} or @var{w} is 0, the input width is used for the output.
14778 If the @var{height} or @var{h} is 0, the input height is used for the output.
14780 If one of the values is -1, the zscale filter will use a value that
14781 maintains the aspect ratio of the input image, calculated from the
14782 other specified dimension. If both of them are -1, the input size is
14785 If one of the values is -n with n > 1, the zscale filter will also use a value
14786 that maintains the aspect ratio of the input image, calculated from the other
14787 specified dimension. After that it will, however, make sure that the calculated
14788 dimension is divisible by n and adjust the value if necessary.
14790 See below for the list of accepted constants for use in the dimension
14794 Set the video size. For the syntax of this option, check the
14795 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14798 Set the dither type.
14800 Possible values are:
14805 @item error_diffusion
14811 Set the resize filter type.
14813 Possible values are:
14823 Default is bilinear.
14826 Set the color range.
14828 Possible values are:
14835 Default is same as input.
14838 Set the color primaries.
14840 Possible values are:
14850 Default is same as input.
14853 Set the transfer characteristics.
14855 Possible values are:
14869 Default is same as input.
14872 Set the colorspace matrix.
14874 Possible value are:
14885 Default is same as input.
14888 Set the input color range.
14890 Possible values are:
14897 Default is same as input.
14899 @item primariesin, pin
14900 Set the input color primaries.
14902 Possible values are:
14912 Default is same as input.
14914 @item transferin, tin
14915 Set the input transfer characteristics.
14917 Possible values are:
14928 Default is same as input.
14930 @item matrixin, min
14931 Set the input colorspace matrix.
14933 Possible value are:
14945 Set the output chroma location.
14947 Possible values are:
14958 @item chromalin, cin
14959 Set the input chroma location.
14961 Possible values are:
14973 Set the nominal peak luminance.
14976 The values of the @option{w} and @option{h} options are expressions
14977 containing the following constants:
14982 The input width and height
14986 These are the same as @var{in_w} and @var{in_h}.
14990 The output (scaled) width and height
14994 These are the same as @var{out_w} and @var{out_h}
14997 The same as @var{iw} / @var{ih}
15000 input sample aspect ratio
15003 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
15007 horizontal and vertical input chroma subsample values. For example for the
15008 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15012 horizontal and vertical output chroma subsample values. For example for the
15013 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15019 @c man end VIDEO FILTERS
15021 @chapter Video Sources
15022 @c man begin VIDEO SOURCES
15024 Below is a description of the currently available video sources.
15028 Buffer video frames, and make them available to the filter chain.
15030 This source is mainly intended for a programmatic use, in particular
15031 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
15033 It accepts the following parameters:
15038 Specify the size (width and height) of the buffered video frames. For the
15039 syntax of this option, check the
15040 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15043 The input video width.
15046 The input video height.
15049 A string representing the pixel format of the buffered video frames.
15050 It may be a number corresponding to a pixel format, or a pixel format
15054 Specify the timebase assumed by the timestamps of the buffered frames.
15057 Specify the frame rate expected for the video stream.
15059 @item pixel_aspect, sar
15060 The sample (pixel) aspect ratio of the input video.
15063 Specify the optional parameters to be used for the scale filter which
15064 is automatically inserted when an input change is detected in the
15065 input size or format.
15067 @item hw_frames_ctx
15068 When using a hardware pixel format, this should be a reference to an
15069 AVHWFramesContext describing input frames.
15074 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
15077 will instruct the source to accept video frames with size 320x240 and
15078 with format "yuv410p", assuming 1/24 as the timestamps timebase and
15079 square pixels (1:1 sample aspect ratio).
15080 Since the pixel format with name "yuv410p" corresponds to the number 6
15081 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
15082 this example corresponds to:
15084 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
15087 Alternatively, the options can be specified as a flat string, but this
15088 syntax is deprecated:
15090 @var{width}:@var{height}:@var{pix_fmt}:@var{time_base.num}:@var{time_base.den}:@var{pixel_aspect.num}:@var{pixel_aspect.den}[:@var{sws_param}]
15094 Create a pattern generated by an elementary cellular automaton.
15096 The initial state of the cellular automaton can be defined through the
15097 @option{filename} and @option{pattern} options. If such options are
15098 not specified an initial state is created randomly.
15100 At each new frame a new row in the video is filled with the result of
15101 the cellular automaton next generation. The behavior when the whole
15102 frame is filled is defined by the @option{scroll} option.
15104 This source accepts the following options:
15108 Read the initial cellular automaton state, i.e. the starting row, from
15109 the specified file.
15110 In the file, each non-whitespace character is considered an alive
15111 cell, a newline will terminate the row, and further characters in the
15112 file will be ignored.
15115 Read the initial cellular automaton state, i.e. the starting row, from
15116 the specified string.
15118 Each non-whitespace character in the string is considered an alive
15119 cell, a newline will terminate the row, and further characters in the
15120 string will be ignored.
15123 Set the video rate, that is the number of frames generated per second.
15126 @item random_fill_ratio, ratio
15127 Set the random fill ratio for the initial cellular automaton row. It
15128 is a floating point number value ranging from 0 to 1, defaults to
15131 This option is ignored when a file or a pattern is specified.
15133 @item random_seed, seed
15134 Set the seed for filling randomly the initial row, must be an integer
15135 included between 0 and UINT32_MAX. If not specified, or if explicitly
15136 set to -1, the filter will try to use a good random seed on a best
15140 Set the cellular automaton rule, it is a number ranging from 0 to 255.
15141 Default value is 110.
15144 Set the size of the output video. For the syntax of this option, check the
15145 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15147 If @option{filename} or @option{pattern} is specified, the size is set
15148 by default to the width of the specified initial state row, and the
15149 height is set to @var{width} * PHI.
15151 If @option{size} is set, it must contain the width of the specified
15152 pattern string, and the specified pattern will be centered in the
15155 If a filename or a pattern string is not specified, the size value
15156 defaults to "320x518" (used for a randomly generated initial state).
15159 If set to 1, scroll the output upward when all the rows in the output
15160 have been already filled. If set to 0, the new generated row will be
15161 written over the top row just after the bottom row is filled.
15164 @item start_full, full
15165 If set to 1, completely fill the output with generated rows before
15166 outputting the first frame.
15167 This is the default behavior, for disabling set the value to 0.
15170 If set to 1, stitch the left and right row edges together.
15171 This is the default behavior, for disabling set the value to 0.
15174 @subsection Examples
15178 Read the initial state from @file{pattern}, and specify an output of
15181 cellauto=f=pattern:s=200x400
15185 Generate a random initial row with a width of 200 cells, with a fill
15188 cellauto=ratio=2/3:s=200x200
15192 Create a pattern generated by rule 18 starting by a single alive cell
15193 centered on an initial row with width 100:
15195 cellauto=p=@@:s=100x400:full=0:rule=18
15199 Specify a more elaborated initial pattern:
15201 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
15206 @anchor{coreimagesrc}
15207 @section coreimagesrc
15208 Video source generated on GPU using Apple's CoreImage API on OSX.
15210 This video source is a specialized version of the @ref{coreimage} video filter.
15211 Use a core image generator at the beginning of the applied filterchain to
15212 generate the content.
15214 The coreimagesrc video source accepts the following options:
15216 @item list_generators
15217 List all available generators along with all their respective options as well as
15218 possible minimum and maximum values along with the default values.
15220 list_generators=true
15224 Specify the size of the sourced video. For the syntax of this option, check the
15225 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15226 The default value is @code{320x240}.
15229 Specify the frame rate of the sourced video, as the number of frames
15230 generated per second. It has to be a string in the format
15231 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15232 number or a valid video frame rate abbreviation. The default value is
15236 Set the sample aspect ratio of the sourced video.
15239 Set the duration of the sourced video. See
15240 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15241 for the accepted syntax.
15243 If not specified, or the expressed duration is negative, the video is
15244 supposed to be generated forever.
15247 Additionally, all options of the @ref{coreimage} video filter are accepted.
15248 A complete filterchain can be used for further processing of the
15249 generated input without CPU-HOST transfer. See @ref{coreimage} documentation
15250 and examples for details.
15252 @subsection Examples
15257 Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
15258 given as complete and escaped command-line for Apple's standard bash shell:
15260 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
15262 This example is equivalent to the QRCode example of @ref{coreimage} without the
15263 need for a nullsrc video source.
15267 @section mandelbrot
15269 Generate a Mandelbrot set fractal, and progressively zoom towards the
15270 point specified with @var{start_x} and @var{start_y}.
15272 This source accepts the following options:
15277 Set the terminal pts value. Default value is 400.
15280 Set the terminal scale value.
15281 Must be a floating point value. Default value is 0.3.
15284 Set the inner coloring mode, that is the algorithm used to draw the
15285 Mandelbrot fractal internal region.
15287 It shall assume one of the following values:
15292 Show time until convergence.
15294 Set color based on point closest to the origin of the iterations.
15299 Default value is @var{mincol}.
15302 Set the bailout value. Default value is 10.0.
15305 Set the maximum of iterations performed by the rendering
15306 algorithm. Default value is 7189.
15309 Set outer coloring mode.
15310 It shall assume one of following values:
15312 @item iteration_count
15313 Set iteration cound mode.
15314 @item normalized_iteration_count
15315 set normalized iteration count mode.
15317 Default value is @var{normalized_iteration_count}.
15320 Set frame rate, expressed as number of frames per second. Default
15324 Set frame size. For the syntax of this option, check the "Video
15325 size" section in the ffmpeg-utils manual. Default value is "640x480".
15328 Set the initial scale value. Default value is 3.0.
15331 Set the initial x position. Must be a floating point value between
15332 -100 and 100. Default value is -0.743643887037158704752191506114774.
15335 Set the initial y position. Must be a floating point value between
15336 -100 and 100. Default value is -0.131825904205311970493132056385139.
15341 Generate various test patterns, as generated by the MPlayer test filter.
15343 The size of the generated video is fixed, and is 256x256.
15344 This source is useful in particular for testing encoding features.
15346 This source accepts the following options:
15351 Specify the frame rate of the sourced video, as the number of frames
15352 generated per second. It has to be a string in the format
15353 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15354 number or a valid video frame rate abbreviation. The default value is
15358 Set the duration of the sourced video. See
15359 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15360 for the accepted syntax.
15362 If not specified, or the expressed duration is negative, the video is
15363 supposed to be generated forever.
15367 Set the number or the name of the test to perform. Supported tests are:
15383 Default value is "all", which will cycle through the list of all tests.
15388 mptestsrc=t=dc_luma
15391 will generate a "dc_luma" test pattern.
15393 @section frei0r_src
15395 Provide a frei0r source.
15397 To enable compilation of this filter you need to install the frei0r
15398 header and configure FFmpeg with @code{--enable-frei0r}.
15400 This source accepts the following parameters:
15405 The size of the video to generate. For the syntax of this option, check the
15406 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15409 The framerate of the generated video. It may be a string of the form
15410 @var{num}/@var{den} or a frame rate abbreviation.
15413 The name to the frei0r source to load. For more information regarding frei0r and
15414 how to set the parameters, read the @ref{frei0r} section in the video filters
15417 @item filter_params
15418 A '|'-separated list of parameters to pass to the frei0r source.
15422 For example, to generate a frei0r partik0l source with size 200x200
15423 and frame rate 10 which is overlaid on the overlay filter main input:
15425 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
15430 Generate a life pattern.
15432 This source is based on a generalization of John Conway's life game.
15434 The sourced input represents a life grid, each pixel represents a cell
15435 which can be in one of two possible states, alive or dead. Every cell
15436 interacts with its eight neighbours, which are the cells that are
15437 horizontally, vertically, or diagonally adjacent.
15439 At each interaction the grid evolves according to the adopted rule,
15440 which specifies the number of neighbor alive cells which will make a
15441 cell stay alive or born. The @option{rule} option allows one to specify
15444 This source accepts the following options:
15448 Set the file from which to read the initial grid state. In the file,
15449 each non-whitespace character is considered an alive cell, and newline
15450 is used to delimit the end of each row.
15452 If this option is not specified, the initial grid is generated
15456 Set the video rate, that is the number of frames generated per second.
15459 @item random_fill_ratio, ratio
15460 Set the random fill ratio for the initial random grid. It is a
15461 floating point number value ranging from 0 to 1, defaults to 1/PHI.
15462 It is ignored when a file is specified.
15464 @item random_seed, seed
15465 Set the seed for filling the initial random grid, must be an integer
15466 included between 0 and UINT32_MAX. If not specified, or if explicitly
15467 set to -1, the filter will try to use a good random seed on a best
15473 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
15474 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
15475 @var{NS} specifies the number of alive neighbor cells which make a
15476 live cell stay alive, and @var{NB} the number of alive neighbor cells
15477 which make a dead cell to become alive (i.e. to "born").
15478 "s" and "b" can be used in place of "S" and "B", respectively.
15480 Alternatively a rule can be specified by an 18-bits integer. The 9
15481 high order bits are used to encode the next cell state if it is alive
15482 for each number of neighbor alive cells, the low order bits specify
15483 the rule for "borning" new cells. Higher order bits encode for an
15484 higher number of neighbor cells.
15485 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
15486 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
15488 Default value is "S23/B3", which is the original Conway's game of life
15489 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
15490 cells, and will born a new cell if there are three alive cells around
15494 Set the size of the output video. For the syntax of this option, check the
15495 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15497 If @option{filename} is specified, the size is set by default to the
15498 same size of the input file. If @option{size} is set, it must contain
15499 the size specified in the input file, and the initial grid defined in
15500 that file is centered in the larger resulting area.
15502 If a filename is not specified, the size value defaults to "320x240"
15503 (used for a randomly generated initial grid).
15506 If set to 1, stitch the left and right grid edges together, and the
15507 top and bottom edges also. Defaults to 1.
15510 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
15511 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
15512 value from 0 to 255.
15515 Set the color of living (or new born) cells.
15518 Set the color of dead cells. If @option{mold} is set, this is the first color
15519 used to represent a dead cell.
15522 Set mold color, for definitely dead and moldy cells.
15524 For the syntax of these 3 color options, check the "Color" section in the
15525 ffmpeg-utils manual.
15528 @subsection Examples
15532 Read a grid from @file{pattern}, and center it on a grid of size
15535 life=f=pattern:s=300x300
15539 Generate a random grid of size 200x200, with a fill ratio of 2/3:
15541 life=ratio=2/3:s=200x200
15545 Specify a custom rule for evolving a randomly generated grid:
15551 Full example with slow death effect (mold) using @command{ffplay}:
15553 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
15560 @anchor{haldclutsrc}
15562 @anchor{rgbtestsrc}
15564 @anchor{smptehdbars}
15567 @anchor{yuvtestsrc}
15568 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2, yuvtestsrc
15570 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
15572 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
15574 The @code{color} source provides an uniformly colored input.
15576 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
15577 @ref{haldclut} filter.
15579 The @code{nullsrc} source returns unprocessed video frames. It is
15580 mainly useful to be employed in analysis / debugging tools, or as the
15581 source for filters which ignore the input data.
15583 The @code{rgbtestsrc} source generates an RGB test pattern useful for
15584 detecting RGB vs BGR issues. You should see a red, green and blue
15585 stripe from top to bottom.
15587 The @code{smptebars} source generates a color bars pattern, based on
15588 the SMPTE Engineering Guideline EG 1-1990.
15590 The @code{smptehdbars} source generates a color bars pattern, based on
15591 the SMPTE RP 219-2002.
15593 The @code{testsrc} source generates a test video pattern, showing a
15594 color pattern, a scrolling gradient and a timestamp. This is mainly
15595 intended for testing purposes.
15597 The @code{testsrc2} source is similar to testsrc, but supports more
15598 pixel formats instead of just @code{rgb24}. This allows using it as an
15599 input for other tests without requiring a format conversion.
15601 The @code{yuvtestsrc} source generates an YUV test pattern. You should
15602 see a y, cb and cr stripe from top to bottom.
15604 The sources accept the following parameters:
15609 Specify the color of the source, only available in the @code{color}
15610 source. For the syntax of this option, check the "Color" section in the
15611 ffmpeg-utils manual.
15614 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
15615 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
15616 pixels to be used as identity matrix for 3D lookup tables. Each component is
15617 coded on a @code{1/(N*N)} scale.
15620 Specify the size of the sourced video. For the syntax of this option, check the
15621 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15622 The default value is @code{320x240}.
15624 This option is not available with the @code{haldclutsrc} filter.
15627 Specify the frame rate of the sourced video, as the number of frames
15628 generated per second. It has to be a string in the format
15629 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15630 number or a valid video frame rate abbreviation. The default value is
15634 Set the sample aspect ratio of the sourced video.
15637 Set the duration of the sourced video. See
15638 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15639 for the accepted syntax.
15641 If not specified, or the expressed duration is negative, the video is
15642 supposed to be generated forever.
15645 Set the number of decimals to show in the timestamp, only available in the
15646 @code{testsrc} source.
15648 The displayed timestamp value will correspond to the original
15649 timestamp value multiplied by the power of 10 of the specified
15650 value. Default value is 0.
15653 For example the following:
15655 testsrc=duration=5.3:size=qcif:rate=10
15658 will generate a video with a duration of 5.3 seconds, with size
15659 176x144 and a frame rate of 10 frames per second.
15661 The following graph description will generate a red source
15662 with an opacity of 0.2, with size "qcif" and a frame rate of 10
15665 color=c=red@@0.2:s=qcif:r=10
15668 If the input content is to be ignored, @code{nullsrc} can be used. The
15669 following command generates noise in the luminance plane by employing
15670 the @code{geq} filter:
15672 nullsrc=s=256x256, geq=random(1)*255:128:128
15675 @subsection Commands
15677 The @code{color} source supports the following commands:
15681 Set the color of the created image. Accepts the same syntax of the
15682 corresponding @option{color} option.
15685 @c man end VIDEO SOURCES
15687 @chapter Video Sinks
15688 @c man begin VIDEO SINKS
15690 Below is a description of the currently available video sinks.
15692 @section buffersink
15694 Buffer video frames, and make them available to the end of the filter
15697 This sink is mainly intended for programmatic use, in particular
15698 through the interface defined in @file{libavfilter/buffersink.h}
15699 or the options system.
15701 It accepts a pointer to an AVBufferSinkContext structure, which
15702 defines the incoming buffers' formats, to be passed as the opaque
15703 parameter to @code{avfilter_init_filter} for initialization.
15707 Null video sink: do absolutely nothing with the input video. It is
15708 mainly useful as a template and for use in analysis / debugging
15711 @c man end VIDEO SINKS
15713 @chapter Multimedia Filters
15714 @c man begin MULTIMEDIA FILTERS
15716 Below is a description of the currently available multimedia filters.
15720 Convert input audio to a video output, displaying the audio bit scope.
15722 The filter accepts the following options:
15726 Set frame rate, expressed as number of frames per second. Default
15730 Specify the video size for the output. For the syntax of this option, check the
15731 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15732 Default value is @code{1024x256}.
15735 Specify list of colors separated by space or by '|' which will be used to
15736 draw channels. Unrecognized or missing colors will be replaced
15740 @section ahistogram
15742 Convert input audio to a video output, displaying the volume histogram.
15744 The filter accepts the following options:
15748 Specify how histogram is calculated.
15750 It accepts the following values:
15753 Use single histogram for all channels.
15755 Use separate histogram for each channel.
15757 Default is @code{single}.
15760 Set frame rate, expressed as number of frames per second. Default
15764 Specify the video size for the output. For the syntax of this option, check the
15765 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15766 Default value is @code{hd720}.
15771 It accepts the following values:
15782 reverse logarithmic
15784 Default is @code{log}.
15787 Set amplitude scale.
15789 It accepts the following values:
15796 Default is @code{log}.
15799 Set how much frames to accumulate in histogram.
15800 Defauls is 1. Setting this to -1 accumulates all frames.
15803 Set histogram ratio of window height.
15806 Set sonogram sliding.
15808 It accepts the following values:
15811 replace old rows with new ones.
15813 scroll from top to bottom.
15815 Default is @code{replace}.
15818 @section aphasemeter
15820 Convert input audio to a video output, displaying the audio phase.
15822 The filter accepts the following options:
15826 Set the output frame rate. Default value is @code{25}.
15829 Set the video size for the output. For the syntax of this option, check the
15830 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15831 Default value is @code{800x400}.
15836 Specify the red, green, blue contrast. Default values are @code{2},
15837 @code{7} and @code{1}.
15838 Allowed range is @code{[0, 255]}.
15841 Set color which will be used for drawing median phase. If color is
15842 @code{none} which is default, no median phase value will be drawn.
15845 Enable video output. Default is enabled.
15848 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
15849 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
15850 The @code{-1} means left and right channels are completely out of phase and
15851 @code{1} means channels are in phase.
15853 @section avectorscope
15855 Convert input audio to a video output, representing the audio vector
15858 The filter is used to measure the difference between channels of stereo
15859 audio stream. A monoaural signal, consisting of identical left and right
15860 signal, results in straight vertical line. Any stereo separation is visible
15861 as a deviation from this line, creating a Lissajous figure.
15862 If the straight (or deviation from it) but horizontal line appears this
15863 indicates that the left and right channels are out of phase.
15865 The filter accepts the following options:
15869 Set the vectorscope mode.
15871 Available values are:
15874 Lissajous rotated by 45 degrees.
15877 Same as above but not rotated.
15880 Shape resembling half of circle.
15883 Default value is @samp{lissajous}.
15886 Set the video size for the output. For the syntax of this option, check the
15887 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15888 Default value is @code{400x400}.
15891 Set the output frame rate. Default value is @code{25}.
15897 Specify the red, green, blue and alpha contrast. Default values are @code{40},
15898 @code{160}, @code{80} and @code{255}.
15899 Allowed range is @code{[0, 255]}.
15905 Specify the red, green, blue and alpha fade. Default values are @code{15},
15906 @code{10}, @code{5} and @code{5}.
15907 Allowed range is @code{[0, 255]}.
15910 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
15913 Set the vectorscope drawing mode.
15915 Available values are:
15918 Draw dot for each sample.
15921 Draw line between previous and current sample.
15924 Default value is @samp{dot}.
15927 Specify amplitude scale of audio samples.
15929 Available values are:
15946 @subsection Examples
15950 Complete example using @command{ffplay}:
15952 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
15953 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
15957 @section bench, abench
15959 Benchmark part of a filtergraph.
15961 The filter accepts the following options:
15965 Start or stop a timer.
15967 Available values are:
15970 Get the current time, set it as frame metadata (using the key
15971 @code{lavfi.bench.start_time}), and forward the frame to the next filter.
15974 Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
15975 the input frame metadata to get the time difference. Time difference, average,
15976 maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
15977 @code{min}) are then printed. The timestamps are expressed in seconds.
15981 @subsection Examples
15985 Benchmark @ref{selectivecolor} filter:
15987 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
15993 Concatenate audio and video streams, joining them together one after the
15996 The filter works on segments of synchronized video and audio streams. All
15997 segments must have the same number of streams of each type, and that will
15998 also be the number of streams at output.
16000 The filter accepts the following options:
16005 Set the number of segments. Default is 2.
16008 Set the number of output video streams, that is also the number of video
16009 streams in each segment. Default is 1.
16012 Set the number of output audio streams, that is also the number of audio
16013 streams in each segment. Default is 0.
16016 Activate unsafe mode: do not fail if segments have a different format.
16020 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
16021 @var{a} audio outputs.
16023 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
16024 segment, in the same order as the outputs, then the inputs for the second
16027 Related streams do not always have exactly the same duration, for various
16028 reasons including codec frame size or sloppy authoring. For that reason,
16029 related synchronized streams (e.g. a video and its audio track) should be
16030 concatenated at once. The concat filter will use the duration of the longest
16031 stream in each segment (except the last one), and if necessary pad shorter
16032 audio streams with silence.
16034 For this filter to work correctly, all segments must start at timestamp 0.
16036 All corresponding streams must have the same parameters in all segments; the
16037 filtering system will automatically select a common pixel format for video
16038 streams, and a common sample format, sample rate and channel layout for
16039 audio streams, but other settings, such as resolution, must be converted
16040 explicitly by the user.
16042 Different frame rates are acceptable but will result in variable frame rate
16043 at output; be sure to configure the output file to handle it.
16045 @subsection Examples
16049 Concatenate an opening, an episode and an ending, all in bilingual version
16050 (video in stream 0, audio in streams 1 and 2):
16052 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
16053 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
16054 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
16055 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
16059 Concatenate two parts, handling audio and video separately, using the
16060 (a)movie sources, and adjusting the resolution:
16062 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
16063 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
16064 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
16066 Note that a desync will happen at the stitch if the audio and video streams
16067 do not have exactly the same duration in the first file.
16071 @section drawgraph, adrawgraph
16073 Draw a graph using input video or audio metadata.
16075 It accepts the following parameters:
16079 Set 1st frame metadata key from which metadata values will be used to draw a graph.
16082 Set 1st foreground color expression.
16085 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
16088 Set 2nd foreground color expression.
16091 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
16094 Set 3rd foreground color expression.
16097 Set 4th frame metadata key from which metadata values will be used to draw a graph.
16100 Set 4th foreground color expression.
16103 Set minimal value of metadata value.
16106 Set maximal value of metadata value.
16109 Set graph background color. Default is white.
16114 Available values for mode is:
16121 Default is @code{line}.
16126 Available values for slide is:
16129 Draw new frame when right border is reached.
16132 Replace old columns with new ones.
16135 Scroll from right to left.
16138 Scroll from left to right.
16141 Draw single picture.
16144 Default is @code{frame}.
16147 Set size of graph video. For the syntax of this option, check the
16148 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16149 The default value is @code{900x256}.
16151 The foreground color expressions can use the following variables:
16154 Minimal value of metadata value.
16157 Maximal value of metadata value.
16160 Current metadata key value.
16163 The color is defined as 0xAABBGGRR.
16166 Example using metadata from @ref{signalstats} filter:
16168 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
16171 Example using metadata from @ref{ebur128} filter:
16173 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
16179 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
16180 it unchanged. By default, it logs a message at a frequency of 10Hz with the
16181 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
16182 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
16184 The filter also has a video output (see the @var{video} option) with a real
16185 time graph to observe the loudness evolution. The graphic contains the logged
16186 message mentioned above, so it is not printed anymore when this option is set,
16187 unless the verbose logging is set. The main graphing area contains the
16188 short-term loudness (3 seconds of analysis), and the gauge on the right is for
16189 the momentary loudness (400 milliseconds).
16191 More information about the Loudness Recommendation EBU R128 on
16192 @url{http://tech.ebu.ch/loudness}.
16194 The filter accepts the following options:
16199 Activate the video output. The audio stream is passed unchanged whether this
16200 option is set or no. The video stream will be the first output stream if
16201 activated. Default is @code{0}.
16204 Set the video size. This option is for video only. For the syntax of this
16206 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16207 Default and minimum resolution is @code{640x480}.
16210 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
16211 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
16212 other integer value between this range is allowed.
16215 Set metadata injection. If set to @code{1}, the audio input will be segmented
16216 into 100ms output frames, each of them containing various loudness information
16217 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
16219 Default is @code{0}.
16222 Force the frame logging level.
16224 Available values are:
16227 information logging level
16229 verbose logging level
16232 By default, the logging level is set to @var{info}. If the @option{video} or
16233 the @option{metadata} options are set, it switches to @var{verbose}.
16238 Available modes can be cumulated (the option is a @code{flag} type). Possible
16242 Disable any peak mode (default).
16244 Enable sample-peak mode.
16246 Simple peak mode looking for the higher sample value. It logs a message
16247 for sample-peak (identified by @code{SPK}).
16249 Enable true-peak mode.
16251 If enabled, the peak lookup is done on an over-sampled version of the input
16252 stream for better peak accuracy. It logs a message for true-peak.
16253 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
16254 This mode requires a build with @code{libswresample}.
16258 Treat mono input files as "dual mono". If a mono file is intended for playback
16259 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
16260 If set to @code{true}, this option will compensate for this effect.
16261 Multi-channel input files are not affected by this option.
16264 Set a specific pan law to be used for the measurement of dual mono files.
16265 This parameter is optional, and has a default value of -3.01dB.
16268 @subsection Examples
16272 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
16274 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
16278 Run an analysis with @command{ffmpeg}:
16280 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
16284 @section interleave, ainterleave
16286 Temporally interleave frames from several inputs.
16288 @code{interleave} works with video inputs, @code{ainterleave} with audio.
16290 These filters read frames from several inputs and send the oldest
16291 queued frame to the output.
16293 Input streams must have well defined, monotonically increasing frame
16296 In order to submit one frame to output, these filters need to enqueue
16297 at least one frame for each input, so they cannot work in case one
16298 input is not yet terminated and will not receive incoming frames.
16300 For example consider the case when one input is a @code{select} filter
16301 which always drops input frames. The @code{interleave} filter will keep
16302 reading from that input, but it will never be able to send new frames
16303 to output until the input sends an end-of-stream signal.
16305 Also, depending on inputs synchronization, the filters will drop
16306 frames in case one input receives more frames than the other ones, and
16307 the queue is already filled.
16309 These filters accept the following options:
16313 Set the number of different inputs, it is 2 by default.
16316 @subsection Examples
16320 Interleave frames belonging to different streams using @command{ffmpeg}:
16322 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
16326 Add flickering blur effect:
16328 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
16332 @section metadata, ametadata
16334 Manipulate frame metadata.
16336 This filter accepts the following options:
16340 Set mode of operation of the filter.
16342 Can be one of the following:
16346 If both @code{value} and @code{key} is set, select frames
16347 which have such metadata. If only @code{key} is set, select
16348 every frame that has such key in metadata.
16351 Add new metadata @code{key} and @code{value}. If key is already available
16355 Modify value of already present key.
16358 If @code{value} is set, delete only keys that have such value.
16359 Otherwise, delete key. If @code{key} is not set, delete all metadata values in
16363 Print key and its value if metadata was found. If @code{key} is not set print all
16364 metadata values available in frame.
16368 Set key used with all modes. Must be set for all modes except @code{print} and @code{delete}.
16371 Set metadata value which will be used. This option is mandatory for
16372 @code{modify} and @code{add} mode.
16375 Which function to use when comparing metadata value and @code{value}.
16377 Can be one of following:
16381 Values are interpreted as strings, returns true if metadata value is same as @code{value}.
16384 Values are interpreted as strings, returns true if metadata value starts with
16385 the @code{value} option string.
16388 Values are interpreted as floats, returns true if metadata value is less than @code{value}.
16391 Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
16394 Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
16397 Values are interpreted as floats, returns true if expression from option @code{expr}
16402 Set expression which is used when @code{function} is set to @code{expr}.
16403 The expression is evaluated through the eval API and can contain the following
16408 Float representation of @code{value} from metadata key.
16411 Float representation of @code{value} as supplied by user in @code{value} option.
16415 If specified in @code{print} mode, output is written to the named file. Instead of
16416 plain filename any writable url can be specified. Filename ``-'' is a shorthand
16417 for standard output. If @code{file} option is not set, output is written to the log
16418 with AV_LOG_INFO loglevel.
16422 @subsection Examples
16426 Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
16429 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
16432 Print silencedetect output to file @file{metadata.txt}.
16434 silencedetect,ametadata=mode=print:file=metadata.txt
16437 Direct all metadata to a pipe with file descriptor 4.
16439 metadata=mode=print:file='pipe\:4'
16443 @section perms, aperms
16445 Set read/write permissions for the output frames.
16447 These filters are mainly aimed at developers to test direct path in the
16448 following filter in the filtergraph.
16450 The filters accept the following options:
16454 Select the permissions mode.
16456 It accepts the following values:
16459 Do nothing. This is the default.
16461 Set all the output frames read-only.
16463 Set all the output frames directly writable.
16465 Make the frame read-only if writable, and writable if read-only.
16467 Set each output frame read-only or writable randomly.
16471 Set the seed for the @var{random} mode, must be an integer included between
16472 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
16473 @code{-1}, the filter will try to use a good random seed on a best effort
16477 Note: in case of auto-inserted filter between the permission filter and the
16478 following one, the permission might not be received as expected in that
16479 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
16480 perms/aperms filter can avoid this problem.
16482 @section realtime, arealtime
16484 Slow down filtering to match real time approximatively.
16486 These filters will pause the filtering for a variable amount of time to
16487 match the output rate with the input timestamps.
16488 They are similar to the @option{re} option to @code{ffmpeg}.
16490 They accept the following options:
16494 Time limit for the pauses. Any pause longer than that will be considered
16495 a timestamp discontinuity and reset the timer. Default is 2 seconds.
16499 @section select, aselect
16501 Select frames to pass in output.
16503 This filter accepts the following options:
16508 Set expression, which is evaluated for each input frame.
16510 If the expression is evaluated to zero, the frame is discarded.
16512 If the evaluation result is negative or NaN, the frame is sent to the
16513 first output; otherwise it is sent to the output with index
16514 @code{ceil(val)-1}, assuming that the input index starts from 0.
16516 For example a value of @code{1.2} corresponds to the output with index
16517 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
16520 Set the number of outputs. The output to which to send the selected
16521 frame is based on the result of the evaluation. Default value is 1.
16524 The expression can contain the following constants:
16528 The (sequential) number of the filtered frame, starting from 0.
16531 The (sequential) number of the selected frame, starting from 0.
16533 @item prev_selected_n
16534 The sequential number of the last selected frame. It's NAN if undefined.
16537 The timebase of the input timestamps.
16540 The PTS (Presentation TimeStamp) of the filtered video frame,
16541 expressed in @var{TB} units. It's NAN if undefined.
16544 The PTS of the filtered video frame,
16545 expressed in seconds. It's NAN if undefined.
16548 The PTS of the previously filtered video frame. It's NAN if undefined.
16550 @item prev_selected_pts
16551 The PTS of the last previously filtered video frame. It's NAN if undefined.
16553 @item prev_selected_t
16554 The PTS of the last previously selected video frame. It's NAN if undefined.
16557 The PTS of the first video frame in the video. It's NAN if undefined.
16560 The time of the first video frame in the video. It's NAN if undefined.
16562 @item pict_type @emph{(video only)}
16563 The type of the filtered frame. It can assume one of the following
16575 @item interlace_type @emph{(video only)}
16576 The frame interlace type. It can assume one of the following values:
16579 The frame is progressive (not interlaced).
16581 The frame is top-field-first.
16583 The frame is bottom-field-first.
16586 @item consumed_sample_n @emph{(audio only)}
16587 the number of selected samples before the current frame
16589 @item samples_n @emph{(audio only)}
16590 the number of samples in the current frame
16592 @item sample_rate @emph{(audio only)}
16593 the input sample rate
16596 This is 1 if the filtered frame is a key-frame, 0 otherwise.
16599 the position in the file of the filtered frame, -1 if the information
16600 is not available (e.g. for synthetic video)
16602 @item scene @emph{(video only)}
16603 value between 0 and 1 to indicate a new scene; a low value reflects a low
16604 probability for the current frame to introduce a new scene, while a higher
16605 value means the current frame is more likely to be one (see the example below)
16607 @item concatdec_select
16608 The concat demuxer can select only part of a concat input file by setting an
16609 inpoint and an outpoint, but the output packets may not be entirely contained
16610 in the selected interval. By using this variable, it is possible to skip frames
16611 generated by the concat demuxer which are not exactly contained in the selected
16614 This works by comparing the frame pts against the @var{lavf.concat.start_time}
16615 and the @var{lavf.concat.duration} packet metadata values which are also
16616 present in the decoded frames.
16618 The @var{concatdec_select} variable is -1 if the frame pts is at least
16619 start_time and either the duration metadata is missing or the frame pts is less
16620 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
16623 That basically means that an input frame is selected if its pts is within the
16624 interval set by the concat demuxer.
16628 The default value of the select expression is "1".
16630 @subsection Examples
16634 Select all frames in input:
16639 The example above is the same as:
16651 Select only I-frames:
16653 select='eq(pict_type\,I)'
16657 Select one frame every 100:
16659 select='not(mod(n\,100))'
16663 Select only frames contained in the 10-20 time interval:
16665 select=between(t\,10\,20)
16669 Select only I-frames contained in the 10-20 time interval:
16671 select=between(t\,10\,20)*eq(pict_type\,I)
16675 Select frames with a minimum distance of 10 seconds:
16677 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
16681 Use aselect to select only audio frames with samples number > 100:
16683 aselect='gt(samples_n\,100)'
16687 Create a mosaic of the first scenes:
16689 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
16692 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
16696 Send even and odd frames to separate outputs, and compose them:
16698 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
16702 Select useful frames from an ffconcat file which is using inpoints and
16703 outpoints but where the source files are not intra frame only.
16705 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
16709 @section sendcmd, asendcmd
16711 Send commands to filters in the filtergraph.
16713 These filters read commands to be sent to other filters in the
16716 @code{sendcmd} must be inserted between two video filters,
16717 @code{asendcmd} must be inserted between two audio filters, but apart
16718 from that they act the same way.
16720 The specification of commands can be provided in the filter arguments
16721 with the @var{commands} option, or in a file specified by the
16722 @var{filename} option.
16724 These filters accept the following options:
16727 Set the commands to be read and sent to the other filters.
16729 Set the filename of the commands to be read and sent to the other
16733 @subsection Commands syntax
16735 A commands description consists of a sequence of interval
16736 specifications, comprising a list of commands to be executed when a
16737 particular event related to that interval occurs. The occurring event
16738 is typically the current frame time entering or leaving a given time
16741 An interval is specified by the following syntax:
16743 @var{START}[-@var{END}] @var{COMMANDS};
16746 The time interval is specified by the @var{START} and @var{END} times.
16747 @var{END} is optional and defaults to the maximum time.
16749 The current frame time is considered within the specified interval if
16750 it is included in the interval [@var{START}, @var{END}), that is when
16751 the time is greater or equal to @var{START} and is lesser than
16754 @var{COMMANDS} consists of a sequence of one or more command
16755 specifications, separated by ",", relating to that interval. The
16756 syntax of a command specification is given by:
16758 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
16761 @var{FLAGS} is optional and specifies the type of events relating to
16762 the time interval which enable sending the specified command, and must
16763 be a non-null sequence of identifier flags separated by "+" or "|" and
16764 enclosed between "[" and "]".
16766 The following flags are recognized:
16769 The command is sent when the current frame timestamp enters the
16770 specified interval. In other words, the command is sent when the
16771 previous frame timestamp was not in the given interval, and the
16775 The command is sent when the current frame timestamp leaves the
16776 specified interval. In other words, the command is sent when the
16777 previous frame timestamp was in the given interval, and the
16781 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
16784 @var{TARGET} specifies the target of the command, usually the name of
16785 the filter class or a specific filter instance name.
16787 @var{COMMAND} specifies the name of the command for the target filter.
16789 @var{ARG} is optional and specifies the optional list of argument for
16790 the given @var{COMMAND}.
16792 Between one interval specification and another, whitespaces, or
16793 sequences of characters starting with @code{#} until the end of line,
16794 are ignored and can be used to annotate comments.
16796 A simplified BNF description of the commands specification syntax
16799 @var{COMMAND_FLAG} ::= "enter" | "leave"
16800 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
16801 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
16802 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
16803 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
16804 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
16807 @subsection Examples
16811 Specify audio tempo change at second 4:
16813 asendcmd=c='4.0 atempo tempo 1.5',atempo
16817 Specify a list of drawtext and hue commands in a file.
16819 # show text in the interval 5-10
16820 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
16821 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
16823 # desaturate the image in the interval 15-20
16824 15.0-20.0 [enter] hue s 0,
16825 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
16827 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
16829 # apply an exponential saturation fade-out effect, starting from time 25
16830 25 [enter] hue s exp(25-t)
16833 A filtergraph allowing to read and process the above command list
16834 stored in a file @file{test.cmd}, can be specified with:
16836 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
16841 @section setpts, asetpts
16843 Change the PTS (presentation timestamp) of the input frames.
16845 @code{setpts} works on video frames, @code{asetpts} on audio frames.
16847 This filter accepts the following options:
16852 The expression which is evaluated for each frame to construct its timestamp.
16856 The expression is evaluated through the eval API and can contain the following
16861 frame rate, only defined for constant frame-rate video
16864 The presentation timestamp in input
16867 The count of the input frame for video or the number of consumed samples,
16868 not including the current frame for audio, starting from 0.
16870 @item NB_CONSUMED_SAMPLES
16871 The number of consumed samples, not including the current frame (only
16874 @item NB_SAMPLES, S
16875 The number of samples in the current frame (only audio)
16877 @item SAMPLE_RATE, SR
16878 The audio sample rate.
16881 The PTS of the first frame.
16884 the time in seconds of the first frame
16887 State whether the current frame is interlaced.
16890 the time in seconds of the current frame
16893 original position in the file of the frame, or undefined if undefined
16894 for the current frame
16897 The previous input PTS.
16900 previous input time in seconds
16903 The previous output PTS.
16906 previous output time in seconds
16909 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
16913 The wallclock (RTC) time at the start of the movie in microseconds.
16916 The timebase of the input timestamps.
16920 @subsection Examples
16924 Start counting PTS from zero
16926 setpts=PTS-STARTPTS
16930 Apply fast motion effect:
16936 Apply slow motion effect:
16942 Set fixed rate of 25 frames per second:
16948 Set fixed rate 25 fps with some jitter:
16950 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
16954 Apply an offset of 10 seconds to the input PTS:
16960 Generate timestamps from a "live source" and rebase onto the current timebase:
16962 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
16966 Generate timestamps by counting samples:
16973 @section settb, asettb
16975 Set the timebase to use for the output frames timestamps.
16976 It is mainly useful for testing timebase configuration.
16978 It accepts the following parameters:
16983 The expression which is evaluated into the output timebase.
16987 The value for @option{tb} is an arithmetic expression representing a
16988 rational. The expression can contain the constants "AVTB" (the default
16989 timebase), "intb" (the input timebase) and "sr" (the sample rate,
16990 audio only). Default value is "intb".
16992 @subsection Examples
16996 Set the timebase to 1/25:
17002 Set the timebase to 1/10:
17008 Set the timebase to 1001/1000:
17014 Set the timebase to 2*intb:
17020 Set the default timebase value:
17027 Convert input audio to a video output representing frequency spectrum
17028 logarithmically using Brown-Puckette constant Q transform algorithm with
17029 direct frequency domain coefficient calculation (but the transform itself
17030 is not really constant Q, instead the Q factor is actually variable/clamped),
17031 with musical tone scale, from E0 to D#10.
17033 The filter accepts the following options:
17037 Specify the video size for the output. It must be even. For the syntax of this option,
17038 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17039 Default value is @code{1920x1080}.
17042 Set the output frame rate. Default value is @code{25}.
17045 Set the bargraph height. It must be even. Default value is @code{-1} which
17046 computes the bargraph height automatically.
17049 Set the axis height. It must be even. Default value is @code{-1} which computes
17050 the axis height automatically.
17053 Set the sonogram height. It must be even. Default value is @code{-1} which
17054 computes the sonogram height automatically.
17057 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
17058 instead. Default value is @code{1}.
17060 @item sono_v, volume
17061 Specify the sonogram volume expression. It can contain variables:
17064 the @var{bar_v} evaluated expression
17065 @item frequency, freq, f
17066 the frequency where it is evaluated
17067 @item timeclamp, tc
17068 the value of @var{timeclamp} option
17072 @item a_weighting(f)
17073 A-weighting of equal loudness
17074 @item b_weighting(f)
17075 B-weighting of equal loudness
17076 @item c_weighting(f)
17077 C-weighting of equal loudness.
17079 Default value is @code{16}.
17081 @item bar_v, volume2
17082 Specify the bargraph volume expression. It can contain variables:
17085 the @var{sono_v} evaluated expression
17086 @item frequency, freq, f
17087 the frequency where it is evaluated
17088 @item timeclamp, tc
17089 the value of @var{timeclamp} option
17093 @item a_weighting(f)
17094 A-weighting of equal loudness
17095 @item b_weighting(f)
17096 B-weighting of equal loudness
17097 @item c_weighting(f)
17098 C-weighting of equal loudness.
17100 Default value is @code{sono_v}.
17102 @item sono_g, gamma
17103 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
17104 higher gamma makes the spectrum having more range. Default value is @code{3}.
17105 Acceptable range is @code{[1, 7]}.
17107 @item bar_g, gamma2
17108 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
17112 Specify the bargraph transparency level. Lower value makes the bargraph sharper.
17113 Default value is @code{1}. Acceptable range is @code{[0, 1]}.
17115 @item timeclamp, tc
17116 Specify the transform timeclamp. At low frequency, there is trade-off between
17117 accuracy in time domain and frequency domain. If timeclamp is lower,
17118 event in time domain is represented more accurately (such as fast bass drum),
17119 otherwise event in frequency domain is represented more accurately
17120 (such as bass guitar). Acceptable range is @code{[0.002, 1]}. Default value is @code{0.17}.
17123 Specify the transform base frequency. Default value is @code{20.01523126408007475},
17124 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
17127 Specify the transform end frequency. Default value is @code{20495.59681441799654},
17128 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
17131 This option is deprecated and ignored.
17134 Specify the transform length in time domain. Use this option to control accuracy
17135 trade-off between time domain and frequency domain at every frequency sample.
17136 It can contain variables:
17138 @item frequency, freq, f
17139 the frequency where it is evaluated
17140 @item timeclamp, tc
17141 the value of @var{timeclamp} option.
17143 Default value is @code{384*tc/(384+tc*f)}.
17146 Specify the transform count for every video frame. Default value is @code{6}.
17147 Acceptable range is @code{[1, 30]}.
17150 Specify the transform count for every single pixel. Default value is @code{0},
17151 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
17154 Specify font file for use with freetype to draw the axis. If not specified,
17155 use embedded font. Note that drawing with font file or embedded font is not
17156 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
17160 Specify fontconfig pattern. This has lower priority than @var{fontfile}.
17161 The : in the pattern may be replaced by | to avoid unnecessary escaping.
17164 Specify font color expression. This is arithmetic expression that should return
17165 integer value 0xRRGGBB. It can contain variables:
17167 @item frequency, freq, f
17168 the frequency where it is evaluated
17169 @item timeclamp, tc
17170 the value of @var{timeclamp} option
17175 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
17176 @item r(x), g(x), b(x)
17177 red, green, and blue value of intensity x.
17179 Default value is @code{st(0, (midi(f)-59.5)/12);
17180 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
17181 r(1-ld(1)) + b(ld(1))}.
17184 Specify image file to draw the axis. This option override @var{fontfile} and
17185 @var{fontcolor} option.
17188 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
17189 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
17190 Default value is @code{1}.
17193 Set colorspace. The accepted values are:
17196 Unspecified (default)
17205 BT.470BG or BT.601-6 625
17208 SMPTE-170M or BT.601-6 525
17214 BT.2020 with non-constant luminance
17219 Set spectrogram color scheme. This is list of floating point values with format
17220 @code{left_r|left_g|left_b|right_r|right_g|right_b}.
17221 The default is @code{1|0.5|0|0|0.5|1}.
17225 @subsection Examples
17229 Playing audio while showing the spectrum:
17231 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
17235 Same as above, but with frame rate 30 fps:
17237 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
17241 Playing at 1280x720:
17243 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
17247 Disable sonogram display:
17253 A1 and its harmonics: A1, A2, (near)E3, A3:
17255 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),
17256 asplit[a][out1]; [a] showcqt [out0]'
17260 Same as above, but with more accuracy in frequency domain:
17262 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),
17263 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
17269 bar_v=10:sono_v=bar_v*a_weighting(f)
17273 Custom gamma, now spectrum is linear to the amplitude.
17279 Custom tlength equation:
17281 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)))'
17285 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
17287 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
17291 Custom font using fontconfig:
17293 font='Courier New,Monospace,mono|bold'
17297 Custom frequency range with custom axis using image file:
17299 axisfile=myaxis.png:basefreq=40:endfreq=10000
17305 Convert input audio to video output representing the audio power spectrum.
17306 Audio amplitude is on Y-axis while frequency is on X-axis.
17308 The filter accepts the following options:
17312 Specify size of video. For the syntax of this option, check the
17313 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17314 Default is @code{1024x512}.
17318 This set how each frequency bin will be represented.
17320 It accepts the following values:
17326 Default is @code{bar}.
17329 Set amplitude scale.
17331 It accepts the following values:
17345 Default is @code{log}.
17348 Set frequency scale.
17350 It accepts the following values:
17359 Reverse logarithmic scale.
17361 Default is @code{lin}.
17366 It accepts the following values:
17382 Default is @code{w2048}
17385 Set windowing function.
17387 It accepts the following values:
17409 Default is @code{hanning}.
17412 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
17413 which means optimal overlap for selected window function will be picked.
17416 Set time averaging. Setting this to 0 will display current maximal peaks.
17417 Default is @code{1}, which means time averaging is disabled.
17420 Specify list of colors separated by space or by '|' which will be used to
17421 draw channel frequencies. Unrecognized or missing colors will be replaced
17425 Set channel display mode.
17427 It accepts the following values:
17432 Default is @code{combined}.
17435 Set minimum amplitude used in @code{log} amplitude scaler.
17439 @anchor{showspectrum}
17440 @section showspectrum
17442 Convert input audio to a video output, representing the audio frequency
17445 The filter accepts the following options:
17449 Specify the video size for the output. For the syntax of this option, check the
17450 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17451 Default value is @code{640x512}.
17454 Specify how the spectrum should slide along the window.
17456 It accepts the following values:
17459 the samples start again on the left when they reach the right
17461 the samples scroll from right to left
17463 frames are only produced when the samples reach the right
17465 the samples scroll from left to right
17468 Default value is @code{replace}.
17471 Specify display mode.
17473 It accepts the following values:
17476 all channels are displayed in the same row
17478 all channels are displayed in separate rows
17481 Default value is @samp{combined}.
17484 Specify display color mode.
17486 It accepts the following values:
17489 each channel is displayed in a separate color
17491 each channel is displayed using the same color scheme
17493 each channel is displayed using the rainbow color scheme
17495 each channel is displayed using the moreland color scheme
17497 each channel is displayed using the nebulae color scheme
17499 each channel is displayed using the fire color scheme
17501 each channel is displayed using the fiery color scheme
17503 each channel is displayed using the fruit color scheme
17505 each channel is displayed using the cool color scheme
17508 Default value is @samp{channel}.
17511 Specify scale used for calculating intensity color values.
17513 It accepts the following values:
17518 square root, default
17529 Default value is @samp{sqrt}.
17532 Set saturation modifier for displayed colors. Negative values provide
17533 alternative color scheme. @code{0} is no saturation at all.
17534 Saturation must be in [-10.0, 10.0] range.
17535 Default value is @code{1}.
17538 Set window function.
17540 It accepts the following values:
17564 Default value is @code{hann}.
17567 Set orientation of time vs frequency axis. Can be @code{vertical} or
17568 @code{horizontal}. Default is @code{vertical}.
17571 Set ratio of overlap window. Default value is @code{0}.
17572 When value is @code{1} overlap is set to recommended size for specific
17573 window function currently used.
17576 Set scale gain for calculating intensity color values.
17577 Default value is @code{1}.
17580 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
17583 Set color rotation, must be in [-1.0, 1.0] range.
17584 Default value is @code{0}.
17587 The usage is very similar to the showwaves filter; see the examples in that
17590 @subsection Examples
17594 Large window with logarithmic color scaling:
17596 showspectrum=s=1280x480:scale=log
17600 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
17602 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
17603 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
17607 @section showspectrumpic
17609 Convert input audio to a single video frame, representing the audio frequency
17612 The filter accepts the following options:
17616 Specify the video size for the output. For the syntax of this option, check the
17617 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17618 Default value is @code{4096x2048}.
17621 Specify display mode.
17623 It accepts the following values:
17626 all channels are displayed in the same row
17628 all channels are displayed in separate rows
17630 Default value is @samp{combined}.
17633 Specify display color mode.
17635 It accepts the following values:
17638 each channel is displayed in a separate color
17640 each channel is displayed using the same color scheme
17642 each channel is displayed using the rainbow color scheme
17644 each channel is displayed using the moreland color scheme
17646 each channel is displayed using the nebulae color scheme
17648 each channel is displayed using the fire color scheme
17650 each channel is displayed using the fiery color scheme
17652 each channel is displayed using the fruit color scheme
17654 each channel is displayed using the cool color scheme
17656 Default value is @samp{intensity}.
17659 Specify scale used for calculating intensity color values.
17661 It accepts the following values:
17666 square root, default
17676 Default value is @samp{log}.
17679 Set saturation modifier for displayed colors. Negative values provide
17680 alternative color scheme. @code{0} is no saturation at all.
17681 Saturation must be in [-10.0, 10.0] range.
17682 Default value is @code{1}.
17685 Set window function.
17687 It accepts the following values:
17710 Default value is @code{hann}.
17713 Set orientation of time vs frequency axis. Can be @code{vertical} or
17714 @code{horizontal}. Default is @code{vertical}.
17717 Set scale gain for calculating intensity color values.
17718 Default value is @code{1}.
17721 Draw time and frequency axes and legends. Default is enabled.
17724 Set color rotation, must be in [-1.0, 1.0] range.
17725 Default value is @code{0}.
17728 @subsection Examples
17732 Extract an audio spectrogram of a whole audio track
17733 in a 1024x1024 picture using @command{ffmpeg}:
17735 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
17739 @section showvolume
17741 Convert input audio volume to a video output.
17743 The filter accepts the following options:
17750 Set border width, allowed range is [0, 5]. Default is 1.
17753 Set channel width, allowed range is [80, 8192]. Default is 400.
17756 Set channel height, allowed range is [1, 900]. Default is 20.
17759 Set fade, allowed range is [0.001, 1]. Default is 0.95.
17762 Set volume color expression.
17764 The expression can use the following variables:
17768 Current max volume of channel in dB.
17774 Current channel number, starting from 0.
17778 If set, displays channel names. Default is enabled.
17781 If set, displays volume values. Default is enabled.
17784 Set orientation, can be @code{horizontal} or @code{vertical},
17785 default is @code{horizontal}.
17788 Set step size, allowed range s [0, 5]. Default is 0, which means
17794 Convert input audio to a video output, representing the samples waves.
17796 The filter accepts the following options:
17800 Specify the video size for the output. For the syntax of this option, check the
17801 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17802 Default value is @code{600x240}.
17807 Available values are:
17810 Draw a point for each sample.
17813 Draw a vertical line for each sample.
17816 Draw a point for each sample and a line between them.
17819 Draw a centered vertical line for each sample.
17822 Default value is @code{point}.
17825 Set the number of samples which are printed on the same column. A
17826 larger value will decrease the frame rate. Must be a positive
17827 integer. This option can be set only if the value for @var{rate}
17828 is not explicitly specified.
17831 Set the (approximate) output frame rate. This is done by setting the
17832 option @var{n}. Default value is "25".
17834 @item split_channels
17835 Set if channels should be drawn separately or overlap. Default value is 0.
17838 Set colors separated by '|' which are going to be used for drawing of each channel.
17841 Set amplitude scale.
17843 Available values are:
17861 @subsection Examples
17865 Output the input file audio and the corresponding video representation
17868 amovie=a.mp3,asplit[out0],showwaves[out1]
17872 Create a synthetic signal and show it with showwaves, forcing a
17873 frame rate of 30 frames per second:
17875 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
17879 @section showwavespic
17881 Convert input audio to a single video frame, representing the samples waves.
17883 The filter accepts the following options:
17887 Specify the video size for the output. For the syntax of this option, check the
17888 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17889 Default value is @code{600x240}.
17891 @item split_channels
17892 Set if channels should be drawn separately or overlap. Default value is 0.
17895 Set colors separated by '|' which are going to be used for drawing of each channel.
17898 Set amplitude scale. Can be linear @code{lin} or logarithmic @code{log}.
17902 @subsection Examples
17906 Extract a channel split representation of the wave form of a whole audio track
17907 in a 1024x800 picture using @command{ffmpeg}:
17909 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
17913 @section sidedata, asidedata
17915 Delete frame side data, or select frames based on it.
17917 This filter accepts the following options:
17921 Set mode of operation of the filter.
17923 Can be one of the following:
17927 Select every frame with side data of @code{type}.
17930 Delete side data of @code{type}. If @code{type} is not set, delete all side
17936 Set side data type used with all modes. Must be set for @code{select} mode. For
17937 the list of frame side data types, refer to the @code{AVFrameSideDataType} enum
17938 in @file{libavutil/frame.h}. For example, to choose
17939 @code{AV_FRAME_DATA_PANSCAN} side data, you must specify @code{PANSCAN}.
17943 @section spectrumsynth
17945 Sythesize audio from 2 input video spectrums, first input stream represents
17946 magnitude across time and second represents phase across time.
17947 The filter will transform from frequency domain as displayed in videos back
17948 to time domain as presented in audio output.
17950 This filter is primarily created for reversing processed @ref{showspectrum}
17951 filter outputs, but can synthesize sound from other spectrograms too.
17952 But in such case results are going to be poor if the phase data is not
17953 available, because in such cases phase data need to be recreated, usually
17954 its just recreated from random noise.
17955 For best results use gray only output (@code{channel} color mode in
17956 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
17957 @code{lin} scale for phase video. To produce phase, for 2nd video, use
17958 @code{data} option. Inputs videos should generally use @code{fullframe}
17959 slide mode as that saves resources needed for decoding video.
17961 The filter accepts the following options:
17965 Specify sample rate of output audio, the sample rate of audio from which
17966 spectrum was generated may differ.
17969 Set number of channels represented in input video spectrums.
17972 Set scale which was used when generating magnitude input spectrum.
17973 Can be @code{lin} or @code{log}. Default is @code{log}.
17976 Set slide which was used when generating inputs spectrums.
17977 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
17978 Default is @code{fullframe}.
17981 Set window function used for resynthesis.
17984 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
17985 which means optimal overlap for selected window function will be picked.
17988 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
17989 Default is @code{vertical}.
17992 @subsection Examples
17996 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
17997 then resynthesize videos back to audio with spectrumsynth:
17999 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
18000 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
18001 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
18005 @section split, asplit
18007 Split input into several identical outputs.
18009 @code{asplit} works with audio input, @code{split} with video.
18011 The filter accepts a single parameter which specifies the number of outputs. If
18012 unspecified, it defaults to 2.
18014 @subsection Examples
18018 Create two separate outputs from the same input:
18020 [in] split [out0][out1]
18024 To create 3 or more outputs, you need to specify the number of
18027 [in] asplit=3 [out0][out1][out2]
18031 Create two separate outputs from the same input, one cropped and
18034 [in] split [splitout1][splitout2];
18035 [splitout1] crop=100:100:0:0 [cropout];
18036 [splitout2] pad=200:200:100:100 [padout];
18040 Create 5 copies of the input audio with @command{ffmpeg}:
18042 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
18048 Receive commands sent through a libzmq client, and forward them to
18049 filters in the filtergraph.
18051 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
18052 must be inserted between two video filters, @code{azmq} between two
18055 To enable these filters you need to install the libzmq library and
18056 headers and configure FFmpeg with @code{--enable-libzmq}.
18058 For more information about libzmq see:
18059 @url{http://www.zeromq.org/}
18061 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
18062 receives messages sent through a network interface defined by the
18063 @option{bind_address} option.
18065 The received message must be in the form:
18067 @var{TARGET} @var{COMMAND} [@var{ARG}]
18070 @var{TARGET} specifies the target of the command, usually the name of
18071 the filter class or a specific filter instance name.
18073 @var{COMMAND} specifies the name of the command for the target filter.
18075 @var{ARG} is optional and specifies the optional argument list for the
18076 given @var{COMMAND}.
18078 Upon reception, the message is processed and the corresponding command
18079 is injected into the filtergraph. Depending on the result, the filter
18080 will send a reply to the client, adopting the format:
18082 @var{ERROR_CODE} @var{ERROR_REASON}
18086 @var{MESSAGE} is optional.
18088 @subsection Examples
18090 Look at @file{tools/zmqsend} for an example of a zmq client which can
18091 be used to send commands processed by these filters.
18093 Consider the following filtergraph generated by @command{ffplay}
18095 ffplay -dumpgraph 1 -f lavfi "
18096 color=s=100x100:c=red [l];
18097 color=s=100x100:c=blue [r];
18098 nullsrc=s=200x100, zmq [bg];
18099 [bg][l] overlay [bg+l];
18100 [bg+l][r] overlay=x=100 "
18103 To change the color of the left side of the video, the following
18104 command can be used:
18106 echo Parsed_color_0 c yellow | tools/zmqsend
18109 To change the right side:
18111 echo Parsed_color_1 c pink | tools/zmqsend
18114 @c man end MULTIMEDIA FILTERS
18116 @chapter Multimedia Sources
18117 @c man begin MULTIMEDIA SOURCES
18119 Below is a description of the currently available multimedia sources.
18123 This is the same as @ref{movie} source, except it selects an audio
18129 Read audio and/or video stream(s) from a movie container.
18131 It accepts the following parameters:
18135 The name of the resource to read (not necessarily a file; it can also be a
18136 device or a stream accessed through some protocol).
18138 @item format_name, f
18139 Specifies the format assumed for the movie to read, and can be either
18140 the name of a container or an input device. If not specified, the
18141 format is guessed from @var{movie_name} or by probing.
18143 @item seek_point, sp
18144 Specifies the seek point in seconds. The frames will be output
18145 starting from this seek point. The parameter is evaluated with
18146 @code{av_strtod}, so the numerical value may be suffixed by an IS
18147 postfix. The default value is "0".
18150 Specifies the streams to read. Several streams can be specified,
18151 separated by "+". The source will then have as many outputs, in the
18152 same order. The syntax is explained in the ``Stream specifiers''
18153 section in the ffmpeg manual. Two special names, "dv" and "da" specify
18154 respectively the default (best suited) video and audio stream. Default
18155 is "dv", or "da" if the filter is called as "amovie".
18157 @item stream_index, si
18158 Specifies the index of the video stream to read. If the value is -1,
18159 the most suitable video stream will be automatically selected. The default
18160 value is "-1". Deprecated. If the filter is called "amovie", it will select
18161 audio instead of video.
18164 Specifies how many times to read the stream in sequence.
18165 If the value is 0, the stream will be looped infinitely.
18166 Default value is "1".
18168 Note that when the movie is looped the source timestamps are not
18169 changed, so it will generate non monotonically increasing timestamps.
18171 @item discontinuity
18172 Specifies the time difference between frames above which the point is
18173 considered a timestamp discontinuity which is removed by adjusting the later
18177 It allows overlaying a second video on top of the main input of
18178 a filtergraph, as shown in this graph:
18180 input -----------> deltapts0 --> overlay --> output
18183 movie --> scale--> deltapts1 -------+
18185 @subsection Examples
18189 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
18190 on top of the input labelled "in":
18192 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
18193 [in] setpts=PTS-STARTPTS [main];
18194 [main][over] overlay=16:16 [out]
18198 Read from a video4linux2 device, and overlay it on top of the input
18201 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
18202 [in] setpts=PTS-STARTPTS [main];
18203 [main][over] overlay=16:16 [out]
18207 Read the first video stream and the audio stream with id 0x81 from
18208 dvd.vob; the video is connected to the pad named "video" and the audio is
18209 connected to the pad named "audio":
18211 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
18215 @subsection Commands
18217 Both movie and amovie support the following commands:
18220 Perform seek using "av_seek_frame".
18221 The syntax is: seek @var{stream_index}|@var{timestamp}|@var{flags}
18224 @var{stream_index}: If stream_index is -1, a default
18225 stream is selected, and @var{timestamp} is automatically converted
18226 from AV_TIME_BASE units to the stream specific time_base.
18228 @var{timestamp}: Timestamp in AVStream.time_base units
18229 or, if no stream is specified, in AV_TIME_BASE units.
18231 @var{flags}: Flags which select direction and seeking mode.
18235 Get movie duration in AV_TIME_BASE units.
18239 @c man end MULTIMEDIA SOURCES