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 @c man end FILTERGRAPH DESCRIPTION
311 @chapter Audio Filters
312 @c man begin AUDIO FILTERS
314 When you configure your FFmpeg build, you can disable any of the
315 existing filters using @code{--disable-filters}.
316 The configure output will show the audio filters included in your
319 Below is a description of the currently available audio filters.
323 A compressor is mainly used to reduce the dynamic range of a signal.
324 Especially modern music is mostly compressed at a high ratio to
325 improve the overall loudness. It's done to get the highest attention
326 of a listener, "fatten" the sound and bring more "power" to the track.
327 If a signal is compressed too much it may sound dull or "dead"
328 afterwards or it may start to "pump" (which could be a powerful effect
329 but can also destroy a track completely).
330 The right compression is the key to reach a professional sound and is
331 the high art of mixing and mastering. Because of its complex settings
332 it may take a long time to get the right feeling for this kind of effect.
334 Compression is done by detecting the volume above a chosen level
335 @code{threshold} and dividing it by the factor set with @code{ratio}.
336 So if you set the threshold to -12dB and your signal reaches -6dB a ratio
337 of 2:1 will result in a signal at -9dB. Because an exact manipulation of
338 the signal would cause distortion of the waveform the reduction can be
339 levelled over the time. This is done by setting "Attack" and "Release".
340 @code{attack} determines how long the signal has to rise above the threshold
341 before any reduction will occur and @code{release} sets the time the signal
342 has to fall below the threshold to reduce the reduction again. Shorter signals
343 than the chosen attack time will be left untouched.
344 The overall reduction of the signal can be made up afterwards with the
345 @code{makeup} setting. So compressing the peaks of a signal about 6dB and
346 raising the makeup to this level results in a signal twice as loud than the
347 source. To gain a softer entry in the compression the @code{knee} flattens the
348 hard edge at the threshold in the range of the chosen decibels.
350 The filter accepts the following options:
354 Set input gain. Default is 1. Range is between 0.015625 and 64.
357 If a signal of second stream rises above this level it will affect the gain
358 reduction of the first stream.
359 By default it is 0.125. Range is between 0.00097563 and 1.
362 Set a ratio by which the signal is reduced. 1:2 means that if the level
363 rose 4dB above the threshold, it will be only 2dB above after the reduction.
364 Default is 2. Range is between 1 and 20.
367 Amount of milliseconds the signal has to rise above the threshold before gain
368 reduction starts. Default is 20. Range is between 0.01 and 2000.
371 Amount of milliseconds the signal has to fall below the threshold before
372 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
375 Set the amount by how much signal will be amplified after processing.
376 Default is 2. Range is from 1 and 64.
379 Curve the sharp knee around the threshold to enter gain reduction more softly.
380 Default is 2.82843. Range is between 1 and 8.
383 Choose if the @code{average} level between all channels of input stream
384 or the louder(@code{maximum}) channel of input stream affects the
385 reduction. Default is @code{average}.
388 Should the exact signal be taken in case of @code{peak} or an RMS one in case
389 of @code{rms}. Default is @code{rms} which is mostly smoother.
392 How much to use compressed signal in output. Default is 1.
393 Range is between 0 and 1.
398 Apply cross fade from one input audio stream to another input audio stream.
399 The cross fade is applied for specified duration near the end of first stream.
401 The filter accepts the following options:
405 Specify the number of samples for which the cross fade effect has to last.
406 At the end of the cross fade effect the first input audio will be completely
407 silent. Default is 44100.
410 Specify the duration of the cross fade effect. See
411 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
412 for the accepted syntax.
413 By default the duration is determined by @var{nb_samples}.
414 If set this option is used instead of @var{nb_samples}.
417 Should first stream end overlap with second stream start. Default is enabled.
420 Set curve for cross fade transition for first stream.
423 Set curve for cross fade transition for second stream.
425 For description of available curve types see @ref{afade} filter description.
432 Cross fade from one input to another:
434 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
438 Cross fade from one input to another but without overlapping:
440 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
446 Delay one or more audio channels.
448 Samples in delayed channel are filled with silence.
450 The filter accepts the following option:
454 Set list of delays in milliseconds for each channel separated by '|'.
455 At least one delay greater than 0 should be provided.
456 Unused delays will be silently ignored. If number of given delays is
457 smaller than number of channels all remaining channels will not be delayed.
464 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
465 the second channel (and any other channels that may be present) unchanged.
473 Apply echoing to the input audio.
475 Echoes are reflected sound and can occur naturally amongst mountains
476 (and sometimes large buildings) when talking or shouting; digital echo
477 effects emulate this behaviour and are often used to help fill out the
478 sound of a single instrument or vocal. The time difference between the
479 original signal and the reflection is the @code{delay}, and the
480 loudness of the reflected signal is the @code{decay}.
481 Multiple echoes can have different delays and decays.
483 A description of the accepted parameters follows.
487 Set input gain of reflected signal. Default is @code{0.6}.
490 Set output gain of reflected signal. Default is @code{0.3}.
493 Set list of time intervals in milliseconds between original signal and reflections
494 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
495 Default is @code{1000}.
498 Set list of loudnesses of reflected signals separated by '|'.
499 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
500 Default is @code{0.5}.
507 Make it sound as if there are twice as many instruments as are actually playing:
509 aecho=0.8:0.88:60:0.4
513 If delay is very short, then it sound like a (metallic) robot playing music:
519 A longer delay will sound like an open air concert in the mountains:
521 aecho=0.8:0.9:1000:0.3
525 Same as above but with one more mountain:
527 aecho=0.8:0.9:1000|1800:0.3|0.25
532 Audio emphasis filter creates or restores material directly taken from LPs or
533 emphased CDs with different filter curves. E.g. to store music on vinyl the
534 signal has to be altered by a filter first to even out the disadvantages of
535 this recording medium.
536 Once the material is played back the inverse filter has to be applied to
537 restore the distortion of the frequency response.
539 The filter accepts the following options:
549 Set filter mode. For restoring material use @code{reproduction} mode, otherwise
550 use @code{production} mode. Default is @code{reproduction} mode.
553 Set filter type. Selects medium. Can be one of the following:
565 select Compact Disc (CD).
571 select 50µs (FM-KF).
573 select 75µs (FM-KF).
579 Modify an audio signal according to the specified expressions.
581 This filter accepts one or more expressions (one for each channel),
582 which are evaluated and used to modify a corresponding audio signal.
584 It accepts the following parameters:
588 Set the '|'-separated expressions list for each separate channel. If
589 the number of input channels is greater than the number of
590 expressions, the last specified expression is used for the remaining
593 @item channel_layout, c
594 Set output channel layout. If not specified, the channel layout is
595 specified by the number of expressions. If set to @samp{same}, it will
596 use by default the same input channel layout.
599 Each expression in @var{exprs} can contain the following constants and functions:
603 channel number of the current expression
606 number of the evaluated sample, starting from 0
612 time of the evaluated sample expressed in seconds
615 @item nb_out_channels
616 input and output number of channels
619 the value of input channel with number @var{CH}
622 Note: this filter is slow. For faster processing you should use a
631 aeval=val(ch)/2:c=same
635 Invert phase of the second channel:
644 Apply fade-in/out effect to input audio.
646 A description of the accepted parameters follows.
650 Specify the effect type, can be either @code{in} for fade-in, or
651 @code{out} for a fade-out effect. Default is @code{in}.
653 @item start_sample, ss
654 Specify the number of the start sample for starting to apply the fade
655 effect. Default is 0.
658 Specify the number of samples for which the fade effect has to last. At
659 the end of the fade-in effect the output audio will have the same
660 volume as the input audio, at the end of the fade-out transition
661 the output audio will be silence. Default is 44100.
664 Specify the start time of the fade effect. Default is 0.
665 The value must be specified as a time duration; see
666 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
667 for the accepted syntax.
668 If set this option is used instead of @var{start_sample}.
671 Specify the duration of the fade effect. See
672 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
673 for the accepted syntax.
674 At the end of the fade-in effect the output audio will have the same
675 volume as the input audio, at the end of the fade-out transition
676 the output audio will be silence.
677 By default the duration is determined by @var{nb_samples}.
678 If set this option is used instead of @var{nb_samples}.
681 Set curve for fade transition.
683 It accepts the following values:
686 select triangular, linear slope (default)
688 select quarter of sine wave
690 select half of sine wave
692 select exponential sine wave
696 select inverted parabola
710 select inverted quarter of sine wave
712 select inverted half of sine wave
714 select double-exponential seat
716 select double-exponential sigmoid
724 Fade in first 15 seconds of audio:
730 Fade out last 25 seconds of a 900 seconds audio:
732 afade=t=out:st=875:d=25
737 Apply arbitrary expressions to samples in frequency domain.
741 Set frequency domain real expression for each separate channel separated
742 by '|'. Default is "1".
743 If the number of input channels is greater than the number of
744 expressions, the last specified expression is used for the remaining
748 Set frequency domain imaginary expression for each separate channel
749 separated by '|'. If not set, @var{real} option is used.
751 Each expression in @var{real} and @var{imag} can contain the following
759 current frequency bin number
762 number of available bins
765 channel number of the current expression
777 It accepts the following values:
793 Default is @code{w4096}
796 Set window function. Default is @code{hann}.
799 Set window overlap. If set to 1, the recommended overlap for selected
800 window function will be picked. Default is @code{0.75}.
807 Leave almost only low frequencies in audio:
809 afftfilt="1-clip((b/nb)*b,0,1)"
816 Set output format constraints for the input audio. The framework will
817 negotiate the most appropriate format to minimize conversions.
819 It accepts the following parameters:
823 A '|'-separated list of requested sample formats.
826 A '|'-separated list of requested sample rates.
828 @item channel_layouts
829 A '|'-separated list of requested channel layouts.
831 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
832 for the required syntax.
835 If a parameter is omitted, all values are allowed.
837 Force the output to either unsigned 8-bit or signed 16-bit stereo
839 aformat=sample_fmts=u8|s16:channel_layouts=stereo
844 A gate is mainly used to reduce lower parts of a signal. This kind of signal
845 processing reduces disturbing noise between useful signals.
847 Gating is done by detecting the volume below a chosen level @var{threshold}
848 and divide it by the factor set with @var{ratio}. The bottom of the noise
849 floor is set via @var{range}. Because an exact manipulation of the signal
850 would cause distortion of the waveform the reduction can be levelled over
851 time. This is done by setting @var{attack} and @var{release}.
853 @var{attack} determines how long the signal has to fall below the threshold
854 before any reduction will occur and @var{release} sets the time the signal
855 has to raise above the threshold to reduce the reduction again.
856 Shorter signals than the chosen attack time will be left untouched.
860 Set input level before filtering.
861 Default is 1. Allowed range is from 0.015625 to 64.
864 Set the level of gain reduction when the signal is below the threshold.
865 Default is 0.06125. Allowed range is from 0 to 1.
868 If a signal rises above this level the gain reduction is released.
869 Default is 0.125. Allowed range is from 0 to 1.
872 Set a ratio about which the signal is reduced.
873 Default is 2. Allowed range is from 1 to 9000.
876 Amount of milliseconds the signal has to rise above the threshold before gain
878 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
881 Amount of milliseconds the signal has to fall below the threshold before the
882 reduction is increased again. Default is 250 milliseconds.
883 Allowed range is from 0.01 to 9000.
886 Set amount of amplification of signal after processing.
887 Default is 1. Allowed range is from 1 to 64.
890 Curve the sharp knee around the threshold to enter gain reduction more softly.
891 Default is 2.828427125. Allowed range is from 1 to 8.
894 Choose if exact signal should be taken for detection or an RMS like one.
895 Default is rms. Can be peak or rms.
898 Choose if the average level between all channels or the louder channel affects
900 Default is average. Can be average or maximum.
905 The limiter prevents input signal from raising over a desired threshold.
906 This limiter uses lookahead technology to prevent your signal from distorting.
907 It means that there is a small delay after signal is processed. Keep in mind
908 that the delay it produces is the attack time you set.
910 The filter accepts the following options:
914 Set input gain. Default is 1.
917 Set output gain. Default is 1.
920 Don't let signals above this level pass the limiter. Default is 1.
923 The limiter will reach its attenuation level in this amount of time in
924 milliseconds. Default is 5 milliseconds.
927 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
928 Default is 50 milliseconds.
931 When gain reduction is always needed ASC takes care of releasing to an
932 average reduction level rather than reaching a reduction of 0 in the release
936 Select how much the release time is affected by ASC, 0 means nearly no changes
937 in release time while 1 produces higher release times.
940 Auto level output signal. Default is enabled.
941 This normalizes audio back to 0dB if enabled.
944 Depending on picked setting it is recommended to upsample input 2x or 4x times
945 with @ref{aresample} before applying this filter.
949 Apply a two-pole all-pass filter with central frequency (in Hz)
950 @var{frequency}, and filter-width @var{width}.
951 An all-pass filter changes the audio's frequency to phase relationship
952 without changing its frequency to amplitude relationship.
954 The filter accepts the following options:
961 Set method to specify band-width of filter.
974 Specify the band-width of a filter in width_type units.
981 The filter accepts the following options:
985 Set the number of loops.
988 Set maximal number of samples.
991 Set first sample of loop.
997 Merge two or more audio streams into a single multi-channel stream.
999 The filter accepts the following options:
1004 Set the number of inputs. Default is 2.
1008 If the channel layouts of the inputs are disjoint, and therefore compatible,
1009 the channel layout of the output will be set accordingly and the channels
1010 will be reordered as necessary. If the channel layouts of the inputs are not
1011 disjoint, the output will have all the channels of the first input then all
1012 the channels of the second input, in that order, and the channel layout of
1013 the output will be the default value corresponding to the total number of
1016 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
1017 is FC+BL+BR, then the output will be in 5.1, with the channels in the
1018 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
1019 first input, b1 is the first channel of the second input).
1021 On the other hand, if both input are in stereo, the output channels will be
1022 in the default order: a1, a2, b1, b2, and the channel layout will be
1023 arbitrarily set to 4.0, which may or may not be the expected value.
1025 All inputs must have the same sample rate, and format.
1027 If inputs do not have the same duration, the output will stop with the
1030 @subsection Examples
1034 Merge two mono files into a stereo stream:
1036 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
1040 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
1042 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
1048 Mixes multiple audio inputs into a single output.
1050 Note that this filter only supports float samples (the @var{amerge}
1051 and @var{pan} audio filters support many formats). If the @var{amix}
1052 input has integer samples then @ref{aresample} will be automatically
1053 inserted to perform the conversion to float samples.
1057 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
1059 will mix 3 input audio streams to a single output with the same duration as the
1060 first input and a dropout transition time of 3 seconds.
1062 It accepts the following parameters:
1066 The number of inputs. If unspecified, it defaults to 2.
1069 How to determine the end-of-stream.
1073 The duration of the longest input. (default)
1076 The duration of the shortest input.
1079 The duration of the first input.
1083 @item dropout_transition
1084 The transition time, in seconds, for volume renormalization when an input
1085 stream ends. The default value is 2 seconds.
1089 @section anequalizer
1091 High-order parametric multiband equalizer for each channel.
1093 It accepts the following parameters:
1097 This option string is in format:
1098 "c@var{chn} f=@var{cf} w=@var{w} g=@var{g} t=@var{f} | ..."
1099 Each equalizer band is separated by '|'.
1103 Set channel number to which equalization will be applied.
1104 If input doesn't have that channel the entry is ignored.
1107 Set central frequency for band.
1108 If input doesn't have that frequency the entry is ignored.
1111 Set band width in hertz.
1114 Set band gain in dB.
1117 Set filter type for band, optional, can be:
1121 Butterworth, this is default.
1132 With this option activated frequency response of anequalizer is displayed
1136 Set video stream size. Only useful if curves option is activated.
1139 Set max gain that will be displayed. Only useful if curves option is activated.
1140 Setting this to reasonable value allows to display gain which is derived from
1141 neighbour bands which are too close to each other and thus produce higher gain
1142 when both are activated.
1145 Set frequency scale used to draw frequency response in video output.
1146 Can be linear or logarithmic. Default is logarithmic.
1149 Set color for each channel curve which is going to be displayed in video stream.
1150 This is list of color names separated by space or by '|'.
1151 Unrecognised or missing colors will be replaced by white color.
1154 @subsection Examples
1158 Lower gain by 10 of central frequency 200Hz and width 100 Hz
1159 for first 2 channels using Chebyshev type 1 filter:
1161 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1165 @subsection Commands
1167 This filter supports the following commands:
1170 Alter existing filter parameters.
1171 Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1173 @var{fN} is existing filter number, starting from 0, if no such filter is available
1175 @var{freq} set new frequency parameter.
1176 @var{width} set new width parameter in herz.
1177 @var{gain} set new gain parameter in dB.
1179 Full filter invocation with asendcmd may look like this:
1180 asendcmd=c='4.0 anequalizer change 0|f=200|w=50|g=1',anequalizer=...
1185 Pass the audio source unchanged to the output.
1189 Pad the end of an audio stream with silence.
1191 This can be used together with @command{ffmpeg} @option{-shortest} to
1192 extend audio streams to the same length as the video stream.
1194 A description of the accepted options follows.
1198 Set silence packet size. Default value is 4096.
1201 Set the number of samples of silence to add to the end. After the
1202 value is reached, the stream is terminated. This option is mutually
1203 exclusive with @option{whole_len}.
1206 Set the minimum total number of samples in the output audio stream. If
1207 the value is longer than the input audio length, silence is added to
1208 the end, until the value is reached. This option is mutually exclusive
1209 with @option{pad_len}.
1212 If neither the @option{pad_len} nor the @option{whole_len} option is
1213 set, the filter will add silence to the end of the input stream
1216 @subsection Examples
1220 Add 1024 samples of silence to the end of the input:
1226 Make sure the audio output will contain at least 10000 samples, pad
1227 the input with silence if required:
1229 apad=whole_len=10000
1233 Use @command{ffmpeg} to pad the audio input with silence, so that the
1234 video stream will always result the shortest and will be converted
1235 until the end in the output file when using the @option{shortest}
1238 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1243 Add a phasing effect to the input audio.
1245 A phaser filter creates series of peaks and troughs in the frequency spectrum.
1246 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1248 A description of the accepted parameters follows.
1252 Set input gain. Default is 0.4.
1255 Set output gain. Default is 0.74
1258 Set delay in milliseconds. Default is 3.0.
1261 Set decay. Default is 0.4.
1264 Set modulation speed in Hz. Default is 0.5.
1267 Set modulation type. Default is triangular.
1269 It accepts the following values:
1278 Audio pulsator is something between an autopanner and a tremolo.
1279 But it can produce funny stereo effects as well. Pulsator changes the volume
1280 of the left and right channel based on a LFO (low frequency oscillator) with
1281 different waveforms and shifted phases.
1282 This filter have the ability to define an offset between left and right
1283 channel. An offset of 0 means that both LFO shapes match each other.
1284 The left and right channel are altered equally - a conventional tremolo.
1285 An offset of 50% means that the shape of the right channel is exactly shifted
1286 in phase (or moved backwards about half of the frequency) - pulsator acts as
1287 an autopanner. At 1 both curves match again. Every setting in between moves the
1288 phase shift gapless between all stages and produces some "bypassing" sounds with
1289 sine and triangle waveforms. The more you set the offset near 1 (starting from
1290 the 0.5) the faster the signal passes from the left to the right speaker.
1292 The filter accepts the following options:
1296 Set input gain. By default it is 1. Range is [0.015625 - 64].
1299 Set output gain. By default it is 1. Range is [0.015625 - 64].
1302 Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1303 sawup or sawdown. Default is sine.
1306 Set modulation. Define how much of original signal is affected by the LFO.
1309 Set left channel offset. Default is 0. Allowed range is [0 - 1].
1312 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1315 Set pulse width. Default is 1. Allowed range is [0 - 2].
1318 Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1321 Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1325 Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1329 Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1330 if timing is set to hz.
1336 Resample the input audio to the specified parameters, using the
1337 libswresample library. If none are specified then the filter will
1338 automatically convert between its input and output.
1340 This filter is also able to stretch/squeeze the audio data to make it match
1341 the timestamps or to inject silence / cut out audio to make it match the
1342 timestamps, do a combination of both or do neither.
1344 The filter accepts the syntax
1345 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1346 expresses a sample rate and @var{resampler_options} is a list of
1347 @var{key}=@var{value} pairs, separated by ":". See the
1348 ffmpeg-resampler manual for the complete list of supported options.
1350 @subsection Examples
1354 Resample the input audio to 44100Hz:
1360 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1361 samples per second compensation:
1363 aresample=async=1000
1369 Reverse an audio clip.
1371 Warning: This filter requires memory to buffer the entire clip, so trimming
1374 @subsection Examples
1378 Take the first 5 seconds of a clip, and reverse it.
1380 atrim=end=5,areverse
1384 @section asetnsamples
1386 Set the number of samples per each output audio frame.
1388 The last output packet may contain a different number of samples, as
1389 the filter will flush all the remaining samples when the input audio
1392 The filter accepts the following options:
1396 @item nb_out_samples, n
1397 Set the number of frames per each output audio frame. The number is
1398 intended as the number of samples @emph{per each channel}.
1399 Default value is 1024.
1402 If set to 1, the filter will pad the last audio frame with zeroes, so
1403 that the last frame will contain the same number of samples as the
1404 previous ones. Default value is 1.
1407 For example, to set the number of per-frame samples to 1234 and
1408 disable padding for the last frame, use:
1410 asetnsamples=n=1234:p=0
1415 Set the sample rate without altering the PCM data.
1416 This will result in a change of speed and pitch.
1418 The filter accepts the following options:
1421 @item sample_rate, r
1422 Set the output sample rate. Default is 44100 Hz.
1427 Show a line containing various information for each input audio frame.
1428 The input audio is not modified.
1430 The shown line contains a sequence of key/value pairs of the form
1431 @var{key}:@var{value}.
1433 The following values are shown in the output:
1437 The (sequential) number of the input frame, starting from 0.
1440 The presentation timestamp of the input frame, in time base units; the time base
1441 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1444 The presentation timestamp of the input frame in seconds.
1447 position of the frame in the input stream, -1 if this information in
1448 unavailable and/or meaningless (for example in case of synthetic audio)
1457 The sample rate for the audio frame.
1460 The number of samples (per channel) in the frame.
1463 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1464 audio, the data is treated as if all the planes were concatenated.
1466 @item plane_checksums
1467 A list of Adler-32 checksums for each data plane.
1473 Display time domain statistical information about the audio channels.
1474 Statistics are calculated and displayed for each audio channel and,
1475 where applicable, an overall figure is also given.
1477 It accepts the following option:
1480 Short window length in seconds, used for peak and trough RMS measurement.
1481 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1485 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1486 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1489 Available keys for each channel are:
1520 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1521 this @code{lavfi.astats.Overall.Peak_count}.
1523 For description what each key means read below.
1526 Set number of frame after which stats are going to be recalculated.
1527 Default is disabled.
1530 A description of each shown parameter follows:
1534 Mean amplitude displacement from zero.
1537 Minimal sample level.
1540 Maximal sample level.
1542 @item Min difference
1543 Minimal difference between two consecutive samples.
1545 @item Max difference
1546 Maximal difference between two consecutive samples.
1548 @item Mean difference
1549 Mean difference between two consecutive samples.
1550 The average of each difference between two consecutive samples.
1554 Standard peak and RMS level measured in dBFS.
1558 Peak and trough values for RMS level measured over a short window.
1561 Standard ratio of peak to RMS level (note: not in dB).
1564 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1565 (i.e. either @var{Min level} or @var{Max level}).
1568 Number of occasions (not the number of samples) that the signal attained either
1569 @var{Min level} or @var{Max level}.
1572 Overall bit depth of audio. Number of bits used for each sample.
1577 Synchronize audio data with timestamps by squeezing/stretching it and/or
1578 dropping samples/adding silence when needed.
1580 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1582 It accepts the following parameters:
1586 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1587 by default. When disabled, time gaps are covered with silence.
1590 The minimum difference between timestamps and audio data (in seconds) to trigger
1591 adding/dropping samples. The default value is 0.1. If you get an imperfect
1592 sync with this filter, try setting this parameter to 0.
1595 The maximum compensation in samples per second. Only relevant with compensate=1.
1596 The default value is 500.
1599 Assume that the first PTS should be this value. The time base is 1 / sample
1600 rate. This allows for padding/trimming at the start of the stream. By default,
1601 no assumption is made about the first frame's expected PTS, so no padding or
1602 trimming is done. For example, this could be set to 0 to pad the beginning with
1603 silence if an audio stream starts after the video stream or to trim any samples
1604 with a negative PTS due to encoder delay.
1612 The filter accepts exactly one parameter, the audio tempo. If not
1613 specified then the filter will assume nominal 1.0 tempo. Tempo must
1614 be in the [0.5, 2.0] range.
1616 @subsection Examples
1620 Slow down audio to 80% tempo:
1626 To speed up audio to 125% tempo:
1634 Trim the input so that the output contains one continuous subpart of the input.
1636 It accepts the following parameters:
1639 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1640 sample with the timestamp @var{start} will be the first sample in the output.
1643 Specify time of the first audio sample that will be dropped, i.e. the
1644 audio sample immediately preceding the one with the timestamp @var{end} will be
1645 the last sample in the output.
1648 Same as @var{start}, except this option sets the start timestamp in samples
1652 Same as @var{end}, except this option sets the end timestamp in samples instead
1656 The maximum duration of the output in seconds.
1659 The number of the first sample that should be output.
1662 The number of the first sample that should be dropped.
1665 @option{start}, @option{end}, and @option{duration} are expressed as time
1666 duration specifications; see
1667 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1669 Note that the first two sets of the start/end options and the @option{duration}
1670 option look at the frame timestamp, while the _sample options simply count the
1671 samples that pass through the filter. So start/end_pts and start/end_sample will
1672 give different results when the timestamps are wrong, inexact or do not start at
1673 zero. Also note that this filter does not modify the timestamps. If you wish
1674 to have the output timestamps start at zero, insert the asetpts filter after the
1677 If multiple start or end options are set, this filter tries to be greedy and
1678 keep all samples that match at least one of the specified constraints. To keep
1679 only the part that matches all the constraints at once, chain multiple atrim
1682 The defaults are such that all the input is kept. So it is possible to set e.g.
1683 just the end values to keep everything before the specified time.
1688 Drop everything except the second minute of input:
1690 ffmpeg -i INPUT -af atrim=60:120
1694 Keep only the first 1000 samples:
1696 ffmpeg -i INPUT -af atrim=end_sample=1000
1703 Apply a two-pole Butterworth band-pass filter with central
1704 frequency @var{frequency}, and (3dB-point) band-width width.
1705 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1706 instead of the default: constant 0dB peak gain.
1707 The filter roll off at 6dB per octave (20dB per decade).
1709 The filter accepts the following options:
1713 Set the filter's central frequency. Default is @code{3000}.
1716 Constant skirt gain if set to 1. Defaults to 0.
1719 Set method to specify band-width of filter.
1732 Specify the band-width of a filter in width_type units.
1737 Apply a two-pole Butterworth band-reject filter with central
1738 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1739 The filter roll off at 6dB per octave (20dB per decade).
1741 The filter accepts the following options:
1745 Set the filter's central frequency. Default is @code{3000}.
1748 Set method to specify band-width of filter.
1761 Specify the band-width of a filter in width_type units.
1766 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1767 shelving filter with a response similar to that of a standard
1768 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1770 The filter accepts the following options:
1774 Give the gain at 0 Hz. Its useful range is about -20
1775 (for a large cut) to +20 (for a large boost).
1776 Beware of clipping when using a positive gain.
1779 Set the filter's central frequency and so can be used
1780 to extend or reduce the frequency range to be boosted or cut.
1781 The default value is @code{100} Hz.
1784 Set method to specify band-width of filter.
1797 Determine how steep is the filter's shelf transition.
1802 Apply a biquad IIR filter with the given coefficients.
1803 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1804 are the numerator and denominator coefficients respectively.
1807 Bauer stereo to binaural transformation, which improves headphone listening of
1808 stereo audio records.
1810 It accepts the following parameters:
1814 Pre-defined crossfeed level.
1818 Default level (fcut=700, feed=50).
1821 Chu Moy circuit (fcut=700, feed=60).
1824 Jan Meier circuit (fcut=650, feed=95).
1829 Cut frequency (in Hz).
1838 Remap input channels to new locations.
1840 It accepts the following parameters:
1842 @item channel_layout
1843 The channel layout of the output stream.
1846 Map channels from input to output. The argument is a '|'-separated list of
1847 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1848 @var{in_channel} form. @var{in_channel} can be either the name of the input
1849 channel (e.g. FL for front left) or its index in the input channel layout.
1850 @var{out_channel} is the name of the output channel or its index in the output
1851 channel layout. If @var{out_channel} is not given then it is implicitly an
1852 index, starting with zero and increasing by one for each mapping.
1855 If no mapping is present, the filter will implicitly map input channels to
1856 output channels, preserving indices.
1858 For example, assuming a 5.1+downmix input MOV file,
1860 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1862 will create an output WAV file tagged as stereo from the downmix channels of
1865 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1867 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1870 @section channelsplit
1872 Split each channel from an input audio stream into a separate output stream.
1874 It accepts the following parameters:
1876 @item channel_layout
1877 The channel layout of the input stream. The default is "stereo".
1880 For example, assuming a stereo input MP3 file,
1882 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1884 will create an output Matroska file with two audio streams, one containing only
1885 the left channel and the other the right channel.
1887 Split a 5.1 WAV file into per-channel files:
1889 ffmpeg -i in.wav -filter_complex
1890 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1891 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1892 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1897 Add a chorus effect to the audio.
1899 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1901 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1902 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1903 The modulation depth defines the range the modulated delay is played before or after
1904 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1905 sound tuned around the original one, like in a chorus where some vocals are slightly
1908 It accepts the following parameters:
1911 Set input gain. Default is 0.4.
1914 Set output gain. Default is 0.4.
1917 Set delays. A typical delay is around 40ms to 60ms.
1929 @subsection Examples
1935 chorus=0.7:0.9:55:0.4:0.25:2
1941 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1945 Fuller sounding chorus with three delays:
1947 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
1952 Compress or expand the audio's dynamic range.
1954 It accepts the following parameters:
1960 A list of times in seconds for each channel over which the instantaneous level
1961 of the input signal is averaged to determine its volume. @var{attacks} refers to
1962 increase of volume and @var{decays} refers to decrease of volume. For most
1963 situations, the attack time (response to the audio getting louder) should be
1964 shorter than the decay time, because the human ear is more sensitive to sudden
1965 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
1966 a typical value for decay is 0.8 seconds.
1967 If specified number of attacks & decays is lower than number of channels, the last
1968 set attack/decay will be used for all remaining channels.
1971 A list of points for the transfer function, specified in dB relative to the
1972 maximum possible signal amplitude. Each key points list must be defined using
1973 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
1974 @code{x0/y0 x1/y1 x2/y2 ....}
1976 The input values must be in strictly increasing order but the transfer function
1977 does not have to be monotonically rising. The point @code{0/0} is assumed but
1978 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
1979 function are @code{-70/-70|-60/-20}.
1982 Set the curve radius in dB for all joints. It defaults to 0.01.
1985 Set the additional gain in dB to be applied at all points on the transfer
1986 function. This allows for easy adjustment of the overall gain.
1990 Set an initial volume, in dB, to be assumed for each channel when filtering
1991 starts. This permits the user to supply a nominal level initially, so that, for
1992 example, a very large gain is not applied to initial signal levels before the
1993 companding has begun to operate. A typical value for audio which is initially
1994 quiet is -90 dB. It defaults to 0.
1997 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
1998 delayed before being fed to the volume adjuster. Specifying a delay
1999 approximately equal to the attack/decay times allows the filter to effectively
2000 operate in predictive rather than reactive mode. It defaults to 0.
2004 @subsection Examples
2008 Make music with both quiet and loud passages suitable for listening to in a
2011 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
2014 Another example for audio with whisper and explosion parts:
2016 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
2020 A noise gate for when the noise is at a lower level than the signal:
2022 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
2026 Here is another noise gate, this time for when the noise is at a higher level
2027 than the signal (making it, in some ways, similar to squelch):
2029 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
2033 2:1 compression starting at -6dB:
2035 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
2039 2:1 compression starting at -9dB:
2041 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
2045 2:1 compression starting at -12dB:
2047 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
2051 2:1 compression starting at -18dB:
2053 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
2057 3:1 compression starting at -15dB:
2059 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
2065 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
2071 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
2075 Hard limiter at -6dB:
2077 compand=attacks=0:points=-80/-80|-6/-6|20/-6
2081 Hard limiter at -12dB:
2083 compand=attacks=0:points=-80/-80|-12/-12|20/-12
2087 Hard noise gate at -35 dB:
2089 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
2095 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
2099 @section compensationdelay
2101 Compensation Delay Line is a metric based delay to compensate differing
2102 positions of microphones or speakers.
2104 For example, you have recorded guitar with two microphones placed in
2105 different location. Because the front of sound wave has fixed speed in
2106 normal conditions, the phasing of microphones can vary and depends on
2107 their location and interposition. The best sound mix can be achieved when
2108 these microphones are in phase (synchronized). Note that distance of
2109 ~30 cm between microphones makes one microphone to capture signal in
2110 antiphase to another microphone. That makes the final mix sounding moody.
2111 This filter helps to solve phasing problems by adding different delays
2112 to each microphone track and make them synchronized.
2114 The best result can be reached when you take one track as base and
2115 synchronize other tracks one by one with it.
2116 Remember that synchronization/delay tolerance depends on sample rate, too.
2117 Higher sample rates will give more tolerance.
2119 It accepts the following parameters:
2123 Set millimeters distance. This is compensation distance for fine tuning.
2127 Set cm distance. This is compensation distance for tightening distance setup.
2131 Set meters distance. This is compensation distance for hard distance setup.
2135 Set dry amount. Amount of unprocessed (dry) signal.
2139 Set wet amount. Amount of processed (wet) signal.
2143 Set temperature degree in Celsius. This is the temperature of the environment.
2147 @section crystalizer
2148 Simple algorithm to expand audio dynamic range.
2150 The filter accepts the following options:
2154 Sets the intensity of effect (default: 2.0). Must be in range between 0.0
2155 (unchanged sound) to 10.0 (maximum effect).
2158 Enable clipping. By default is enabled.
2162 Apply a DC shift to the audio.
2164 This can be useful to remove a DC offset (caused perhaps by a hardware problem
2165 in the recording chain) from the audio. The effect of a DC offset is reduced
2166 headroom and hence volume. The @ref{astats} filter can be used to determine if
2167 a signal has a DC offset.
2171 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
2175 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
2176 used to prevent clipping.
2180 Dynamic Audio Normalizer.
2182 This filter applies a certain amount of gain to the input audio in order
2183 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
2184 contrast to more "simple" normalization algorithms, the Dynamic Audio
2185 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
2186 This allows for applying extra gain to the "quiet" sections of the audio
2187 while avoiding distortions or clipping the "loud" sections. In other words:
2188 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
2189 sections, in the sense that the volume of each section is brought to the
2190 same target level. Note, however, that the Dynamic Audio Normalizer achieves
2191 this goal *without* applying "dynamic range compressing". It will retain 100%
2192 of the dynamic range *within* each section of the audio file.
2196 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
2197 Default is 500 milliseconds.
2198 The Dynamic Audio Normalizer processes the input audio in small chunks,
2199 referred to as frames. This is required, because a peak magnitude has no
2200 meaning for just a single sample value. Instead, we need to determine the
2201 peak magnitude for a contiguous sequence of sample values. While a "standard"
2202 normalizer would simply use the peak magnitude of the complete file, the
2203 Dynamic Audio Normalizer determines the peak magnitude individually for each
2204 frame. The length of a frame is specified in milliseconds. By default, the
2205 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
2206 been found to give good results with most files.
2207 Note that the exact frame length, in number of samples, will be determined
2208 automatically, based on the sampling rate of the individual input audio file.
2211 Set the Gaussian filter window size. In range from 3 to 301, must be odd
2212 number. Default is 31.
2213 Probably the most important parameter of the Dynamic Audio Normalizer is the
2214 @code{window size} of the Gaussian smoothing filter. The filter's window size
2215 is specified in frames, centered around the current frame. For the sake of
2216 simplicity, this must be an odd number. Consequently, the default value of 31
2217 takes into account the current frame, as well as the 15 preceding frames and
2218 the 15 subsequent frames. Using a larger window results in a stronger
2219 smoothing effect and thus in less gain variation, i.e. slower gain
2220 adaptation. Conversely, using a smaller window results in a weaker smoothing
2221 effect and thus in more gain variation, i.e. faster gain adaptation.
2222 In other words, the more you increase this value, the more the Dynamic Audio
2223 Normalizer will behave like a "traditional" normalization filter. On the
2224 contrary, the more you decrease this value, the more the Dynamic Audio
2225 Normalizer will behave like a dynamic range compressor.
2228 Set the target peak value. This specifies the highest permissible magnitude
2229 level for the normalized audio input. This filter will try to approach the
2230 target peak magnitude as closely as possible, but at the same time it also
2231 makes sure that the normalized signal will never exceed the peak magnitude.
2232 A frame's maximum local gain factor is imposed directly by the target peak
2233 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
2234 It is not recommended to go above this value.
2237 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
2238 The Dynamic Audio Normalizer determines the maximum possible (local) gain
2239 factor for each input frame, i.e. the maximum gain factor that does not
2240 result in clipping or distortion. The maximum gain factor is determined by
2241 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
2242 additionally bounds the frame's maximum gain factor by a predetermined
2243 (global) maximum gain factor. This is done in order to avoid excessive gain
2244 factors in "silent" or almost silent frames. By default, the maximum gain
2245 factor is 10.0, For most inputs the default value should be sufficient and
2246 it usually is not recommended to increase this value. Though, for input
2247 with an extremely low overall volume level, it may be necessary to allow even
2248 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
2249 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
2250 Instead, a "sigmoid" threshold function will be applied. This way, the
2251 gain factors will smoothly approach the threshold value, but never exceed that
2255 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
2256 By default, the Dynamic Audio Normalizer performs "peak" normalization.
2257 This means that the maximum local gain factor for each frame is defined
2258 (only) by the frame's highest magnitude sample. This way, the samples can
2259 be amplified as much as possible without exceeding the maximum signal
2260 level, i.e. without clipping. Optionally, however, the Dynamic Audio
2261 Normalizer can also take into account the frame's root mean square,
2262 abbreviated RMS. In electrical engineering, the RMS is commonly used to
2263 determine the power of a time-varying signal. It is therefore considered
2264 that the RMS is a better approximation of the "perceived loudness" than
2265 just looking at the signal's peak magnitude. Consequently, by adjusting all
2266 frames to a constant RMS value, a uniform "perceived loudness" can be
2267 established. If a target RMS value has been specified, a frame's local gain
2268 factor is defined as the factor that would result in exactly that RMS value.
2269 Note, however, that the maximum local gain factor is still restricted by the
2270 frame's highest magnitude sample, in order to prevent clipping.
2273 Enable channels coupling. By default is enabled.
2274 By default, the Dynamic Audio Normalizer will amplify all channels by the same
2275 amount. This means the same gain factor will be applied to all channels, i.e.
2276 the maximum possible gain factor is determined by the "loudest" channel.
2277 However, in some recordings, it may happen that the volume of the different
2278 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
2279 In this case, this option can be used to disable the channel coupling. This way,
2280 the gain factor will be determined independently for each channel, depending
2281 only on the individual channel's highest magnitude sample. This allows for
2282 harmonizing the volume of the different channels.
2285 Enable DC bias correction. By default is disabled.
2286 An audio signal (in the time domain) is a sequence of sample values.
2287 In the Dynamic Audio Normalizer these sample values are represented in the
2288 -1.0 to 1.0 range, regardless of the original input format. Normally, the
2289 audio signal, or "waveform", should be centered around the zero point.
2290 That means if we calculate the mean value of all samples in a file, or in a
2291 single frame, then the result should be 0.0 or at least very close to that
2292 value. If, however, there is a significant deviation of the mean value from
2293 0.0, in either positive or negative direction, this is referred to as a
2294 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
2295 Audio Normalizer provides optional DC bias correction.
2296 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
2297 the mean value, or "DC correction" offset, of each input frame and subtract
2298 that value from all of the frame's sample values which ensures those samples
2299 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
2300 boundaries, the DC correction offset values will be interpolated smoothly
2301 between neighbouring frames.
2304 Enable alternative boundary mode. By default is disabled.
2305 The Dynamic Audio Normalizer takes into account a certain neighbourhood
2306 around each frame. This includes the preceding frames as well as the
2307 subsequent frames. However, for the "boundary" frames, located at the very
2308 beginning and at the very end of the audio file, not all neighbouring
2309 frames are available. In particular, for the first few frames in the audio
2310 file, the preceding frames are not known. And, similarly, for the last few
2311 frames in the audio file, the subsequent frames are not known. Thus, the
2312 question arises which gain factors should be assumed for the missing frames
2313 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
2314 to deal with this situation. The default boundary mode assumes a gain factor
2315 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
2316 "fade out" at the beginning and at the end of the input, respectively.
2319 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
2320 By default, the Dynamic Audio Normalizer does not apply "traditional"
2321 compression. This means that signal peaks will not be pruned and thus the
2322 full dynamic range will be retained within each local neighbourhood. However,
2323 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
2324 normalization algorithm with a more "traditional" compression.
2325 For this purpose, the Dynamic Audio Normalizer provides an optional compression
2326 (thresholding) function. If (and only if) the compression feature is enabled,
2327 all input frames will be processed by a soft knee thresholding function prior
2328 to the actual normalization process. Put simply, the thresholding function is
2329 going to prune all samples whose magnitude exceeds a certain threshold value.
2330 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
2331 value. Instead, the threshold value will be adjusted for each individual
2333 In general, smaller parameters result in stronger compression, and vice versa.
2334 Values below 3.0 are not recommended, because audible distortion may appear.
2339 Make audio easier to listen to on headphones.
2341 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
2342 so that when listened to on headphones the stereo image is moved from
2343 inside your head (standard for headphones) to outside and in front of
2344 the listener (standard for speakers).
2350 Apply a two-pole peaking equalisation (EQ) filter. With this
2351 filter, the signal-level at and around a selected frequency can
2352 be increased or decreased, whilst (unlike bandpass and bandreject
2353 filters) that at all other frequencies is unchanged.
2355 In order to produce complex equalisation curves, this filter can
2356 be given several times, each with a different central frequency.
2358 The filter accepts the following options:
2362 Set the filter's central frequency in Hz.
2365 Set method to specify band-width of filter.
2378 Specify the band-width of a filter in width_type units.
2381 Set the required gain or attenuation in dB.
2382 Beware of clipping when using a positive gain.
2385 @subsection Examples
2388 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2390 equalizer=f=1000:width_type=h:width=200:g=-10
2394 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2396 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2400 @section extrastereo
2402 Linearly increases the difference between left and right channels which
2403 adds some sort of "live" effect to playback.
2405 The filter accepts the following options:
2409 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2410 (average of both channels), with 1.0 sound will be unchanged, with
2411 -1.0 left and right channels will be swapped.
2414 Enable clipping. By default is enabled.
2417 @section firequalizer
2418 Apply FIR Equalization using arbitrary frequency response.
2420 The filter accepts the following option:
2424 Set gain curve equation (in dB). The expression can contain variables:
2427 the evaluated frequency
2431 channel number, set to 0 when multichannels evaluation is disabled
2433 channel id, see libavutil/channel_layout.h, set to the first channel id when
2434 multichannels evaluation is disabled
2438 channel_layout, see libavutil/channel_layout.h
2443 @item gain_interpolate(f)
2444 interpolate gain on frequency f based on gain_entry
2446 This option is also available as command. Default is @code{gain_interpolate(f)}.
2449 Set gain entry for gain_interpolate function. The expression can
2453 store gain entry at frequency f with value g
2455 This option is also available as command.
2458 Set filter delay in seconds. Higher value means more accurate.
2459 Default is @code{0.01}.
2462 Set filter accuracy in Hz. Lower value means more accurate.
2463 Default is @code{5}.
2466 Set window function. Acceptable values are:
2469 rectangular window, useful when gain curve is already smooth
2471 hann window (default)
2477 3-terms continuous 1st derivative nuttall window
2479 minimum 3-terms discontinuous nuttall window
2481 4-terms continuous 1st derivative nuttall window
2483 minimum 4-terms discontinuous nuttall (blackman-nuttall) window
2485 blackman-harris window
2489 If enabled, use fixed number of audio samples. This improves speed when
2490 filtering with large delay. Default is disabled.
2493 Enable multichannels evaluation on gain. Default is disabled.
2496 Enable zero phase mode by substracting timestamp to compensate delay.
2497 Default is disabled.
2500 @subsection Examples
2505 firequalizer=gain='if(lt(f,1000), 0, -INF)'
2508 lowpass at 1000 Hz with gain_entry:
2510 firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
2513 custom equalization:
2515 firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
2518 higher delay with zero phase to compensate delay:
2520 firequalizer=delay=0.1:fixed=on:zero_phase=on
2523 lowpass on left channel, highpass on right channel:
2525 firequalizer=gain='if(eq(chid,1), gain_interpolate(f), if(eq(chid,2), gain_interpolate(1e6+f), 0))'
2526 :gain_entry='entry(1000, 0); entry(1001,-INF); entry(1e6+1000,0)':multi=on
2531 Apply a flanging effect to the audio.
2533 The filter accepts the following options:
2537 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2540 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2543 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2547 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2548 Default value is 71.
2551 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2554 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2555 Default value is @var{sinusoidal}.
2558 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2559 Default value is 25.
2562 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2563 Default is @var{linear}.
2568 Apply a high-pass filter with 3dB point frequency.
2569 The filter can be either single-pole, or double-pole (the default).
2570 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2572 The filter accepts the following options:
2576 Set frequency in Hz. Default is 3000.
2579 Set number of poles. Default is 2.
2582 Set method to specify band-width of filter.
2595 Specify the band-width of a filter in width_type units.
2596 Applies only to double-pole filter.
2597 The default is 0.707q and gives a Butterworth response.
2602 Join multiple input streams into one multi-channel stream.
2604 It accepts the following parameters:
2608 The number of input streams. It defaults to 2.
2610 @item channel_layout
2611 The desired output channel layout. It defaults to stereo.
2614 Map channels from inputs to output. The argument is a '|'-separated list of
2615 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2616 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2617 can be either the name of the input channel (e.g. FL for front left) or its
2618 index in the specified input stream. @var{out_channel} is the name of the output
2622 The filter will attempt to guess the mappings when they are not specified
2623 explicitly. It does so by first trying to find an unused matching input channel
2624 and if that fails it picks the first unused input channel.
2626 Join 3 inputs (with properly set channel layouts):
2628 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2631 Build a 5.1 output from 6 single-channel streams:
2633 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2634 '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'
2640 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2642 To enable compilation of this filter you need to configure FFmpeg with
2643 @code{--enable-ladspa}.
2647 Specifies the name of LADSPA plugin library to load. If the environment
2648 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2649 each one of the directories specified by the colon separated list in
2650 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2651 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2652 @file{/usr/lib/ladspa/}.
2655 Specifies the plugin within the library. Some libraries contain only
2656 one plugin, but others contain many of them. If this is not set filter
2657 will list all available plugins within the specified library.
2660 Set the '|' separated list of controls which are zero or more floating point
2661 values that determine the behavior of the loaded plugin (for example delay,
2663 Controls need to be defined using the following syntax:
2664 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2665 @var{valuei} is the value set on the @var{i}-th control.
2666 Alternatively they can be also defined using the following syntax:
2667 @var{value0}|@var{value1}|@var{value2}|..., where
2668 @var{valuei} is the value set on the @var{i}-th control.
2669 If @option{controls} is set to @code{help}, all available controls and
2670 their valid ranges are printed.
2672 @item sample_rate, s
2673 Specify the sample rate, default to 44100. Only used if plugin have
2677 Set the number of samples per channel per each output frame, default
2678 is 1024. Only used if plugin have zero inputs.
2681 Set the minimum duration of the sourced audio. See
2682 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2683 for the accepted syntax.
2684 Note that the resulting duration may be greater than the specified duration,
2685 as the generated audio is always cut at the end of a complete frame.
2686 If not specified, or the expressed duration is negative, the audio is
2687 supposed to be generated forever.
2688 Only used if plugin have zero inputs.
2692 @subsection Examples
2696 List all available plugins within amp (LADSPA example plugin) library:
2702 List all available controls and their valid ranges for @code{vcf_notch}
2703 plugin from @code{VCF} library:
2705 ladspa=f=vcf:p=vcf_notch:c=help
2709 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2712 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2716 Add reverberation to the audio using TAP-plugins
2717 (Tom's Audio Processing plugins):
2719 ladspa=file=tap_reverb:tap_reverb
2723 Generate white noise, with 0.2 amplitude:
2725 ladspa=file=cmt:noise_source_white:c=c0=.2
2729 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2730 @code{C* Audio Plugin Suite} (CAPS) library:
2732 ladspa=file=caps:Click:c=c1=20'
2736 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2738 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2742 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2743 @code{SWH Plugins} collection:
2745 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2749 Attenuate low frequencies using Multiband EQ from Steve Harris
2750 @code{SWH Plugins} collection:
2752 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2756 @subsection Commands
2758 This filter supports the following commands:
2761 Modify the @var{N}-th control value.
2763 If the specified value is not valid, it is ignored and prior one is kept.
2768 EBU R128 loudness normalization. Includes both dynamic and linear normalization modes.
2769 Support for both single pass (livestreams, files) and double pass (files) modes.
2770 This algorithm can target IL, LRA, and maximum true peak.
2772 To enable compilation of this filter you need to configure FFmpeg with
2773 @code{--enable-libebur128}.
2775 The filter accepts the following options:
2779 Set integrated loudness target.
2780 Range is -70.0 - -5.0. Default value is -24.0.
2783 Set loudness range target.
2784 Range is 1.0 - 20.0. Default value is 7.0.
2787 Set maximum true peak.
2788 Range is -9.0 - +0.0. Default value is -2.0.
2790 @item measured_I, measured_i
2791 Measured IL of input file.
2792 Range is -99.0 - +0.0.
2794 @item measured_LRA, measured_lra
2795 Measured LRA of input file.
2796 Range is 0.0 - 99.0.
2798 @item measured_TP, measured_tp
2799 Measured true peak of input file.
2800 Range is -99.0 - +99.0.
2802 @item measured_thresh
2803 Measured threshold of input file.
2804 Range is -99.0 - +0.0.
2807 Set offset gain. Gain is applied before the true-peak limiter.
2808 Range is -99.0 - +99.0. Default is +0.0.
2811 Normalize linearly if possible.
2812 measured_I, measured_LRA, measured_TP, and measured_thresh must also
2813 to be specified in order to use this mode.
2814 Options are true or false. Default is true.
2817 Treat mono input files as "dual-mono". If a mono file is intended for playback
2818 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
2819 If set to @code{true}, this option will compensate for this effect.
2820 Multi-channel input files are not affected by this option.
2821 Options are true or false. Default is false.
2824 Set print format for stats. Options are summary, json, or none.
2825 Default value is none.
2830 Apply a low-pass filter with 3dB point frequency.
2831 The filter can be either single-pole or double-pole (the default).
2832 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2834 The filter accepts the following options:
2838 Set frequency in Hz. Default is 500.
2841 Set number of poles. Default is 2.
2844 Set method to specify band-width of filter.
2857 Specify the band-width of a filter in width_type units.
2858 Applies only to double-pole filter.
2859 The default is 0.707q and gives a Butterworth response.
2865 Mix channels with specific gain levels. The filter accepts the output
2866 channel layout followed by a set of channels definitions.
2868 This filter is also designed to efficiently remap the channels of an audio
2871 The filter accepts parameters of the form:
2872 "@var{l}|@var{outdef}|@var{outdef}|..."
2876 output channel layout or number of channels
2879 output channel specification, of the form:
2880 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
2883 output channel to define, either a channel name (FL, FR, etc.) or a channel
2884 number (c0, c1, etc.)
2887 multiplicative coefficient for the channel, 1 leaving the volume unchanged
2890 input channel to use, see out_name for details; it is not possible to mix
2891 named and numbered input channels
2894 If the `=' in a channel specification is replaced by `<', then the gains for
2895 that specification will be renormalized so that the total is 1, thus
2896 avoiding clipping noise.
2898 @subsection Mixing examples
2900 For example, if you want to down-mix from stereo to mono, but with a bigger
2901 factor for the left channel:
2903 pan=1c|c0=0.9*c0+0.1*c1
2906 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
2907 7-channels surround:
2909 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
2912 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
2913 that should be preferred (see "-ac" option) unless you have very specific
2916 @subsection Remapping examples
2918 The channel remapping will be effective if, and only if:
2921 @item gain coefficients are zeroes or ones,
2922 @item only one input per channel output,
2925 If all these conditions are satisfied, the filter will notify the user ("Pure
2926 channel mapping detected"), and use an optimized and lossless method to do the
2929 For example, if you have a 5.1 source and want a stereo audio stream by
2930 dropping the extra channels:
2932 pan="stereo| c0=FL | c1=FR"
2935 Given the same source, you can also switch front left and front right channels
2936 and keep the input channel layout:
2938 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
2941 If the input is a stereo audio stream, you can mute the front left channel (and
2942 still keep the stereo channel layout) with:
2947 Still with a stereo audio stream input, you can copy the right channel in both
2948 front left and right:
2950 pan="stereo| c0=FR | c1=FR"
2955 ReplayGain scanner filter. This filter takes an audio stream as an input and
2956 outputs it unchanged.
2957 At end of filtering it displays @code{track_gain} and @code{track_peak}.
2961 Convert the audio sample format, sample rate and channel layout. It is
2962 not meant to be used directly.
2965 Apply time-stretching and pitch-shifting with librubberband.
2967 The filter accepts the following options:
2971 Set tempo scale factor.
2974 Set pitch scale factor.
2977 Set transients detector.
2978 Possible values are:
2987 Possible values are:
2996 Possible values are:
3003 Set processing window size.
3004 Possible values are:
3013 Possible values are:
3020 Enable formant preservation when shift pitching.
3021 Possible values are:
3029 Possible values are:
3038 Possible values are:
3045 @section sidechaincompress
3047 This filter acts like normal compressor but has the ability to compress
3048 detected signal using second input signal.
3049 It needs two input streams and returns one output stream.
3050 First input stream will be processed depending on second stream signal.
3051 The filtered signal then can be filtered with other filters in later stages of
3052 processing. See @ref{pan} and @ref{amerge} filter.
3054 The filter accepts the following options:
3058 Set input gain. Default is 1. Range is between 0.015625 and 64.
3061 If a signal of second stream raises above this level it will affect the gain
3062 reduction of first stream.
3063 By default is 0.125. Range is between 0.00097563 and 1.
3066 Set a ratio about which the signal is reduced. 1:2 means that if the level
3067 raised 4dB above the threshold, it will be only 2dB above after the reduction.
3068 Default is 2. Range is between 1 and 20.
3071 Amount of milliseconds the signal has to rise above the threshold before gain
3072 reduction starts. Default is 20. Range is between 0.01 and 2000.
3075 Amount of milliseconds the signal has to fall below the threshold before
3076 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
3079 Set the amount by how much signal will be amplified after processing.
3080 Default is 2. Range is from 1 and 64.
3083 Curve the sharp knee around the threshold to enter gain reduction more softly.
3084 Default is 2.82843. Range is between 1 and 8.
3087 Choose if the @code{average} level between all channels of side-chain stream
3088 or the louder(@code{maximum}) channel of side-chain stream affects the
3089 reduction. Default is @code{average}.
3092 Should the exact signal be taken in case of @code{peak} or an RMS one in case
3093 of @code{rms}. Default is @code{rms} which is mainly smoother.
3096 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
3099 How much to use compressed signal in output. Default is 1.
3100 Range is between 0 and 1.
3103 @subsection Examples
3107 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
3108 depending on the signal of 2nd input and later compressed signal to be
3109 merged with 2nd input:
3111 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3115 @section sidechaingate
3117 A sidechain gate acts like a normal (wideband) gate but has the ability to
3118 filter the detected signal before sending it to the gain reduction stage.
3119 Normally a gate uses the full range signal to detect a level above the
3121 For example: If you cut all lower frequencies from your sidechain signal
3122 the gate will decrease the volume of your track only if not enough highs
3123 appear. With this technique you are able to reduce the resonation of a
3124 natural drum or remove "rumbling" of muted strokes from a heavily distorted
3126 It needs two input streams and returns one output stream.
3127 First input stream will be processed depending on second stream signal.
3129 The filter accepts the following options:
3133 Set input level before filtering.
3134 Default is 1. Allowed range is from 0.015625 to 64.
3137 Set the level of gain reduction when the signal is below the threshold.
3138 Default is 0.06125. Allowed range is from 0 to 1.
3141 If a signal rises above this level the gain reduction is released.
3142 Default is 0.125. Allowed range is from 0 to 1.
3145 Set a ratio about which the signal is reduced.
3146 Default is 2. Allowed range is from 1 to 9000.
3149 Amount of milliseconds the signal has to rise above the threshold before gain
3151 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
3154 Amount of milliseconds the signal has to fall below the threshold before the
3155 reduction is increased again. Default is 250 milliseconds.
3156 Allowed range is from 0.01 to 9000.
3159 Set amount of amplification of signal after processing.
3160 Default is 1. Allowed range is from 1 to 64.
3163 Curve the sharp knee around the threshold to enter gain reduction more softly.
3164 Default is 2.828427125. Allowed range is from 1 to 8.
3167 Choose if exact signal should be taken for detection or an RMS like one.
3168 Default is rms. Can be peak or rms.
3171 Choose if the average level between all channels or the louder channel affects
3173 Default is average. Can be average or maximum.
3176 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3179 @section silencedetect
3181 Detect silence in an audio stream.
3183 This filter logs a message when it detects that the input audio volume is less
3184 or equal to a noise tolerance value for a duration greater or equal to the
3185 minimum detected noise duration.
3187 The printed times and duration are expressed in seconds.
3189 The filter accepts the following options:
3193 Set silence duration until notification (default is 2 seconds).
3196 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
3197 specified value) or amplitude ratio. Default is -60dB, or 0.001.
3200 @subsection Examples
3204 Detect 5 seconds of silence with -50dB noise tolerance:
3206 silencedetect=n=-50dB:d=5
3210 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3211 tolerance in @file{silence.mp3}:
3213 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3217 @section silenceremove
3219 Remove silence from the beginning, middle or end of the audio.
3221 The filter accepts the following options:
3225 This value is used to indicate if audio should be trimmed at beginning of
3226 the audio. A value of zero indicates no silence should be trimmed from the
3227 beginning. When specifying a non-zero value, it trims audio up until it
3228 finds non-silence. Normally, when trimming silence from beginning of audio
3229 the @var{start_periods} will be @code{1} but it can be increased to higher
3230 values to trim all audio up to specific count of non-silence periods.
3231 Default value is @code{0}.
3233 @item start_duration
3234 Specify the amount of time that non-silence must be detected before it stops
3235 trimming audio. By increasing the duration, bursts of noises can be treated
3236 as silence and trimmed off. Default value is @code{0}.
3238 @item start_threshold
3239 This indicates what sample value should be treated as silence. For digital
3240 audio, a value of @code{0} may be fine but for audio recorded from analog,
3241 you may wish to increase the value to account for background noise.
3242 Can be specified in dB (in case "dB" is appended to the specified value)
3243 or amplitude ratio. Default value is @code{0}.
3246 Set the count for trimming silence from the end of audio.
3247 To remove silence from the middle of a file, specify a @var{stop_periods}
3248 that is negative. This value is then treated as a positive value and is
3249 used to indicate the effect should restart processing as specified by
3250 @var{start_periods}, making it suitable for removing periods of silence
3251 in the middle of the audio.
3252 Default value is @code{0}.
3255 Specify a duration of silence that must exist before audio is not copied any
3256 more. By specifying a higher duration, silence that is wanted can be left in
3258 Default value is @code{0}.
3260 @item stop_threshold
3261 This is the same as @option{start_threshold} but for trimming silence from
3263 Can be specified in dB (in case "dB" is appended to the specified value)
3264 or amplitude ratio. Default value is @code{0}.
3267 This indicate that @var{stop_duration} length of audio should be left intact
3268 at the beginning of each period of silence.
3269 For example, if you want to remove long pauses between words but do not want
3270 to remove the pauses completely. Default value is @code{0}.
3273 Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
3274 and works better with digital silence which is exactly 0.
3275 Default value is @code{rms}.
3278 Set ratio used to calculate size of window for detecting silence.
3279 Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
3282 @subsection Examples
3286 The following example shows how this filter can be used to start a recording
3287 that does not contain the delay at the start which usually occurs between
3288 pressing the record button and the start of the performance:
3290 silenceremove=1:5:0.02
3294 Trim all silence encountered from beginning to end where there is more than 1
3295 second of silence in audio:
3297 silenceremove=0:0:0:-1:1:-90dB
3303 SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3304 loudspeakers around the user for binaural listening via headphones (audio
3305 formats up to 9 channels supported).
3306 The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3307 SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3308 Austrian Academy of Sciences.
3310 To enable compilation of this filter you need to configure FFmpeg with
3311 @code{--enable-netcdf}.
3313 The filter accepts the following options:
3317 Set the SOFA file used for rendering.
3320 Set gain applied to audio. Value is in dB. Default is 0.
3323 Set rotation of virtual loudspeakers in deg. Default is 0.
3326 Set elevation of virtual speakers in deg. Default is 0.
3329 Set distance in meters between loudspeakers and the listener with near-field
3330 HRTFs. Default is 1.
3333 Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3334 processing audio in time domain which is slow.
3335 @var{freq} is processing audio in frequency domain which is fast.
3336 Default is @var{freq}.
3339 Set custom positions of virtual loudspeakers. Syntax for this option is:
3340 <CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...].
3341 Each virtual loudspeaker is described with short channel name following with
3342 azimuth and elevation in degreees.
3343 Each virtual loudspeaker description is separated by '|'.
3344 For example to override front left and front right channel positions use:
3345 'speakers=FL 45 15|FR 345 15'.
3346 Descriptions with unrecognised channel names are ignored.
3349 @subsection Examples
3353 Using ClubFritz6 sofa file:
3355 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3359 Using ClubFritz12 sofa file and bigger radius with small rotation:
3361 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
3365 Similar as above but with custom speaker positions for front left, front right, rear left and rear right
3366 and also with custom gain:
3368 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|RL 135|RR 225:gain=28"
3372 @section stereotools
3374 This filter has some handy utilities to manage stereo signals, for converting
3375 M/S stereo recordings to L/R signal while having control over the parameters
3376 or spreading the stereo image of master track.
3378 The filter accepts the following options:
3382 Set input level before filtering for both channels. Defaults is 1.
3383 Allowed range is from 0.015625 to 64.
3386 Set output level after filtering for both channels. Defaults is 1.
3387 Allowed range is from 0.015625 to 64.
3390 Set input balance between both channels. Default is 0.
3391 Allowed range is from -1 to 1.
3394 Set output balance between both channels. Default is 0.
3395 Allowed range is from -1 to 1.
3398 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3399 clipping. Disabled by default.
3402 Mute the left channel. Disabled by default.
3405 Mute the right channel. Disabled by default.
3408 Change the phase of the left channel. Disabled by default.
3411 Change the phase of the right channel. Disabled by default.
3414 Set stereo mode. Available values are:
3418 Left/Right to Left/Right, this is default.
3421 Left/Right to Mid/Side.
3424 Mid/Side to Left/Right.
3427 Left/Right to Left/Left.
3430 Left/Right to Right/Right.
3433 Left/Right to Left + Right.
3436 Left/Right to Right/Left.
3440 Set level of side signal. Default is 1.
3441 Allowed range is from 0.015625 to 64.
3444 Set balance of side signal. Default is 0.
3445 Allowed range is from -1 to 1.
3448 Set level of the middle signal. Default is 1.
3449 Allowed range is from 0.015625 to 64.
3452 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3455 Set stereo base between mono and inversed channels. Default is 0.
3456 Allowed range is from -1 to 1.
3459 Set delay in milliseconds how much to delay left from right channel and
3460 vice versa. Default is 0. Allowed range is from -20 to 20.
3463 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3466 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3469 @subsection Examples
3473 Apply karaoke like effect:
3475 stereotools=mlev=0.015625
3479 Convert M/S signal to L/R:
3481 "stereotools=mode=ms>lr"
3485 @section stereowiden
3487 This filter enhance the stereo effect by suppressing signal common to both
3488 channels and by delaying the signal of left into right and vice versa,
3489 thereby widening the stereo effect.
3491 The filter accepts the following options:
3495 Time in milliseconds of the delay of left signal into right and vice versa.
3496 Default is 20 milliseconds.
3499 Amount of gain in delayed signal into right and vice versa. Gives a delay
3500 effect of left signal in right output and vice versa which gives widening
3501 effect. Default is 0.3.
3504 Cross feed of left into right with inverted phase. This helps in suppressing
3505 the mono. If the value is 1 it will cancel all the signal common to both
3506 channels. Default is 0.3.
3509 Set level of input signal of original channel. Default is 0.8.
3514 Boost or cut treble (upper) frequencies of the audio using a two-pole
3515 shelving filter with a response similar to that of a standard
3516 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3518 The filter accepts the following options:
3522 Give the gain at whichever is the lower of ~22 kHz and the
3523 Nyquist frequency. Its useful range is about -20 (for a large cut)
3524 to +20 (for a large boost). Beware of clipping when using a positive gain.
3527 Set the filter's central frequency and so can be used
3528 to extend or reduce the frequency range to be boosted or cut.
3529 The default value is @code{3000} Hz.
3532 Set method to specify band-width of filter.
3545 Determine how steep is the filter's shelf transition.
3550 Sinusoidal amplitude modulation.
3552 The filter accepts the following options:
3556 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3557 (20 Hz or lower) will result in a tremolo effect.
3558 This filter may also be used as a ring modulator by specifying
3559 a modulation frequency higher than 20 Hz.
3560 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3563 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3564 Default value is 0.5.
3569 Sinusoidal phase modulation.
3571 The filter accepts the following options:
3575 Modulation frequency in Hertz.
3576 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3579 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3580 Default value is 0.5.
3585 Adjust the input audio volume.
3587 It accepts the following parameters:
3591 Set audio volume expression.
3593 Output values are clipped to the maximum value.
3595 The output audio volume is given by the relation:
3597 @var{output_volume} = @var{volume} * @var{input_volume}
3600 The default value for @var{volume} is "1.0".
3603 This parameter represents the mathematical precision.
3605 It determines which input sample formats will be allowed, which affects the
3606 precision of the volume scaling.
3610 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3612 32-bit floating-point; this limits input sample format to FLT. (default)
3614 64-bit floating-point; this limits input sample format to DBL.
3618 Choose the behaviour on encountering ReplayGain side data in input frames.
3622 Remove ReplayGain side data, ignoring its contents (the default).
3625 Ignore ReplayGain side data, but leave it in the frame.
3628 Prefer the track gain, if present.
3631 Prefer the album gain, if present.
3634 @item replaygain_preamp
3635 Pre-amplification gain in dB to apply to the selected replaygain gain.
3637 Default value for @var{replaygain_preamp} is 0.0.
3640 Set when the volume expression is evaluated.
3642 It accepts the following values:
3645 only evaluate expression once during the filter initialization, or
3646 when the @samp{volume} command is sent
3649 evaluate expression for each incoming frame
3652 Default value is @samp{once}.
3655 The volume expression can contain the following parameters.
3659 frame number (starting at zero)
3662 @item nb_consumed_samples
3663 number of samples consumed by the filter
3665 number of samples in the current frame
3667 original frame position in the file
3673 PTS at start of stream
3675 time at start of stream
3681 last set volume value
3684 Note that when @option{eval} is set to @samp{once} only the
3685 @var{sample_rate} and @var{tb} variables are available, all other
3686 variables will evaluate to NAN.
3688 @subsection Commands
3690 This filter supports the following commands:
3693 Modify the volume expression.
3694 The command accepts the same syntax of the corresponding option.
3696 If the specified expression is not valid, it is kept at its current
3698 @item replaygain_noclip
3699 Prevent clipping by limiting the gain applied.
3701 Default value for @var{replaygain_noclip} is 1.
3705 @subsection Examples
3709 Halve the input audio volume:
3713 volume=volume=-6.0206dB
3716 In all the above example the named key for @option{volume} can be
3717 omitted, for example like in:
3723 Increase input audio power by 6 decibels using fixed-point precision:
3725 volume=volume=6dB:precision=fixed
3729 Fade volume after time 10 with an annihilation period of 5 seconds:
3731 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3735 @section volumedetect
3737 Detect the volume of the input video.
3739 The filter has no parameters. The input is not modified. Statistics about
3740 the volume will be printed in the log when the input stream end is reached.
3742 In particular it will show the mean volume (root mean square), maximum
3743 volume (on a per-sample basis), and the beginning of a histogram of the
3744 registered volume values (from the maximum value to a cumulated 1/1000 of
3747 All volumes are in decibels relative to the maximum PCM value.
3749 @subsection Examples
3751 Here is an excerpt of the output:
3753 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3754 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3755 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3756 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3757 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3758 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3759 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3760 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3761 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3767 The mean square energy is approximately -27 dB, or 10^-2.7.
3769 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3771 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3774 In other words, raising the volume by +4 dB does not cause any clipping,
3775 raising it by +5 dB causes clipping for 6 samples, etc.
3777 @c man end AUDIO FILTERS
3779 @chapter Audio Sources
3780 @c man begin AUDIO SOURCES
3782 Below is a description of the currently available audio sources.
3786 Buffer audio frames, and make them available to the filter chain.
3788 This source is mainly intended for a programmatic use, in particular
3789 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3791 It accepts the following parameters:
3795 The timebase which will be used for timestamps of submitted frames. It must be
3796 either a floating-point number or in @var{numerator}/@var{denominator} form.
3799 The sample rate of the incoming audio buffers.
3802 The sample format of the incoming audio buffers.
3803 Either a sample format name or its corresponding integer representation from
3804 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3806 @item channel_layout
3807 The channel layout of the incoming audio buffers.
3808 Either a channel layout name from channel_layout_map in
3809 @file{libavutil/channel_layout.c} or its corresponding integer representation
3810 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3813 The number of channels of the incoming audio buffers.
3814 If both @var{channels} and @var{channel_layout} are specified, then they
3819 @subsection Examples
3822 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3825 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3826 Since the sample format with name "s16p" corresponds to the number
3827 6 and the "stereo" channel layout corresponds to the value 0x3, this is
3830 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3835 Generate an audio signal specified by an expression.
3837 This source accepts in input one or more expressions (one for each
3838 channel), which are evaluated and used to generate a corresponding
3841 This source accepts the following options:
3845 Set the '|'-separated expressions list for each separate channel. In case the
3846 @option{channel_layout} option is not specified, the selected channel layout
3847 depends on the number of provided expressions. Otherwise the last
3848 specified expression is applied to the remaining output channels.
3850 @item channel_layout, c
3851 Set the channel layout. The number of channels in the specified layout
3852 must be equal to the number of specified expressions.
3855 Set the minimum duration of the sourced audio. See
3856 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
3857 for the accepted syntax.
3858 Note that the resulting duration may be greater than the specified
3859 duration, as the generated audio is always cut at the end of a
3862 If not specified, or the expressed duration is negative, the audio is
3863 supposed to be generated forever.
3866 Set the number of samples per channel per each output frame,
3869 @item sample_rate, s
3870 Specify the sample rate, default to 44100.
3873 Each expression in @var{exprs} can contain the following constants:
3877 number of the evaluated sample, starting from 0
3880 time of the evaluated sample expressed in seconds, starting from 0
3887 @subsection Examples
3897 Generate a sin signal with frequency of 440 Hz, set sample rate to
3900 aevalsrc="sin(440*2*PI*t):s=8000"
3904 Generate a two channels signal, specify the channel layout (Front
3905 Center + Back Center) explicitly:
3907 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
3911 Generate white noise:
3913 aevalsrc="-2+random(0)"
3917 Generate an amplitude modulated signal:
3919 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
3923 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
3925 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
3932 The null audio source, return unprocessed audio frames. It is mainly useful
3933 as a template and to be employed in analysis / debugging tools, or as
3934 the source for filters which ignore the input data (for example the sox
3937 This source accepts the following options:
3941 @item channel_layout, cl
3943 Specifies the channel layout, and can be either an integer or a string
3944 representing a channel layout. The default value of @var{channel_layout}
3947 Check the channel_layout_map definition in
3948 @file{libavutil/channel_layout.c} for the mapping between strings and
3949 channel layout values.
3951 @item sample_rate, r
3952 Specifies the sample rate, and defaults to 44100.
3955 Set the number of samples per requested frames.
3959 @subsection Examples
3963 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
3965 anullsrc=r=48000:cl=4
3969 Do the same operation with a more obvious syntax:
3971 anullsrc=r=48000:cl=mono
3975 All the parameters need to be explicitly defined.
3979 Synthesize a voice utterance using the libflite library.
3981 To enable compilation of this filter you need to configure FFmpeg with
3982 @code{--enable-libflite}.
3984 Note that the flite library is not thread-safe.
3986 The filter accepts the following options:
3991 If set to 1, list the names of the available voices and exit
3992 immediately. Default value is 0.
3995 Set the maximum number of samples per frame. Default value is 512.
3998 Set the filename containing the text to speak.
4001 Set the text to speak.
4004 Set the voice to use for the speech synthesis. Default value is
4005 @code{kal}. See also the @var{list_voices} option.
4008 @subsection Examples
4012 Read from file @file{speech.txt}, and synthesize the text using the
4013 standard flite voice:
4015 flite=textfile=speech.txt
4019 Read the specified text selecting the @code{slt} voice:
4021 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4025 Input text to ffmpeg:
4027 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4031 Make @file{ffplay} speak the specified text, using @code{flite} and
4032 the @code{lavfi} device:
4034 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
4038 For more information about libflite, check:
4039 @url{http://www.speech.cs.cmu.edu/flite/}
4043 Generate a noise audio signal.
4045 The filter accepts the following options:
4048 @item sample_rate, r
4049 Specify the sample rate. Default value is 48000 Hz.
4052 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
4056 Specify the duration of the generated audio stream. Not specifying this option
4057 results in noise with an infinite length.
4059 @item color, colour, c
4060 Specify the color of noise. Available noise colors are white, pink, and brown.
4061 Default color is white.
4064 Specify a value used to seed the PRNG.
4067 Set the number of samples per each output frame, default is 1024.
4070 @subsection Examples
4075 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
4077 anoisesrc=d=60:c=pink:r=44100:a=0.5
4083 Generate an audio signal made of a sine wave with amplitude 1/8.
4085 The audio signal is bit-exact.
4087 The filter accepts the following options:
4092 Set the carrier frequency. Default is 440 Hz.
4094 @item beep_factor, b
4095 Enable a periodic beep every second with frequency @var{beep_factor} times
4096 the carrier frequency. Default is 0, meaning the beep is disabled.
4098 @item sample_rate, r
4099 Specify the sample rate, default is 44100.
4102 Specify the duration of the generated audio stream.
4104 @item samples_per_frame
4105 Set the number of samples per output frame.
4107 The expression can contain the following constants:
4111 The (sequential) number of the output audio frame, starting from 0.
4114 The PTS (Presentation TimeStamp) of the output audio frame,
4115 expressed in @var{TB} units.
4118 The PTS of the output audio frame, expressed in seconds.
4121 The timebase of the output audio frames.
4124 Default is @code{1024}.
4127 @subsection Examples
4132 Generate a simple 440 Hz sine wave:
4138 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4142 sine=frequency=220:beep_factor=4:duration=5
4146 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4149 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4153 @c man end AUDIO SOURCES
4155 @chapter Audio Sinks
4156 @c man begin AUDIO SINKS
4158 Below is a description of the currently available audio sinks.
4160 @section abuffersink
4162 Buffer audio frames, and make them available to the end of filter chain.
4164 This sink is mainly intended for programmatic use, in particular
4165 through the interface defined in @file{libavfilter/buffersink.h}
4166 or the options system.
4168 It accepts a pointer to an AVABufferSinkContext structure, which
4169 defines the incoming buffers' formats, to be passed as the opaque
4170 parameter to @code{avfilter_init_filter} for initialization.
4173 Null audio sink; do absolutely nothing with the input audio. It is
4174 mainly useful as a template and for use in analysis / debugging
4177 @c man end AUDIO SINKS
4179 @chapter Video Filters
4180 @c man begin VIDEO FILTERS
4182 When you configure your FFmpeg build, you can disable any of the
4183 existing filters using @code{--disable-filters}.
4184 The configure output will show the video filters included in your
4187 Below is a description of the currently available video filters.
4189 @section alphaextract
4191 Extract the alpha component from the input as a grayscale video. This
4192 is especially useful with the @var{alphamerge} filter.
4196 Add or replace the alpha component of the primary input with the
4197 grayscale value of a second input. This is intended for use with
4198 @var{alphaextract} to allow the transmission or storage of frame
4199 sequences that have alpha in a format that doesn't support an alpha
4202 For example, to reconstruct full frames from a normal YUV-encoded video
4203 and a separate video created with @var{alphaextract}, you might use:
4205 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
4208 Since this filter is designed for reconstruction, it operates on frame
4209 sequences without considering timestamps, and terminates when either
4210 input reaches end of stream. This will cause problems if your encoding
4211 pipeline drops frames. If you're trying to apply an image as an
4212 overlay to a video stream, consider the @var{overlay} filter instead.
4216 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
4217 and libavformat to work. On the other hand, it is limited to ASS (Advanced
4218 Substation Alpha) subtitles files.
4220 This filter accepts the following option in addition to the common options from
4221 the @ref{subtitles} filter:
4225 Set the shaping engine
4227 Available values are:
4230 The default libass shaping engine, which is the best available.
4232 Fast, font-agnostic shaper that can do only substitutions
4234 Slower shaper using OpenType for substitutions and positioning
4237 The default is @code{auto}.
4241 Apply an Adaptive Temporal Averaging Denoiser to the video input.
4243 The filter accepts the following options:
4247 Set threshold A for 1st plane. Default is 0.02.
4248 Valid range is 0 to 0.3.
4251 Set threshold B for 1st plane. Default is 0.04.
4252 Valid range is 0 to 5.
4255 Set threshold A for 2nd plane. Default is 0.02.
4256 Valid range is 0 to 0.3.
4259 Set threshold B for 2nd plane. Default is 0.04.
4260 Valid range is 0 to 5.
4263 Set threshold A for 3rd plane. Default is 0.02.
4264 Valid range is 0 to 0.3.
4267 Set threshold B for 3rd plane. Default is 0.04.
4268 Valid range is 0 to 5.
4270 Threshold A is designed to react on abrupt changes in the input signal and
4271 threshold B is designed to react on continuous changes in the input signal.
4274 Set number of frames filter will use for averaging. Default is 33. Must be odd
4275 number in range [5, 129].
4280 Compute the bounding box for the non-black pixels in the input frame
4283 This filter computes the bounding box containing all the pixels with a
4284 luminance value greater than the minimum allowed value.
4285 The parameters describing the bounding box are printed on the filter
4288 The filter accepts the following option:
4292 Set the minimal luminance value. Default is @code{16}.
4295 @section blackdetect
4297 Detect video intervals that are (almost) completely black. Can be
4298 useful to detect chapter transitions, commercials, or invalid
4299 recordings. Output lines contains the time for the start, end and
4300 duration of the detected black interval expressed in seconds.
4302 In order to display the output lines, you need to set the loglevel at
4303 least to the AV_LOG_INFO value.
4305 The filter accepts the following options:
4308 @item black_min_duration, d
4309 Set the minimum detected black duration expressed in seconds. It must
4310 be a non-negative floating point number.
4312 Default value is 2.0.
4314 @item picture_black_ratio_th, pic_th
4315 Set the threshold for considering a picture "black".
4316 Express the minimum value for the ratio:
4318 @var{nb_black_pixels} / @var{nb_pixels}
4321 for which a picture is considered black.
4322 Default value is 0.98.
4324 @item pixel_black_th, pix_th
4325 Set the threshold for considering a pixel "black".
4327 The threshold expresses the maximum pixel luminance value for which a
4328 pixel is considered "black". The provided value is scaled according to
4329 the following equation:
4331 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
4334 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
4335 the input video format, the range is [0-255] for YUV full-range
4336 formats and [16-235] for YUV non full-range formats.
4338 Default value is 0.10.
4341 The following example sets the maximum pixel threshold to the minimum
4342 value, and detects only black intervals of 2 or more seconds:
4344 blackdetect=d=2:pix_th=0.00
4349 Detect frames that are (almost) completely black. Can be useful to
4350 detect chapter transitions or commercials. Output lines consist of
4351 the frame number of the detected frame, the percentage of blackness,
4352 the position in the file if known or -1 and the timestamp in seconds.
4354 In order to display the output lines, you need to set the loglevel at
4355 least to the AV_LOG_INFO value.
4357 It accepts the following parameters:
4362 The percentage of the pixels that have to be below the threshold; it defaults to
4365 @item threshold, thresh
4366 The threshold below which a pixel value is considered black; it defaults to
4371 @section blend, tblend
4373 Blend two video frames into each other.
4375 The @code{blend} filter takes two input streams and outputs one
4376 stream, the first input is the "top" layer and second input is
4377 "bottom" layer. Output terminates when shortest input terminates.
4379 The @code{tblend} (time blend) filter takes two consecutive frames
4380 from one single stream, and outputs the result obtained by blending
4381 the new frame on top of the old frame.
4383 A description of the accepted options follows.
4391 Set blend mode for specific pixel component or all pixel components in case
4392 of @var{all_mode}. Default value is @code{normal}.
4394 Available values for component modes are:
4435 Set blend opacity for specific pixel component or all pixel components in case
4436 of @var{all_opacity}. Only used in combination with pixel component blend modes.
4443 Set blend expression for specific pixel component or all pixel components in case
4444 of @var{all_expr}. Note that related mode options will be ignored if those are set.
4446 The expressions can use the following variables:
4450 The sequential number of the filtered frame, starting from @code{0}.
4454 the coordinates of the current sample
4458 the width and height of currently filtered plane
4462 Width and height scale depending on the currently filtered plane. It is the
4463 ratio between the corresponding luma plane number of pixels and the current
4464 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4465 @code{0.5,0.5} for chroma planes.
4468 Time of the current frame, expressed in seconds.
4471 Value of pixel component at current location for first video frame (top layer).
4474 Value of pixel component at current location for second video frame (bottom layer).
4478 Force termination when the shortest input terminates. Default is
4479 @code{0}. This option is only defined for the @code{blend} filter.
4482 Continue applying the last bottom frame after the end of the stream. A value of
4483 @code{0} disable the filter after the last frame of the bottom layer is reached.
4484 Default is @code{1}. This option is only defined for the @code{blend} filter.
4487 @subsection Examples
4491 Apply transition from bottom layer to top layer in first 10 seconds:
4493 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4497 Apply 1x1 checkerboard effect:
4499 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4503 Apply uncover left effect:
4505 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4509 Apply uncover down effect:
4511 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4515 Apply uncover up-left effect:
4517 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4521 Split diagonally video and shows top and bottom layer on each side:
4523 blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4527 Display differences between the current and the previous frame:
4529 tblend=all_mode=difference128
4535 Apply a boxblur algorithm to the input video.
4537 It accepts the following parameters:
4541 @item luma_radius, lr
4542 @item luma_power, lp
4543 @item chroma_radius, cr
4544 @item chroma_power, cp
4545 @item alpha_radius, ar
4546 @item alpha_power, ap
4550 A description of the accepted options follows.
4553 @item luma_radius, lr
4554 @item chroma_radius, cr
4555 @item alpha_radius, ar
4556 Set an expression for the box radius in pixels used for blurring the
4557 corresponding input plane.
4559 The radius value must be a non-negative number, and must not be
4560 greater than the value of the expression @code{min(w,h)/2} for the
4561 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4564 Default value for @option{luma_radius} is "2". If not specified,
4565 @option{chroma_radius} and @option{alpha_radius} default to the
4566 corresponding value set for @option{luma_radius}.
4568 The expressions can contain the following constants:
4572 The input width and height in pixels.
4576 The input chroma image width and height in pixels.
4580 The horizontal and vertical chroma subsample values. For example, for the
4581 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4584 @item luma_power, lp
4585 @item chroma_power, cp
4586 @item alpha_power, ap
4587 Specify how many times the boxblur filter is applied to the
4588 corresponding plane.
4590 Default value for @option{luma_power} is 2. If not specified,
4591 @option{chroma_power} and @option{alpha_power} default to the
4592 corresponding value set for @option{luma_power}.
4594 A value of 0 will disable the effect.
4597 @subsection Examples
4601 Apply a boxblur filter with the luma, chroma, and alpha radii
4604 boxblur=luma_radius=2:luma_power=1
4609 Set the luma radius to 2, and alpha and chroma radius to 0:
4611 boxblur=2:1:cr=0:ar=0
4615 Set the luma and chroma radii to a fraction of the video dimension:
4617 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4623 Deinterlace the input video ("bwdif" stands for "Bob Weaver
4624 Deinterlacing Filter").
4626 Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4627 interpolation algorithms.
4628 It accepts the following parameters:
4632 The interlacing mode to adopt. It accepts one of the following values:
4636 Output one frame for each frame.
4638 Output one frame for each field.
4641 The default value is @code{send_field}.
4644 The picture field parity assumed for the input interlaced video. It accepts one
4645 of the following values:
4649 Assume the top field is first.
4651 Assume the bottom field is first.
4653 Enable automatic detection of field parity.
4656 The default value is @code{auto}.
4657 If the interlacing is unknown or the decoder does not export this information,
4658 top field first will be assumed.
4661 Specify which frames to deinterlace. Accept one of the following
4666 Deinterlace all frames.
4668 Only deinterlace frames marked as interlaced.
4671 The default value is @code{all}.
4675 YUV colorspace color/chroma keying.
4677 The filter accepts the following options:
4681 The color which will be replaced with transparency.
4684 Similarity percentage with the key color.
4686 0.01 matches only the exact key color, while 1.0 matches everything.
4691 0.0 makes pixels either fully transparent, or not transparent at all.
4693 Higher values result in semi-transparent pixels, with a higher transparency
4694 the more similar the pixels color is to the key color.
4697 Signals that the color passed is already in YUV instead of RGB.
4699 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4700 This can be used to pass exact YUV values as hexadecimal numbers.
4703 @subsection Examples
4707 Make every green pixel in the input image transparent:
4709 ffmpeg -i input.png -vf chromakey=green out.png
4713 Overlay a greenscreen-video on top of a static black background.
4715 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
4721 Display CIE color diagram with pixels overlaid onto it.
4723 The filter accepts the following options:
4738 @item uhdtv, rec2020
4751 Set what gamuts to draw.
4753 See @code{system} option for available values.
4756 Set ciescope size, by default set to 512.
4759 Set intensity used to map input pixel values to CIE diagram.
4762 Set contrast used to draw tongue colors that are out of active color system gamut.
4765 Correct gamma displayed on scope, by default enabled.
4768 Show white point on CIE diagram, by default disabled.
4771 Set input gamma. Used only with XYZ input color space.
4776 Visualize information exported by some codecs.
4778 Some codecs can export information through frames using side-data or other
4779 means. For example, some MPEG based codecs export motion vectors through the
4780 @var{export_mvs} flag in the codec @option{flags2} option.
4782 The filter accepts the following option:
4786 Set motion vectors to visualize.
4788 Available flags for @var{mv} are:
4792 forward predicted MVs of P-frames
4794 forward predicted MVs of B-frames
4796 backward predicted MVs of B-frames
4800 Display quantization parameters using the chroma planes.
4803 Set motion vectors type to visualize. Includes MVs from all frames unless specified by @var{frame_type} option.
4805 Available flags for @var{mv_type} are:
4809 forward predicted MVs
4811 backward predicted MVs
4814 @item frame_type, ft
4815 Set frame type to visualize motion vectors of.
4817 Available flags for @var{frame_type} are:
4821 intra-coded frames (I-frames)
4823 predicted frames (P-frames)
4825 bi-directionally predicted frames (B-frames)
4829 @subsection Examples
4833 Visualize forward predicted MVs of all frames using @command{ffplay}:
4835 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
4839 Visualize multi-directionals MVs of P and B-Frames using @command{ffplay}:
4841 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
4845 @section colorbalance
4846 Modify intensity of primary colors (red, green and blue) of input frames.
4848 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
4849 regions for the red-cyan, green-magenta or blue-yellow balance.
4851 A positive adjustment value shifts the balance towards the primary color, a negative
4852 value towards the complementary color.
4854 The filter accepts the following options:
4860 Adjust red, green and blue shadows (darkest pixels).
4865 Adjust red, green and blue midtones (medium pixels).
4870 Adjust red, green and blue highlights (brightest pixels).
4872 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
4875 @subsection Examples
4879 Add red color cast to shadows:
4886 RGB colorspace color keying.
4888 The filter accepts the following options:
4892 The color which will be replaced with transparency.
4895 Similarity percentage with the key color.
4897 0.01 matches only the exact key color, while 1.0 matches everything.
4902 0.0 makes pixels either fully transparent, or not transparent at all.
4904 Higher values result in semi-transparent pixels, with a higher transparency
4905 the more similar the pixels color is to the key color.
4908 @subsection Examples
4912 Make every green pixel in the input image transparent:
4914 ffmpeg -i input.png -vf colorkey=green out.png
4918 Overlay a greenscreen-video on top of a static background image.
4920 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
4924 @section colorlevels
4926 Adjust video input frames using levels.
4928 The filter accepts the following options:
4935 Adjust red, green, blue and alpha input black point.
4936 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
4942 Adjust red, green, blue and alpha input white point.
4943 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
4945 Input levels are used to lighten highlights (bright tones), darken shadows
4946 (dark tones), change the balance of bright and dark tones.
4952 Adjust red, green, blue and alpha output black point.
4953 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
4959 Adjust red, green, blue and alpha output white point.
4960 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
4962 Output levels allows manual selection of a constrained output level range.
4965 @subsection Examples
4969 Make video output darker:
4971 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
4977 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
4981 Make video output lighter:
4983 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
4987 Increase brightness:
4989 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
4993 @section colorchannelmixer
4995 Adjust video input frames by re-mixing color channels.
4997 This filter modifies a color channel by adding the values associated to
4998 the other channels of the same pixels. For example if the value to
4999 modify is red, the output value will be:
5001 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
5004 The filter accepts the following options:
5011 Adjust contribution of input red, green, blue and alpha channels for output red channel.
5012 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
5018 Adjust contribution of input red, green, blue and alpha channels for output green channel.
5019 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
5025 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
5026 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
5032 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
5033 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
5035 Allowed ranges for options are @code{[-2.0, 2.0]}.
5038 @subsection Examples
5042 Convert source to grayscale:
5044 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
5047 Simulate sepia tones:
5049 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
5053 @section colormatrix
5055 Convert color matrix.
5057 The filter accepts the following options:
5062 Specify the source and destination color matrix. Both values must be
5065 The accepted values are:
5084 For example to convert from BT.601 to SMPTE-240M, use the command:
5086 colormatrix=bt601:smpte240m
5091 Convert colorspace, transfer characteristics or color primaries.
5093 The filter accepts the following options:
5097 Specify all color properties at once.
5099 The accepted values are:
5128 Specify output colorspace.
5130 The accepted values are:
5139 BT.470BG or BT.601-6 625
5142 SMPTE-170M or BT.601-6 525
5148 BT.2020 with non-constant luminance
5153 Specify output transfer characteristics.
5155 The accepted values are:
5161 Constant gamma of 2.2
5164 Constant gamma of 2.8
5167 SMPTE-170M, BT.601-6 625 or BT.601-6 525
5173 BT.2020 for 10-bits content
5176 BT.2020 for 12-bits content
5181 Specify output color primaries.
5183 The accepted values are:
5192 BT.470BG or BT.601-6 625
5195 SMPTE-170M or BT.601-6 525
5206 Specify output color range.
5208 The accepted values are:
5211 MPEG (restricted) range
5219 Specify output color format.
5221 The accepted values are:
5224 YUV 4:2:0 planar 8-bits
5227 YUV 4:2:0 planar 10-bits
5230 YUV 4:2:0 planar 12-bits
5233 YUV 4:2:2 planar 8-bits
5236 YUV 4:2:2 planar 10-bits
5239 YUV 4:2:2 planar 12-bits
5242 YUV 4:4:4 planar 8-bits
5245 YUV 4:4:4 planar 10-bits
5248 YUV 4:4:4 planar 12-bits
5253 Do a fast conversion, which skips gamma/primary correction. This will take
5254 significantly less CPU, but will be mathematically incorrect. To get output
5255 compatible with that produced by the colormatrix filter, use fast=1.
5258 Specify dithering mode.
5260 The accepted values are:
5266 Floyd-Steinberg dithering
5270 Whitepoint adaptation mode.
5272 The accepted values are:
5275 Bradford whitepoint adaptation
5278 von Kries whitepoint adaptation
5281 identity whitepoint adaptation (i.e. no whitepoint adaptation)
5286 The filter converts the transfer characteristics, color space and color
5287 primaries to the specified user values. The output value, if not specified,
5288 is set to a default value based on the "all" property. If that property is
5289 also not specified, the filter will log an error. The output color range and
5290 format default to the same value as the input color range and format. The
5291 input transfer characteristics, color space, color primaries and color range
5292 should be set on the input data. If any of these are missing, the filter will
5293 log an error and no conversion will take place.
5295 For example to convert the input to SMPTE-240M, use the command:
5297 colorspace=smpte240m
5300 @section convolution
5302 Apply convolution 3x3 or 5x5 filter.
5304 The filter accepts the following options:
5311 Set matrix for each plane.
5312 Matrix is sequence of 9 or 25 signed integers.
5318 Set multiplier for calculated value for each plane.
5324 Set bias for each plane. This value is added to the result of the multiplication.
5325 Useful for making the overall image brighter or darker. Default is 0.0.
5328 @subsection Examples
5334 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"
5340 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"
5346 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"
5352 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"
5358 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"
5364 Copy the input source unchanged to the output. This is mainly useful for
5369 Video filtering on GPU using Apple's CoreImage API on OSX.
5371 Hardware acceleration is based on an OpenGL context. Usually, this means it is
5372 processed by video hardware. However, software-based OpenGL implementations
5373 exist which means there is no guarantee for hardware processing. It depends on
5376 There are many filters and image generators provided by Apple that come with a
5377 large variety of options. The filter has to be referenced by its name along
5380 The coreimage filter accepts the following options:
5383 List all available filters and generators along with all their respective
5384 options as well as possible minimum and maximum values along with the default
5391 Specify all filters by their respective name and options.
5392 Use @var{list_filters} to determine all valid filter names and options.
5393 Numerical options are specified by a float value and are automatically clamped
5394 to their respective value range. Vector and color options have to be specified
5395 by a list of space separated float values. Character escaping has to be done.
5396 A special option name @code{default} is available to use default options for a
5399 It is required to specify either @code{default} or at least one of the filter options.
5400 All omitted options are used with their default values.
5401 The syntax of the filter string is as follows:
5403 filter=<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...][#<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...]][#...]
5407 Specify a rectangle where the output of the filter chain is copied into the
5408 input image. It is given by a list of space separated float values:
5410 output_rect=x\ y\ width\ height
5412 If not given, the output rectangle equals the dimensions of the input image.
5413 The output rectangle is automatically cropped at the borders of the input
5414 image. Negative values are valid for each component.
5416 output_rect=25\ 25\ 100\ 100
5420 Several filters can be chained for successive processing without GPU-HOST
5421 transfers allowing for fast processing of complex filter chains.
5422 Currently, only filters with zero (generators) or exactly one (filters) input
5423 image and one output image are supported. Also, transition filters are not yet
5426 Some filters generate output images with additional padding depending on the
5427 respective filter kernel. The padding is automatically removed to ensure the
5428 filter output has the same size as the input image.
5430 For image generators, the size of the output image is determined by the
5431 previous output image of the filter chain or the input image of the whole
5432 filterchain, respectively. The generators do not use the pixel information of
5433 this image to generate their output. However, the generated output is
5434 blended onto this image, resulting in partial or complete coverage of the
5437 The @ref{coreimagesrc} video source can be used for generating input images
5438 which are directly fed into the filter chain. By using it, providing input
5439 images by another video source or an input video is not required.
5441 @subsection Examples
5446 List all filters available:
5448 coreimage=list_filters=true
5452 Use the CIBoxBlur filter with default options to blur an image:
5454 coreimage=filter=CIBoxBlur@@default
5458 Use a filter chain with CISepiaTone at default values and CIVignetteEffect with
5459 its center at 100x100 and a radius of 50 pixels:
5461 coreimage=filter=CIBoxBlur@@default#CIVignetteEffect@@inputCenter=100\ 100@@inputRadius=50
5465 Use nullsrc and CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
5466 given as complete and escaped command-line for Apple's standard bash shell:
5468 ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
5474 Crop the input video to given dimensions.
5476 It accepts the following parameters:
5480 The width of the output video. It defaults to @code{iw}.
5481 This expression is evaluated only once during the filter
5482 configuration, or when the @samp{w} or @samp{out_w} command is sent.
5485 The height of the output video. It defaults to @code{ih}.
5486 This expression is evaluated only once during the filter
5487 configuration, or when the @samp{h} or @samp{out_h} command is sent.
5490 The horizontal position, in the input video, of the left edge of the output
5491 video. It defaults to @code{(in_w-out_w)/2}.
5492 This expression is evaluated per-frame.
5495 The vertical position, in the input video, of the top edge of the output video.
5496 It defaults to @code{(in_h-out_h)/2}.
5497 This expression is evaluated per-frame.
5500 If set to 1 will force the output display aspect ratio
5501 to be the same of the input, by changing the output sample aspect
5502 ratio. It defaults to 0.
5505 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
5506 expressions containing the following constants:
5511 The computed values for @var{x} and @var{y}. They are evaluated for
5516 The input width and height.
5520 These are the same as @var{in_w} and @var{in_h}.
5524 The output (cropped) width and height.
5528 These are the same as @var{out_w} and @var{out_h}.
5531 same as @var{iw} / @var{ih}
5534 input sample aspect ratio
5537 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
5541 horizontal and vertical chroma subsample values. For example for the
5542 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5545 The number of the input frame, starting from 0.
5548 the position in the file of the input frame, NAN if unknown
5551 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
5555 The expression for @var{out_w} may depend on the value of @var{out_h},
5556 and the expression for @var{out_h} may depend on @var{out_w}, but they
5557 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
5558 evaluated after @var{out_w} and @var{out_h}.
5560 The @var{x} and @var{y} parameters specify the expressions for the
5561 position of the top-left corner of the output (non-cropped) area. They
5562 are evaluated for each frame. If the evaluated value is not valid, it
5563 is approximated to the nearest valid value.
5565 The expression for @var{x} may depend on @var{y}, and the expression
5566 for @var{y} may depend on @var{x}.
5568 @subsection Examples
5572 Crop area with size 100x100 at position (12,34).
5577 Using named options, the example above becomes:
5579 crop=w=100:h=100:x=12:y=34
5583 Crop the central input area with size 100x100:
5589 Crop the central input area with size 2/3 of the input video:
5591 crop=2/3*in_w:2/3*in_h
5595 Crop the input video central square:
5602 Delimit the rectangle with the top-left corner placed at position
5603 100:100 and the right-bottom corner corresponding to the right-bottom
5604 corner of the input image.
5606 crop=in_w-100:in_h-100:100:100
5610 Crop 10 pixels from the left and right borders, and 20 pixels from
5611 the top and bottom borders
5613 crop=in_w-2*10:in_h-2*20
5617 Keep only the bottom right quarter of the input image:
5619 crop=in_w/2:in_h/2:in_w/2:in_h/2
5623 Crop height for getting Greek harmony:
5625 crop=in_w:1/PHI*in_w
5629 Apply trembling effect:
5631 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)
5635 Apply erratic camera effect depending on timestamp:
5637 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)"
5641 Set x depending on the value of y:
5643 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
5647 @subsection Commands
5649 This filter supports the following commands:
5655 Set width/height of the output video and the horizontal/vertical position
5657 The command accepts the same syntax of the corresponding option.
5659 If the specified expression is not valid, it is kept at its current
5665 Auto-detect the crop size.
5667 It calculates the necessary cropping parameters and prints the
5668 recommended parameters via the logging system. The detected dimensions
5669 correspond to the non-black area of the input video.
5671 It accepts the following parameters:
5676 Set higher black value threshold, which can be optionally specified
5677 from nothing (0) to everything (255 for 8-bit based formats). An intensity
5678 value greater to the set value is considered non-black. It defaults to 24.
5679 You can also specify a value between 0.0 and 1.0 which will be scaled depending
5680 on the bitdepth of the pixel format.
5683 The value which the width/height should be divisible by. It defaults to
5684 16. The offset is automatically adjusted to center the video. Use 2 to
5685 get only even dimensions (needed for 4:2:2 video). 16 is best when
5686 encoding to most video codecs.
5688 @item reset_count, reset
5689 Set the counter that determines after how many frames cropdetect will
5690 reset the previously detected largest video area and start over to
5691 detect the current optimal crop area. Default value is 0.
5693 This can be useful when channel logos distort the video area. 0
5694 indicates 'never reset', and returns the largest area encountered during
5701 Apply color adjustments using curves.
5703 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
5704 component (red, green and blue) has its values defined by @var{N} key points
5705 tied from each other using a smooth curve. The x-axis represents the pixel
5706 values from the input frame, and the y-axis the new pixel values to be set for
5709 By default, a component curve is defined by the two points @var{(0;0)} and
5710 @var{(1;1)}. This creates a straight line where each original pixel value is
5711 "adjusted" to its own value, which means no change to the image.
5713 The filter allows you to redefine these two points and add some more. A new
5714 curve (using a natural cubic spline interpolation) will be define to pass
5715 smoothly through all these new coordinates. The new defined points needs to be
5716 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
5717 be in the @var{[0;1]} interval. If the computed curves happened to go outside
5718 the vector spaces, the values will be clipped accordingly.
5720 The filter accepts the following options:
5724 Select one of the available color presets. This option can be used in addition
5725 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
5726 options takes priority on the preset values.
5727 Available presets are:
5730 @item color_negative
5733 @item increase_contrast
5735 @item linear_contrast
5736 @item medium_contrast
5738 @item strong_contrast
5741 Default is @code{none}.
5743 Set the master key points. These points will define a second pass mapping. It
5744 is sometimes called a "luminance" or "value" mapping. It can be used with
5745 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
5746 post-processing LUT.
5748 Set the key points for the red component.
5750 Set the key points for the green component.
5752 Set the key points for the blue component.
5754 Set the key points for all components (not including master).
5755 Can be used in addition to the other key points component
5756 options. In this case, the unset component(s) will fallback on this
5757 @option{all} setting.
5759 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
5761 Save Gnuplot script of the curves in specified file.
5764 To avoid some filtergraph syntax conflicts, each key points list need to be
5765 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
5767 @subsection Examples
5771 Increase slightly the middle level of blue:
5773 curves=blue='0/0 0.5/0.58 1/1'
5779 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'
5781 Here we obtain the following coordinates for each components:
5784 @code{(0;0.11) (0.42;0.51) (1;0.95)}
5786 @code{(0;0) (0.50;0.48) (1;1)}
5788 @code{(0;0.22) (0.49;0.44) (1;0.80)}
5792 The previous example can also be achieved with the associated built-in preset:
5794 curves=preset=vintage
5804 Use a Photoshop preset and redefine the points of the green component:
5806 curves=psfile='MyCurvesPresets/purple.acv':green='0/0 0.45/0.53 1/1'
5810 Check out the curves of the @code{cross_process} profile using @command{ffmpeg}
5811 and @command{gnuplot}:
5813 ffmpeg -f lavfi -i color -vf curves=cross_process:plot=/tmp/curves.plt -frames:v 1 -f null -
5814 gnuplot -p /tmp/curves.plt
5820 Video data analysis filter.
5822 This filter shows hexadecimal pixel values of part of video.
5824 The filter accepts the following options:
5828 Set output video size.
5831 Set x offset from where to pick pixels.
5834 Set y offset from where to pick pixels.
5837 Set scope mode, can be one of the following:
5840 Draw hexadecimal pixel values with white color on black background.
5843 Draw hexadecimal pixel values with input video pixel color on black
5847 Draw hexadecimal pixel values on color background picked from input video,
5848 the text color is picked in such way so its always visible.
5852 Draw rows and columns numbers on left and top of video.
5857 Denoise frames using 2D DCT (frequency domain filtering).
5859 This filter is not designed for real time.
5861 The filter accepts the following options:
5865 Set the noise sigma constant.
5867 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
5868 coefficient (absolute value) below this threshold with be dropped.
5870 If you need a more advanced filtering, see @option{expr}.
5872 Default is @code{0}.
5875 Set number overlapping pixels for each block. Since the filter can be slow, you
5876 may want to reduce this value, at the cost of a less effective filter and the
5877 risk of various artefacts.
5879 If the overlapping value doesn't permit processing the whole input width or
5880 height, a warning will be displayed and according borders won't be denoised.
5882 Default value is @var{blocksize}-1, which is the best possible setting.
5885 Set the coefficient factor expression.
5887 For each coefficient of a DCT block, this expression will be evaluated as a
5888 multiplier value for the coefficient.
5890 If this is option is set, the @option{sigma} option will be ignored.
5892 The absolute value of the coefficient can be accessed through the @var{c}
5896 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
5897 @var{blocksize}, which is the width and height of the processed blocks.
5899 The default value is @var{3} (8x8) and can be raised to @var{4} for a
5900 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
5901 on the speed processing. Also, a larger block size does not necessarily means a
5905 @subsection Examples
5907 Apply a denoise with a @option{sigma} of @code{4.5}:
5912 The same operation can be achieved using the expression system:
5914 dctdnoiz=e='gte(c, 4.5*3)'
5917 Violent denoise using a block size of @code{16x16}:
5924 Remove banding artifacts from input video.
5925 It works by replacing banded pixels with average value of referenced pixels.
5927 The filter accepts the following options:
5934 Set banding detection threshold for each plane. Default is 0.02.
5935 Valid range is 0.00003 to 0.5.
5936 If difference between current pixel and reference pixel is less than threshold,
5937 it will be considered as banded.
5940 Banding detection range in pixels. Default is 16. If positive, random number
5941 in range 0 to set value will be used. If negative, exact absolute value
5943 The range defines square of four pixels around current pixel.
5946 Set direction in radians from which four pixel will be compared. If positive,
5947 random direction from 0 to set direction will be picked. If negative, exact of
5948 absolute value will be picked. For example direction 0, -PI or -2*PI radians
5949 will pick only pixels on same row and -PI/2 will pick only pixels on same
5953 If enabled, current pixel is compared with average value of all four
5954 surrounding pixels. The default is enabled. If disabled current pixel is
5955 compared with all four surrounding pixels. The pixel is considered banded
5956 if only all four differences with surrounding pixels are less than threshold.
5962 Drop duplicated frames at regular intervals.
5964 The filter accepts the following options:
5968 Set the number of frames from which one will be dropped. Setting this to
5969 @var{N} means one frame in every batch of @var{N} frames will be dropped.
5970 Default is @code{5}.
5973 Set the threshold for duplicate detection. If the difference metric for a frame
5974 is less than or equal to this value, then it is declared as duplicate. Default
5978 Set scene change threshold. Default is @code{15}.
5982 Set the size of the x and y-axis blocks used during metric calculations.
5983 Larger blocks give better noise suppression, but also give worse detection of
5984 small movements. Must be a power of two. Default is @code{32}.
5987 Mark main input as a pre-processed input and activate clean source input
5988 stream. This allows the input to be pre-processed with various filters to help
5989 the metrics calculation while keeping the frame selection lossless. When set to
5990 @code{1}, the first stream is for the pre-processed input, and the second
5991 stream is the clean source from where the kept frames are chosen. Default is
5995 Set whether or not chroma is considered in the metric calculations. Default is
6001 Apply deflate effect to the video.
6003 This filter replaces the pixel by the local(3x3) average by taking into account
6004 only values lower than the pixel.
6006 It accepts the following options:
6013 Limit the maximum change for each plane, default is 65535.
6014 If 0, plane will remain unchanged.
6019 Remove judder produced by partially interlaced telecined content.
6021 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
6022 source was partially telecined content then the output of @code{pullup,dejudder}
6023 will have a variable frame rate. May change the recorded frame rate of the
6024 container. Aside from that change, this filter will not affect constant frame
6027 The option available in this filter is:
6031 Specify the length of the window over which the judder repeats.
6033 Accepts any integer greater than 1. Useful values are:
6037 If the original was telecined from 24 to 30 fps (Film to NTSC).
6040 If the original was telecined from 25 to 30 fps (PAL to NTSC).
6043 If a mixture of the two.
6046 The default is @samp{4}.
6051 Suppress a TV station logo by a simple interpolation of the surrounding
6052 pixels. Just set a rectangle covering the logo and watch it disappear
6053 (and sometimes something even uglier appear - your mileage may vary).
6055 It accepts the following parameters:
6060 Specify the top left corner coordinates of the logo. They must be
6065 Specify the width and height of the logo to clear. They must be
6069 Specify the thickness of the fuzzy edge of the rectangle (added to
6070 @var{w} and @var{h}). The default value is 1. This option is
6071 deprecated, setting higher values should no longer be necessary and
6075 When set to 1, a green rectangle is drawn on the screen to simplify
6076 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
6077 The default value is 0.
6079 The rectangle is drawn on the outermost pixels which will be (partly)
6080 replaced with interpolated values. The values of the next pixels
6081 immediately outside this rectangle in each direction will be used to
6082 compute the interpolated pixel values inside the rectangle.
6086 @subsection Examples
6090 Set a rectangle covering the area with top left corner coordinates 0,0
6091 and size 100x77, and a band of size 10:
6093 delogo=x=0:y=0:w=100:h=77:band=10
6100 Attempt to fix small changes in horizontal and/or vertical shift. This
6101 filter helps remove camera shake from hand-holding a camera, bumping a
6102 tripod, moving on a vehicle, etc.
6104 The filter accepts the following options:
6112 Specify a rectangular area where to limit the search for motion
6114 If desired the search for motion vectors can be limited to a
6115 rectangular area of the frame defined by its top left corner, width
6116 and height. These parameters have the same meaning as the drawbox
6117 filter which can be used to visualise the position of the bounding
6120 This is useful when simultaneous movement of subjects within the frame
6121 might be confused for camera motion by the motion vector search.
6123 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
6124 then the full frame is used. This allows later options to be set
6125 without specifying the bounding box for the motion vector search.
6127 Default - search the whole frame.
6131 Specify the maximum extent of movement in x and y directions in the
6132 range 0-64 pixels. Default 16.
6135 Specify how to generate pixels to fill blanks at the edge of the
6136 frame. Available values are:
6139 Fill zeroes at blank locations
6141 Original image at blank locations
6143 Extruded edge value at blank locations
6145 Mirrored edge at blank locations
6147 Default value is @samp{mirror}.
6150 Specify the blocksize to use for motion search. Range 4-128 pixels,
6154 Specify the contrast threshold for blocks. Only blocks with more than
6155 the specified contrast (difference between darkest and lightest
6156 pixels) will be considered. Range 1-255, default 125.
6159 Specify the search strategy. Available values are:
6162 Set exhaustive search
6164 Set less exhaustive search.
6166 Default value is @samp{exhaustive}.
6169 If set then a detailed log of the motion search is written to the
6173 If set to 1, specify using OpenCL capabilities, only available if
6174 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
6180 Apply an exact inverse of the telecine operation. It requires a predefined
6181 pattern specified using the pattern option which must be the same as that passed
6182 to the telecine filter.
6184 This filter accepts the following options:
6193 The default value is @code{top}.
6197 A string of numbers representing the pulldown pattern you wish to apply.
6198 The default value is @code{23}.
6201 A number representing position of the first frame with respect to the telecine
6202 pattern. This is to be used if the stream is cut. The default value is @code{0}.
6207 Apply dilation effect to the video.
6209 This filter replaces the pixel by the local(3x3) maximum.
6211 It accepts the following options:
6218 Limit the maximum change for each plane, default is 65535.
6219 If 0, plane will remain unchanged.
6222 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6225 Flags to local 3x3 coordinates maps like this:
6234 Displace pixels as indicated by second and third input stream.
6236 It takes three input streams and outputs one stream, the first input is the
6237 source, and second and third input are displacement maps.
6239 The second input specifies how much to displace pixels along the
6240 x-axis, while the third input specifies how much to displace pixels
6242 If one of displacement map streams terminates, last frame from that
6243 displacement map will be used.
6245 Note that once generated, displacements maps can be reused over and over again.
6247 A description of the accepted options follows.
6251 Set displace behavior for pixels that are out of range.
6253 Available values are:
6256 Missing pixels are replaced by black pixels.
6259 Adjacent pixels will spread out to replace missing pixels.
6262 Out of range pixels are wrapped so they point to pixels of other side.
6264 Default is @samp{smear}.
6268 @subsection Examples
6272 Add ripple effect to rgb input of video size hd720:
6274 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
6278 Add wave effect to rgb input of video size hd720:
6280 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
6286 Draw a colored box on the input image.
6288 It accepts the following parameters:
6293 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
6297 The expressions which specify the width and height of the box; if 0 they are interpreted as
6298 the input width and height. It defaults to 0.
6301 Specify the color of the box to write. For the general syntax of this option,
6302 check the "Color" section in the ffmpeg-utils manual. If the special
6303 value @code{invert} is used, the box edge color is the same as the
6304 video with inverted luma.
6307 The expression which sets the thickness of the box edge. Default value is @code{3}.
6309 See below for the list of accepted constants.
6312 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6313 following constants:
6317 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6321 horizontal and vertical chroma subsample values. For example for the
6322 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6326 The input width and height.
6329 The input sample aspect ratio.
6333 The x and y offset coordinates where the box is drawn.
6337 The width and height of the drawn box.
6340 The thickness of the drawn box.
6342 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6343 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6347 @subsection Examples
6351 Draw a black box around the edge of the input image:
6357 Draw a box with color red and an opacity of 50%:
6359 drawbox=10:20:200:60:red@@0.5
6362 The previous example can be specified as:
6364 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
6368 Fill the box with pink color:
6370 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
6374 Draw a 2-pixel red 2.40:1 mask:
6376 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
6382 Draw a grid on the input image.
6384 It accepts the following parameters:
6389 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
6393 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
6394 input width and height, respectively, minus @code{thickness}, so image gets
6395 framed. Default to 0.
6398 Specify the color of the grid. For the general syntax of this option,
6399 check the "Color" section in the ffmpeg-utils manual. If the special
6400 value @code{invert} is used, the grid color is the same as the
6401 video with inverted luma.
6404 The expression which sets the thickness of the grid line. Default value is @code{1}.
6406 See below for the list of accepted constants.
6409 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6410 following constants:
6414 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6418 horizontal and vertical chroma subsample values. For example for the
6419 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6423 The input grid cell width and height.
6426 The input sample aspect ratio.
6430 The x and y coordinates of some point of grid intersection (meant to configure offset).
6434 The width and height of the drawn cell.
6437 The thickness of the drawn cell.
6439 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6440 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6444 @subsection Examples
6448 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
6450 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
6454 Draw a white 3x3 grid with an opacity of 50%:
6456 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
6463 Draw a text string or text from a specified file on top of a video, using the
6464 libfreetype library.
6466 To enable compilation of this filter, you need to configure FFmpeg with
6467 @code{--enable-libfreetype}.
6468 To enable default font fallback and the @var{font} option you need to
6469 configure FFmpeg with @code{--enable-libfontconfig}.
6470 To enable the @var{text_shaping} option, you need to configure FFmpeg with
6471 @code{--enable-libfribidi}.
6475 It accepts the following parameters:
6480 Used to draw a box around text using the background color.
6481 The value must be either 1 (enable) or 0 (disable).
6482 The default value of @var{box} is 0.
6485 Set the width of the border to be drawn around the box using @var{boxcolor}.
6486 The default value of @var{boxborderw} is 0.
6489 The color to be used for drawing box around text. For the syntax of this
6490 option, check the "Color" section in the ffmpeg-utils manual.
6492 The default value of @var{boxcolor} is "white".
6495 Set the width of the border to be drawn around the text using @var{bordercolor}.
6496 The default value of @var{borderw} is 0.
6499 Set the color to be used for drawing border around text. For the syntax of this
6500 option, check the "Color" section in the ffmpeg-utils manual.
6502 The default value of @var{bordercolor} is "black".
6505 Select how the @var{text} is expanded. Can be either @code{none},
6506 @code{strftime} (deprecated) or
6507 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
6511 If true, check and fix text coords to avoid clipping.
6514 The color to be used for drawing fonts. For the syntax of this option, check
6515 the "Color" section in the ffmpeg-utils manual.
6517 The default value of @var{fontcolor} is "black".
6519 @item fontcolor_expr
6520 String which is expanded the same way as @var{text} to obtain dynamic
6521 @var{fontcolor} value. By default this option has empty value and is not
6522 processed. When this option is set, it overrides @var{fontcolor} option.
6525 The font family to be used for drawing text. By default Sans.
6528 The font file to be used for drawing text. The path must be included.
6529 This parameter is mandatory if the fontconfig support is disabled.
6532 This option does not exist, please see the timeline system
6535 Draw the text applying alpha blending. The value can
6536 be either a number between 0.0 and 1.0
6537 The expression accepts the same variables @var{x, y} do.
6538 The default value is 1.
6539 Please see fontcolor_expr
6542 The font size to be used for drawing text.
6543 The default value of @var{fontsize} is 16.
6546 If set to 1, attempt to shape the text (for example, reverse the order of
6547 right-to-left text and join Arabic characters) before drawing it.
6548 Otherwise, just draw the text exactly as given.
6549 By default 1 (if supported).
6552 The flags to be used for loading the fonts.
6554 The flags map the corresponding flags supported by libfreetype, and are
6555 a combination of the following values:
6562 @item vertical_layout
6563 @item force_autohint
6566 @item ignore_global_advance_width
6568 @item ignore_transform
6574 Default value is "default".
6576 For more information consult the documentation for the FT_LOAD_*
6580 The color to be used for drawing a shadow behind the drawn text. For the
6581 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
6583 The default value of @var{shadowcolor} is "black".
6587 The x and y offsets for the text shadow position with respect to the
6588 position of the text. They can be either positive or negative
6589 values. The default value for both is "0".
6592 The starting frame number for the n/frame_num variable. The default value
6596 The size in number of spaces to use for rendering the tab.
6600 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
6601 format. It can be used with or without text parameter. @var{timecode_rate}
6602 option must be specified.
6604 @item timecode_rate, rate, r
6605 Set the timecode frame rate (timecode only).
6608 The text string to be drawn. The text must be a sequence of UTF-8
6610 This parameter is mandatory if no file is specified with the parameter
6614 A text file containing text to be drawn. The text must be a sequence
6615 of UTF-8 encoded characters.
6617 This parameter is mandatory if no text string is specified with the
6618 parameter @var{text}.
6620 If both @var{text} and @var{textfile} are specified, an error is thrown.
6623 If set to 1, the @var{textfile} will be reloaded before each frame.
6624 Be sure to update it atomically, or it may be read partially, or even fail.
6628 The expressions which specify the offsets where text will be drawn
6629 within the video frame. They are relative to the top/left border of the
6632 The default value of @var{x} and @var{y} is "0".
6634 See below for the list of accepted constants and functions.
6637 The parameters for @var{x} and @var{y} are expressions containing the
6638 following constants and functions:
6642 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
6646 horizontal and vertical chroma subsample values. For example for the
6647 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6650 the height of each text line
6658 @item max_glyph_a, ascent
6659 the maximum distance from the baseline to the highest/upper grid
6660 coordinate used to place a glyph outline point, for all the rendered
6662 It is a positive value, due to the grid's orientation with the Y axis
6665 @item max_glyph_d, descent
6666 the maximum distance from the baseline to the lowest grid coordinate
6667 used to place a glyph outline point, for all the rendered glyphs.
6668 This is a negative value, due to the grid's orientation, with the Y axis
6672 maximum glyph height, that is the maximum height for all the glyphs
6673 contained in the rendered text, it is equivalent to @var{ascent} -
6677 maximum glyph width, that is the maximum width for all the glyphs
6678 contained in the rendered text
6681 the number of input frame, starting from 0
6683 @item rand(min, max)
6684 return a random number included between @var{min} and @var{max}
6687 The input sample aspect ratio.
6690 timestamp expressed in seconds, NAN if the input timestamp is unknown
6693 the height of the rendered text
6696 the width of the rendered text
6700 the x and y offset coordinates where the text is drawn.
6702 These parameters allow the @var{x} and @var{y} expressions to refer
6703 each other, so you can for example specify @code{y=x/dar}.
6706 @anchor{drawtext_expansion}
6707 @subsection Text expansion
6709 If @option{expansion} is set to @code{strftime},
6710 the filter recognizes strftime() sequences in the provided text and
6711 expands them accordingly. Check the documentation of strftime(). This
6712 feature is deprecated.
6714 If @option{expansion} is set to @code{none}, the text is printed verbatim.
6716 If @option{expansion} is set to @code{normal} (which is the default),
6717 the following expansion mechanism is used.
6719 The backslash character @samp{\}, followed by any character, always expands to
6720 the second character.
6722 Sequence of the form @code{%@{...@}} are expanded. The text between the
6723 braces is a function name, possibly followed by arguments separated by ':'.
6724 If the arguments contain special characters or delimiters (':' or '@}'),
6725 they should be escaped.
6727 Note that they probably must also be escaped as the value for the
6728 @option{text} option in the filter argument string and as the filter
6729 argument in the filtergraph description, and possibly also for the shell,
6730 that makes up to four levels of escaping; using a text file avoids these
6733 The following functions are available:
6738 The expression evaluation result.
6740 It must take one argument specifying the expression to be evaluated,
6741 which accepts the same constants and functions as the @var{x} and
6742 @var{y} values. Note that not all constants should be used, for
6743 example the text size is not known when evaluating the expression, so
6744 the constants @var{text_w} and @var{text_h} will have an undefined
6747 @item expr_int_format, eif
6748 Evaluate the expression's value and output as formatted integer.
6750 The first argument is the expression to be evaluated, just as for the @var{expr} function.
6751 The second argument specifies the output format. Allowed values are @samp{x},
6752 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
6753 @code{printf} function.
6754 The third parameter is optional and sets the number of positions taken by the output.
6755 It can be used to add padding with zeros from the left.
6758 The time at which the filter is running, expressed in UTC.
6759 It can accept an argument: a strftime() format string.
6762 The time at which the filter is running, expressed in the local time zone.
6763 It can accept an argument: a strftime() format string.
6766 Frame metadata. Takes one or two arguments.
6768 The first argument is mandatory and specifies the metadata key.
6770 The second argument is optional and specifies a default value, used when the
6771 metadata key is not found or empty.
6774 The frame number, starting from 0.
6777 A 1 character description of the current picture type.
6780 The timestamp of the current frame.
6781 It can take up to three arguments.
6783 The first argument is the format of the timestamp; it defaults to @code{flt}
6784 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
6785 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
6786 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
6787 @code{localtime} stands for the timestamp of the frame formatted as
6788 local time zone time.
6790 The second argument is an offset added to the timestamp.
6792 If the format is set to @code{localtime} or @code{gmtime},
6793 a third argument may be supplied: a strftime() format string.
6794 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
6797 @subsection Examples
6801 Draw "Test Text" with font FreeSerif, using the default values for the
6802 optional parameters.
6805 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
6809 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
6810 and y=50 (counting from the top-left corner of the screen), text is
6811 yellow with a red box around it. Both the text and the box have an
6815 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
6816 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
6819 Note that the double quotes are not necessary if spaces are not used
6820 within the parameter list.
6823 Show the text at the center of the video frame:
6825 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
6829 Show the text at a random position, switching to a new position every 30 seconds:
6831 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)"
6835 Show a text line sliding from right to left in the last row of the video
6836 frame. The file @file{LONG_LINE} is assumed to contain a single line
6839 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
6843 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
6845 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
6849 Draw a single green letter "g", at the center of the input video.
6850 The glyph baseline is placed at half screen height.
6852 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
6856 Show text for 1 second every 3 seconds:
6858 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
6862 Use fontconfig to set the font. Note that the colons need to be escaped.
6864 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
6868 Print the date of a real-time encoding (see strftime(3)):
6870 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
6874 Show text fading in and out (appearing/disappearing):
6877 DS=1.0 # display start
6878 DE=10.0 # display end
6879 FID=1.5 # fade in duration
6880 FOD=5 # fade out duration
6881 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 @}"
6886 For more information about libfreetype, check:
6887 @url{http://www.freetype.org/}.
6889 For more information about fontconfig, check:
6890 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
6892 For more information about libfribidi, check:
6893 @url{http://fribidi.org/}.
6897 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
6899 The filter accepts the following options:
6904 Set low and high threshold values used by the Canny thresholding
6907 The high threshold selects the "strong" edge pixels, which are then
6908 connected through 8-connectivity with the "weak" edge pixels selected
6909 by the low threshold.
6911 @var{low} and @var{high} threshold values must be chosen in the range
6912 [0,1], and @var{low} should be lesser or equal to @var{high}.
6914 Default value for @var{low} is @code{20/255}, and default value for @var{high}
6918 Define the drawing mode.
6922 Draw white/gray wires on black background.
6925 Mix the colors to create a paint/cartoon effect.
6928 Default value is @var{wires}.
6931 @subsection Examples
6935 Standard edge detection with custom values for the hysteresis thresholding:
6937 edgedetect=low=0.1:high=0.4
6941 Painting effect without thresholding:
6943 edgedetect=mode=colormix:high=0
6948 Set brightness, contrast, saturation and approximate gamma adjustment.
6950 The filter accepts the following options:
6954 Set the contrast expression. The value must be a float value in range
6955 @code{-2.0} to @code{2.0}. The default value is "1".
6958 Set the brightness expression. The value must be a float value in
6959 range @code{-1.0} to @code{1.0}. The default value is "0".
6962 Set the saturation expression. The value must be a float in
6963 range @code{0.0} to @code{3.0}. The default value is "1".
6966 Set the gamma expression. The value must be a float in range
6967 @code{0.1} to @code{10.0}. The default value is "1".
6970 Set the gamma expression for red. The value must be a float in
6971 range @code{0.1} to @code{10.0}. The default value is "1".
6974 Set the gamma expression for green. The value must be a float in range
6975 @code{0.1} to @code{10.0}. The default value is "1".
6978 Set the gamma expression for blue. The value must be a float in range
6979 @code{0.1} to @code{10.0}. The default value is "1".
6982 Set the gamma weight expression. It can be used to reduce the effect
6983 of a high gamma value on bright image areas, e.g. keep them from
6984 getting overamplified and just plain white. The value must be a float
6985 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
6986 gamma correction all the way down while @code{1.0} leaves it at its
6987 full strength. Default is "1".
6990 Set when the expressions for brightness, contrast, saturation and
6991 gamma expressions are evaluated.
6993 It accepts the following values:
6996 only evaluate expressions once during the filter initialization or
6997 when a command is processed
7000 evaluate expressions for each incoming frame
7003 Default value is @samp{init}.
7006 The expressions accept the following parameters:
7009 frame count of the input frame starting from 0
7012 byte position of the corresponding packet in the input file, NAN if
7016 frame rate of the input video, NAN if the input frame rate is unknown
7019 timestamp expressed in seconds, NAN if the input timestamp is unknown
7022 @subsection Commands
7023 The filter supports the following commands:
7027 Set the contrast expression.
7030 Set the brightness expression.
7033 Set the saturation expression.
7036 Set the gamma expression.
7039 Set the gamma_r expression.
7042 Set gamma_g expression.
7045 Set gamma_b expression.
7048 Set gamma_weight expression.
7050 The command accepts the same syntax of the corresponding option.
7052 If the specified expression is not valid, it is kept at its current
7059 Apply erosion effect to the video.
7061 This filter replaces the pixel by the local(3x3) minimum.
7063 It accepts the following options:
7070 Limit the maximum change for each plane, default is 65535.
7071 If 0, plane will remain unchanged.
7074 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
7077 Flags to local 3x3 coordinates maps like this:
7084 @section extractplanes
7086 Extract color channel components from input video stream into
7087 separate grayscale video streams.
7089 The filter accepts the following option:
7093 Set plane(s) to extract.
7095 Available values for planes are:
7106 Choosing planes not available in the input will result in an error.
7107 That means you cannot select @code{r}, @code{g}, @code{b} planes
7108 with @code{y}, @code{u}, @code{v} planes at same time.
7111 @subsection Examples
7115 Extract luma, u and v color channel component from input video frame
7116 into 3 grayscale outputs:
7118 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
7124 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
7126 For each input image, the filter will compute the optimal mapping from
7127 the input to the output given the codebook length, that is the number
7128 of distinct output colors.
7130 This filter accepts the following options.
7133 @item codebook_length, l
7134 Set codebook length. The value must be a positive integer, and
7135 represents the number of distinct output colors. Default value is 256.
7138 Set the maximum number of iterations to apply for computing the optimal
7139 mapping. The higher the value the better the result and the higher the
7140 computation time. Default value is 1.
7143 Set a random seed, must be an integer included between 0 and
7144 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
7145 will try to use a good random seed on a best effort basis.
7148 Set pal8 output pixel format. This option does not work with codebook
7149 length greater than 256.
7154 Apply a fade-in/out effect to the input video.
7156 It accepts the following parameters:
7160 The effect type can be either "in" for a fade-in, or "out" for a fade-out
7162 Default is @code{in}.
7164 @item start_frame, s
7165 Specify the number of the frame to start applying the fade
7166 effect at. Default is 0.
7169 The number of frames that the fade effect lasts. At the end of the
7170 fade-in effect, the output video will have the same intensity as the input video.
7171 At the end of the fade-out transition, the output video will be filled with the
7172 selected @option{color}.
7176 If set to 1, fade only alpha channel, if one exists on the input.
7179 @item start_time, st
7180 Specify the timestamp (in seconds) of the frame to start to apply the fade
7181 effect. If both start_frame and start_time are specified, the fade will start at
7182 whichever comes last. Default is 0.
7185 The number of seconds for which the fade effect has to last. At the end of the
7186 fade-in effect the output video will have the same intensity as the input video,
7187 at the end of the fade-out transition the output video will be filled with the
7188 selected @option{color}.
7189 If both duration and nb_frames are specified, duration is used. Default is 0
7190 (nb_frames is used by default).
7193 Specify the color of the fade. Default is "black".
7196 @subsection Examples
7200 Fade in the first 30 frames of video:
7205 The command above is equivalent to:
7211 Fade out the last 45 frames of a 200-frame video:
7214 fade=type=out:start_frame=155:nb_frames=45
7218 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
7220 fade=in:0:25, fade=out:975:25
7224 Make the first 5 frames yellow, then fade in from frame 5-24:
7226 fade=in:5:20:color=yellow
7230 Fade in alpha over first 25 frames of video:
7232 fade=in:0:25:alpha=1
7236 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
7238 fade=t=in:st=5.5:d=0.5
7244 Apply arbitrary expressions to samples in frequency domain
7248 Adjust the dc value (gain) of the luma plane of the image. The filter
7249 accepts an integer value in range @code{0} to @code{1000}. The default
7250 value is set to @code{0}.
7253 Adjust the dc value (gain) of the 1st chroma plane of the image. The
7254 filter accepts an integer value in range @code{0} to @code{1000}. The
7255 default value is set to @code{0}.
7258 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
7259 filter accepts an integer value in range @code{0} to @code{1000}. The
7260 default value is set to @code{0}.
7263 Set the frequency domain weight expression for the luma plane.
7266 Set the frequency domain weight expression for the 1st chroma plane.
7269 Set the frequency domain weight expression for the 2nd chroma plane.
7271 The filter accepts the following variables:
7274 The coordinates of the current sample.
7278 The width and height of the image.
7281 @subsection Examples
7287 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
7293 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
7299 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
7305 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
7312 Extract a single field from an interlaced image using stride
7313 arithmetic to avoid wasting CPU time. The output frames are marked as
7316 The filter accepts the following options:
7320 Specify whether to extract the top (if the value is @code{0} or
7321 @code{top}) or the bottom field (if the value is @code{1} or
7327 Create new frames by copying the top and bottom fields from surrounding frames
7328 supplied as numbers by the hint file.
7332 Set file containing hints: absolute/relative frame numbers.
7334 There must be one line for each frame in a clip. Each line must contain two
7335 numbers separated by the comma, optionally followed by @code{-} or @code{+}.
7336 Numbers supplied on each line of file can not be out of [N-1,N+1] where N
7337 is current frame number for @code{absolute} mode or out of [-1, 1] range
7338 for @code{relative} mode. First number tells from which frame to pick up top
7339 field and second number tells from which frame to pick up bottom field.
7341 If optionally followed by @code{+} output frame will be marked as interlaced,
7342 else if followed by @code{-} output frame will be marked as progressive, else
7343 it will be marked same as input frame.
7344 If line starts with @code{#} or @code{;} that line is skipped.
7347 Can be item @code{absolute} or @code{relative}. Default is @code{absolute}.
7350 Example of first several lines of @code{hint} file for @code{relative} mode:
7353 1,0 - # second frame, use third's frame top field and second's frame bottom field
7354 1,0 - # third frame, use fourth's frame top field and third's frame bottom field
7371 Field matching filter for inverse telecine. It is meant to reconstruct the
7372 progressive frames from a telecined stream. The filter does not drop duplicated
7373 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
7374 followed by a decimation filter such as @ref{decimate} in the filtergraph.
7376 The separation of the field matching and the decimation is notably motivated by
7377 the possibility of inserting a de-interlacing filter fallback between the two.
7378 If the source has mixed telecined and real interlaced content,
7379 @code{fieldmatch} will not be able to match fields for the interlaced parts.
7380 But these remaining combed frames will be marked as interlaced, and thus can be
7381 de-interlaced by a later filter such as @ref{yadif} before decimation.
7383 In addition to the various configuration options, @code{fieldmatch} can take an
7384 optional second stream, activated through the @option{ppsrc} option. If
7385 enabled, the frames reconstruction will be based on the fields and frames from
7386 this second stream. This allows the first input to be pre-processed in order to
7387 help the various algorithms of the filter, while keeping the output lossless
7388 (assuming the fields are matched properly). Typically, a field-aware denoiser,
7389 or brightness/contrast adjustments can help.
7391 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
7392 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
7393 which @code{fieldmatch} is based on. While the semantic and usage are very
7394 close, some behaviour and options names can differ.
7396 The @ref{decimate} filter currently only works for constant frame rate input.
7397 If your input has mixed telecined (30fps) and progressive content with a lower
7398 framerate like 24fps use the following filterchain to produce the necessary cfr
7399 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
7401 The filter accepts the following options:
7405 Specify the assumed field order of the input stream. Available values are:
7409 Auto detect parity (use FFmpeg's internal parity value).
7411 Assume bottom field first.
7413 Assume top field first.
7416 Note that it is sometimes recommended not to trust the parity announced by the
7419 Default value is @var{auto}.
7422 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
7423 sense that it won't risk creating jerkiness due to duplicate frames when
7424 possible, but if there are bad edits or blended fields it will end up
7425 outputting combed frames when a good match might actually exist. On the other
7426 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
7427 but will almost always find a good frame if there is one. The other values are
7428 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
7429 jerkiness and creating duplicate frames versus finding good matches in sections
7430 with bad edits, orphaned fields, blended fields, etc.
7432 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
7434 Available values are:
7438 2-way matching (p/c)
7440 2-way matching, and trying 3rd match if still combed (p/c + n)
7442 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
7444 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
7445 still combed (p/c + n + u/b)
7447 3-way matching (p/c/n)
7449 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
7450 detected as combed (p/c/n + u/b)
7453 The parenthesis at the end indicate the matches that would be used for that
7454 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
7457 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
7460 Default value is @var{pc_n}.
7463 Mark the main input stream as a pre-processed input, and enable the secondary
7464 input stream as the clean source to pick the fields from. See the filter
7465 introduction for more details. It is similar to the @option{clip2} feature from
7468 Default value is @code{0} (disabled).
7471 Set the field to match from. It is recommended to set this to the same value as
7472 @option{order} unless you experience matching failures with that setting. In
7473 certain circumstances changing the field that is used to match from can have a
7474 large impact on matching performance. Available values are:
7478 Automatic (same value as @option{order}).
7480 Match from the bottom field.
7482 Match from the top field.
7485 Default value is @var{auto}.
7488 Set whether or not chroma is included during the match comparisons. In most
7489 cases it is recommended to leave this enabled. You should set this to @code{0}
7490 only if your clip has bad chroma problems such as heavy rainbowing or other
7491 artifacts. Setting this to @code{0} could also be used to speed things up at
7492 the cost of some accuracy.
7494 Default value is @code{1}.
7498 These define an exclusion band which excludes the lines between @option{y0} and
7499 @option{y1} from being included in the field matching decision. An exclusion
7500 band can be used to ignore subtitles, a logo, or other things that may
7501 interfere with the matching. @option{y0} sets the starting scan line and
7502 @option{y1} sets the ending line; all lines in between @option{y0} and
7503 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
7504 @option{y0} and @option{y1} to the same value will disable the feature.
7505 @option{y0} and @option{y1} defaults to @code{0}.
7508 Set the scene change detection threshold as a percentage of maximum change on
7509 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
7510 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
7511 @option{scthresh} is @code{[0.0, 100.0]}.
7513 Default value is @code{12.0}.
7516 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
7517 account the combed scores of matches when deciding what match to use as the
7518 final match. Available values are:
7522 No final matching based on combed scores.
7524 Combed scores are only used when a scene change is detected.
7526 Use combed scores all the time.
7529 Default is @var{sc}.
7532 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
7533 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
7534 Available values are:
7538 No forced calculation.
7540 Force p/c/n calculations.
7542 Force p/c/n/u/b calculations.
7545 Default value is @var{none}.
7548 This is the area combing threshold used for combed frame detection. This
7549 essentially controls how "strong" or "visible" combing must be to be detected.
7550 Larger values mean combing must be more visible and smaller values mean combing
7551 can be less visible or strong and still be detected. Valid settings are from
7552 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
7553 be detected as combed). This is basically a pixel difference value. A good
7554 range is @code{[8, 12]}.
7556 Default value is @code{9}.
7559 Sets whether or not chroma is considered in the combed frame decision. Only
7560 disable this if your source has chroma problems (rainbowing, etc.) that are
7561 causing problems for the combed frame detection with chroma enabled. Actually,
7562 using @option{chroma}=@var{0} is usually more reliable, except for the case
7563 where there is chroma only combing in the source.
7565 Default value is @code{0}.
7569 Respectively set the x-axis and y-axis size of the window used during combed
7570 frame detection. This has to do with the size of the area in which
7571 @option{combpel} pixels are required to be detected as combed for a frame to be
7572 declared combed. See the @option{combpel} parameter description for more info.
7573 Possible values are any number that is a power of 2 starting at 4 and going up
7576 Default value is @code{16}.
7579 The number of combed pixels inside any of the @option{blocky} by
7580 @option{blockx} size blocks on the frame for the frame to be detected as
7581 combed. While @option{cthresh} controls how "visible" the combing must be, this
7582 setting controls "how much" combing there must be in any localized area (a
7583 window defined by the @option{blockx} and @option{blocky} settings) on the
7584 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
7585 which point no frames will ever be detected as combed). This setting is known
7586 as @option{MI} in TFM/VFM vocabulary.
7588 Default value is @code{80}.
7591 @anchor{p/c/n/u/b meaning}
7592 @subsection p/c/n/u/b meaning
7594 @subsubsection p/c/n
7596 We assume the following telecined stream:
7599 Top fields: 1 2 2 3 4
7600 Bottom fields: 1 2 3 4 4
7603 The numbers correspond to the progressive frame the fields relate to. Here, the
7604 first two frames are progressive, the 3rd and 4th are combed, and so on.
7606 When @code{fieldmatch} is configured to run a matching from bottom
7607 (@option{field}=@var{bottom}) this is how this input stream get transformed:
7612 B 1 2 3 4 4 <-- matching reference
7621 As a result of the field matching, we can see that some frames get duplicated.
7622 To perform a complete inverse telecine, you need to rely on a decimation filter
7623 after this operation. See for instance the @ref{decimate} filter.
7625 The same operation now matching from top fields (@option{field}=@var{top})
7630 T 1 2 2 3 4 <-- matching reference
7640 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
7641 basically, they refer to the frame and field of the opposite parity:
7644 @item @var{p} matches the field of the opposite parity in the previous frame
7645 @item @var{c} matches the field of the opposite parity in the current frame
7646 @item @var{n} matches the field of the opposite parity in the next frame
7651 The @var{u} and @var{b} matching are a bit special in the sense that they match
7652 from the opposite parity flag. In the following examples, we assume that we are
7653 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
7654 'x' is placed above and below each matched fields.
7656 With bottom matching (@option{field}=@var{bottom}):
7661 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7662 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7670 With top matching (@option{field}=@var{top}):
7675 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7676 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7684 @subsection Examples
7686 Simple IVTC of a top field first telecined stream:
7688 fieldmatch=order=tff:combmatch=none, decimate
7691 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
7693 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
7698 Transform the field order of the input video.
7700 It accepts the following parameters:
7705 The output field order. Valid values are @var{tff} for top field first or @var{bff}
7706 for bottom field first.
7709 The default value is @samp{tff}.
7711 The transformation is done by shifting the picture content up or down
7712 by one line, and filling the remaining line with appropriate picture content.
7713 This method is consistent with most broadcast field order converters.
7715 If the input video is not flagged as being interlaced, or it is already
7716 flagged as being of the required output field order, then this filter does
7717 not alter the incoming video.
7719 It is very useful when converting to or from PAL DV material,
7720 which is bottom field first.
7724 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
7727 @section fifo, afifo
7729 Buffer input images and send them when they are requested.
7731 It is mainly useful when auto-inserted by the libavfilter
7734 It does not take parameters.
7738 Find a rectangular object
7740 It accepts the following options:
7744 Filepath of the object image, needs to be in gray8.
7747 Detection threshold, default is 0.5.
7750 Number of mipmaps, default is 3.
7752 @item xmin, ymin, xmax, ymax
7753 Specifies the rectangle in which to search.
7756 @subsection Examples
7760 Generate a representative palette of a given video using @command{ffmpeg}:
7762 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
7768 Cover a rectangular object
7770 It accepts the following options:
7774 Filepath of the optional cover image, needs to be in yuv420.
7779 It accepts the following values:
7782 cover it by the supplied image
7784 cover it by interpolating the surrounding pixels
7787 Default value is @var{blur}.
7790 @subsection Examples
7794 Generate a representative palette of a given video using @command{ffmpeg}:
7796 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
7803 Convert the input video to one of the specified pixel formats.
7804 Libavfilter will try to pick one that is suitable as input to
7807 It accepts the following parameters:
7811 A '|'-separated list of pixel format names, such as
7812 "pix_fmts=yuv420p|monow|rgb24".
7816 @subsection Examples
7820 Convert the input video to the @var{yuv420p} format
7822 format=pix_fmts=yuv420p
7825 Convert the input video to any of the formats in the list
7827 format=pix_fmts=yuv420p|yuv444p|yuv410p
7834 Convert the video to specified constant frame rate by duplicating or dropping
7835 frames as necessary.
7837 It accepts the following parameters:
7841 The desired output frame rate. The default is @code{25}.
7846 Possible values are:
7849 zero round towards 0
7853 round towards -infinity
7855 round towards +infinity
7859 The default is @code{near}.
7862 Assume the first PTS should be the given value, in seconds. This allows for
7863 padding/trimming at the start of stream. By default, no assumption is made
7864 about the first frame's expected PTS, so no padding or trimming is done.
7865 For example, this could be set to 0 to pad the beginning with duplicates of
7866 the first frame if a video stream starts after the audio stream or to trim any
7867 frames with a negative PTS.
7871 Alternatively, the options can be specified as a flat string:
7872 @var{fps}[:@var{round}].
7874 See also the @ref{setpts} filter.
7876 @subsection Examples
7880 A typical usage in order to set the fps to 25:
7886 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
7888 fps=fps=film:round=near
7894 Pack two different video streams into a stereoscopic video, setting proper
7895 metadata on supported codecs. The two views should have the same size and
7896 framerate and processing will stop when the shorter video ends. Please note
7897 that you may conveniently adjust view properties with the @ref{scale} and
7900 It accepts the following parameters:
7904 The desired packing format. Supported values are:
7909 The views are next to each other (default).
7912 The views are on top of each other.
7915 The views are packed by line.
7918 The views are packed by column.
7921 The views are temporally interleaved.
7930 # Convert left and right views into a frame-sequential video
7931 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
7933 # Convert views into a side-by-side video with the same output resolution as the input
7934 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
7939 Change the frame rate by interpolating new video output frames from the source
7942 This filter is not designed to function correctly with interlaced media. If
7943 you wish to change the frame rate of interlaced media then you are required
7944 to deinterlace before this filter and re-interlace after this filter.
7946 A description of the accepted options follows.
7950 Specify the output frames per second. This option can also be specified
7951 as a value alone. The default is @code{50}.
7954 Specify the start of a range where the output frame will be created as a
7955 linear interpolation of two frames. The range is [@code{0}-@code{255}],
7956 the default is @code{15}.
7959 Specify the end of a range where the output frame will be created as a
7960 linear interpolation of two frames. The range is [@code{0}-@code{255}],
7961 the default is @code{240}.
7964 Specify the level at which a scene change is detected as a value between
7965 0 and 100 to indicate a new scene; a low value reflects a low
7966 probability for the current frame to introduce a new scene, while a higher
7967 value means the current frame is more likely to be one.
7968 The default is @code{7}.
7971 Specify flags influencing the filter process.
7973 Available value for @var{flags} is:
7976 @item scene_change_detect, scd
7977 Enable scene change detection using the value of the option @var{scene}.
7978 This flag is enabled by default.
7984 Select one frame every N-th frame.
7986 This filter accepts the following option:
7989 Select frame after every @code{step} frames.
7990 Allowed values are positive integers higher than 0. Default value is @code{1}.
7996 Apply a frei0r effect to the input video.
7998 To enable the compilation of this filter, you need to install the frei0r
7999 header and configure FFmpeg with @code{--enable-frei0r}.
8001 It accepts the following parameters:
8006 The name of the frei0r effect to load. If the environment variable
8007 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
8008 directories specified by the colon-separated list in @env{FREIOR_PATH}.
8009 Otherwise, the standard frei0r paths are searched, in this order:
8010 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
8011 @file{/usr/lib/frei0r-1/}.
8014 A '|'-separated list of parameters to pass to the frei0r effect.
8018 A frei0r effect parameter can be a boolean (its value is either
8019 "y" or "n"), a double, a color (specified as
8020 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
8021 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
8022 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
8023 @var{X} and @var{Y} are floating point numbers) and/or a string.
8025 The number and types of parameters depend on the loaded effect. If an
8026 effect parameter is not specified, the default value is set.
8028 @subsection Examples
8032 Apply the distort0r effect, setting the first two double parameters:
8034 frei0r=filter_name=distort0r:filter_params=0.5|0.01
8038 Apply the colordistance effect, taking a color as the first parameter:
8040 frei0r=colordistance:0.2/0.3/0.4
8041 frei0r=colordistance:violet
8042 frei0r=colordistance:0x112233
8046 Apply the perspective effect, specifying the top left and top right image
8049 frei0r=perspective:0.2/0.2|0.8/0.2
8053 For more information, see
8054 @url{http://frei0r.dyne.org}
8058 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
8060 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
8061 processing filter, one of them is performed once per block, not per pixel.
8062 This allows for much higher speed.
8064 The filter accepts the following options:
8068 Set quality. This option defines the number of levels for averaging. It accepts
8069 an integer in the range 4-5. Default value is @code{4}.
8072 Force a constant quantization parameter. It accepts an integer in range 0-63.
8073 If not set, the filter will use the QP from the video stream (if available).
8076 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
8077 more details but also more artifacts, while higher values make the image smoother
8078 but also blurrier. Default value is @code{0} − PSNR optimal.
8081 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8082 option may cause flicker since the B-Frames have often larger QP. Default is
8083 @code{0} (not enabled).
8089 The filter accepts the following options:
8093 Set the luminance expression.
8095 Set the chrominance blue expression.
8097 Set the chrominance red expression.
8099 Set the alpha expression.
8101 Set the red expression.
8103 Set the green expression.
8105 Set the blue expression.
8108 The colorspace is selected according to the specified options. If one
8109 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
8110 options is specified, the filter will automatically select a YCbCr
8111 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
8112 @option{blue_expr} options is specified, it will select an RGB
8115 If one of the chrominance expression is not defined, it falls back on the other
8116 one. If no alpha expression is specified it will evaluate to opaque value.
8117 If none of chrominance expressions are specified, they will evaluate
8118 to the luminance expression.
8120 The expressions can use the following variables and functions:
8124 The sequential number of the filtered frame, starting from @code{0}.
8128 The coordinates of the current sample.
8132 The width and height of the image.
8136 Width and height scale depending on the currently filtered plane. It is the
8137 ratio between the corresponding luma plane number of pixels and the current
8138 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
8139 @code{0.5,0.5} for chroma planes.
8142 Time of the current frame, expressed in seconds.
8145 Return the value of the pixel at location (@var{x},@var{y}) of the current
8149 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
8153 Return the value of the pixel at location (@var{x},@var{y}) of the
8154 blue-difference chroma plane. Return 0 if there is no such plane.
8157 Return the value of the pixel at location (@var{x},@var{y}) of the
8158 red-difference chroma plane. Return 0 if there is no such plane.
8163 Return the value of the pixel at location (@var{x},@var{y}) of the
8164 red/green/blue component. Return 0 if there is no such component.
8167 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
8168 plane. Return 0 if there is no such plane.
8171 For functions, if @var{x} and @var{y} are outside the area, the value will be
8172 automatically clipped to the closer edge.
8174 @subsection Examples
8178 Flip the image horizontally:
8184 Generate a bidimensional sine wave, with angle @code{PI/3} and a
8185 wavelength of 100 pixels:
8187 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
8191 Generate a fancy enigmatic moving light:
8193 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
8197 Generate a quick emboss effect:
8199 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
8203 Modify RGB components depending on pixel position:
8205 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
8209 Create a radial gradient that is the same size as the input (also see
8210 the @ref{vignette} filter):
8212 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
8218 Fix the banding artifacts that are sometimes introduced into nearly flat
8219 regions by truncation to 8-bit color depth.
8220 Interpolate the gradients that should go where the bands are, and
8223 It is designed for playback only. Do not use it prior to
8224 lossy compression, because compression tends to lose the dither and
8225 bring back the bands.
8227 It accepts the following parameters:
8232 The maximum amount by which the filter will change any one pixel. This is also
8233 the threshold for detecting nearly flat regions. Acceptable values range from
8234 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
8238 The neighborhood to fit the gradient to. A larger radius makes for smoother
8239 gradients, but also prevents the filter from modifying the pixels near detailed
8240 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
8241 values will be clipped to the valid range.
8245 Alternatively, the options can be specified as a flat string:
8246 @var{strength}[:@var{radius}]
8248 @subsection Examples
8252 Apply the filter with a @code{3.5} strength and radius of @code{8}:
8258 Specify radius, omitting the strength (which will fall-back to the default
8269 Apply a Hald CLUT to a video stream.
8271 First input is the video stream to process, and second one is the Hald CLUT.
8272 The Hald CLUT input can be a simple picture or a complete video stream.
8274 The filter accepts the following options:
8278 Force termination when the shortest input terminates. Default is @code{0}.
8280 Continue applying the last CLUT after the end of the stream. A value of
8281 @code{0} disable the filter after the last frame of the CLUT is reached.
8282 Default is @code{1}.
8285 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
8286 filters share the same internals).
8288 More information about the Hald CLUT can be found on Eskil Steenberg's website
8289 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
8291 @subsection Workflow examples
8293 @subsubsection Hald CLUT video stream
8295 Generate an identity Hald CLUT stream altered with various effects:
8297 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
8300 Note: make sure you use a lossless codec.
8302 Then use it with @code{haldclut} to apply it on some random stream:
8304 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
8307 The Hald CLUT will be applied to the 10 first seconds (duration of
8308 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
8309 to the remaining frames of the @code{mandelbrot} stream.
8311 @subsubsection Hald CLUT with preview
8313 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
8314 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
8315 biggest possible square starting at the top left of the picture. The remaining
8316 padding pixels (bottom or right) will be ignored. This area can be used to add
8317 a preview of the Hald CLUT.
8319 Typically, the following generated Hald CLUT will be supported by the
8320 @code{haldclut} filter:
8323 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
8324 pad=iw+320 [padded_clut];
8325 smptebars=s=320x256, split [a][b];
8326 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
8327 [main][b] overlay=W-320" -frames:v 1 clut.png
8330 It contains the original and a preview of the effect of the CLUT: SMPTE color
8331 bars are displayed on the right-top, and below the same color bars processed by
8334 Then, the effect of this Hald CLUT can be visualized with:
8336 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
8341 Decodes High Definition Compatible Digital (HDCD) data. A 16-bit PCM stream with
8342 embedded HDCD codes is expanded into a 20-bit PCM stream.
8344 The filter supports the Peak Extend and Low-level Gain Adjustment features
8345 of HDCD, and detects the Transient Filter flag.
8348 ffmpeg -i HDCD16.flac -af hdcd OUT24.flac
8351 When using the filter with wav, note the default encoding for wav is 16-bit,
8352 so the resulting 20-bit stream will be truncated back to 16-bit. Use something
8353 like @command{-acodec pcm_s24le} after the filter to get 24-bit PCM output.
8355 ffmpeg -i HDCD16.wav -af hdcd OUT16.wav
8356 ffmpeg -i HDCD16.wav -af hdcd -acodec pcm_s24le OUT24.wav
8359 The filter accepts the following options:
8362 @item process_stereo
8363 Process the stereo channels together. If target_gain does not match between
8364 channels, consider it invalid and use the last valid target_gain.
8367 Always extend peaks above -3dBFS even if PE isn't signaled.
8370 Replace audio with a solid tone and adjust the amplitude to signal some
8371 specific aspect of the decoding process. The output file can be loaded in
8372 an audio editor alongside the original to aid analysis.
8374 @code{analyze_mode=pe:force_pe=1} can be used to see all samples above the PE level.
8381 Gain adjustment level at each sample
8383 Samples where peak extend occurs
8385 Samples where the code detect timer is active
8387 Samples where the target gain does not match between channels
8393 Flip the input video horizontally.
8395 For example, to horizontally flip the input video with @command{ffmpeg}:
8397 ffmpeg -i in.avi -vf "hflip" out.avi
8401 This filter applies a global color histogram equalization on a
8404 It can be used to correct video that has a compressed range of pixel
8405 intensities. The filter redistributes the pixel intensities to
8406 equalize their distribution across the intensity range. It may be
8407 viewed as an "automatically adjusting contrast filter". This filter is
8408 useful only for correcting degraded or poorly captured source
8411 The filter accepts the following options:
8415 Determine the amount of equalization to be applied. As the strength
8416 is reduced, the distribution of pixel intensities more-and-more
8417 approaches that of the input frame. The value must be a float number
8418 in the range [0,1] and defaults to 0.200.
8421 Set the maximum intensity that can generated and scale the output
8422 values appropriately. The strength should be set as desired and then
8423 the intensity can be limited if needed to avoid washing-out. The value
8424 must be a float number in the range [0,1] and defaults to 0.210.
8427 Set the antibanding level. If enabled the filter will randomly vary
8428 the luminance of output pixels by a small amount to avoid banding of
8429 the histogram. Possible values are @code{none}, @code{weak} or
8430 @code{strong}. It defaults to @code{none}.
8435 Compute and draw a color distribution histogram for the input video.
8437 The computed histogram is a representation of the color component
8438 distribution in an image.
8440 Standard histogram displays the color components distribution in an image.
8441 Displays color graph for each color component. Shows distribution of
8442 the Y, U, V, A or R, G, B components, depending on input format, in the
8443 current frame. Below each graph a color component scale meter is shown.
8445 The filter accepts the following options:
8449 Set height of level. Default value is @code{200}.
8450 Allowed range is [50, 2048].
8453 Set height of color scale. Default value is @code{12}.
8454 Allowed range is [0, 40].
8458 It accepts the following values:
8461 Per color component graphs are placed below each other.
8464 Presents information identical to that in the @code{parade}, except
8465 that the graphs representing color components are superimposed directly
8468 Default is @code{parade}.
8471 Set mode. Can be either @code{linear}, or @code{logarithmic}.
8472 Default is @code{linear}.
8475 Set what color components to display.
8476 Default is @code{7}.
8479 @subsection Examples
8484 Calculate and draw histogram:
8486 ffplay -i input -vf histogram
8494 This is a high precision/quality 3d denoise filter. It aims to reduce
8495 image noise, producing smooth images and making still images really
8496 still. It should enhance compressibility.
8498 It accepts the following optional parameters:
8502 A non-negative floating point number which specifies spatial luma strength.
8505 @item chroma_spatial
8506 A non-negative floating point number which specifies spatial chroma strength.
8507 It defaults to 3.0*@var{luma_spatial}/4.0.
8510 A floating point number which specifies luma temporal strength. It defaults to
8511 6.0*@var{luma_spatial}/4.0.
8514 A floating point number which specifies chroma temporal strength. It defaults to
8515 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
8518 @anchor{hwupload_cuda}
8519 @section hwupload_cuda
8521 Upload system memory frames to a CUDA device.
8523 It accepts the following optional parameters:
8527 The number of the CUDA device to use
8532 Apply a high-quality magnification filter designed for pixel art. This filter
8533 was originally created by Maxim Stepin.
8535 It accepts the following option:
8539 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
8540 @code{hq3x} and @code{4} for @code{hq4x}.
8541 Default is @code{3}.
8545 Stack input videos horizontally.
8547 All streams must be of same pixel format and of same height.
8549 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
8550 to create same output.
8552 The filter accept the following option:
8556 Set number of input streams. Default is 2.
8559 If set to 1, force the output to terminate when the shortest input
8560 terminates. Default value is 0.
8565 Modify the hue and/or the saturation of the input.
8567 It accepts the following parameters:
8571 Specify the hue angle as a number of degrees. It accepts an expression,
8572 and defaults to "0".
8575 Specify the saturation in the [-10,10] range. It accepts an expression and
8579 Specify the hue angle as a number of radians. It accepts an
8580 expression, and defaults to "0".
8583 Specify the brightness in the [-10,10] range. It accepts an expression and
8587 @option{h} and @option{H} are mutually exclusive, and can't be
8588 specified at the same time.
8590 The @option{b}, @option{h}, @option{H} and @option{s} option values are
8591 expressions containing the following constants:
8595 frame count of the input frame starting from 0
8598 presentation timestamp of the input frame expressed in time base units
8601 frame rate of the input video, NAN if the input frame rate is unknown
8604 timestamp expressed in seconds, NAN if the input timestamp is unknown
8607 time base of the input video
8610 @subsection Examples
8614 Set the hue to 90 degrees and the saturation to 1.0:
8620 Same command but expressing the hue in radians:
8626 Rotate hue and make the saturation swing between 0
8627 and 2 over a period of 1 second:
8629 hue="H=2*PI*t: s=sin(2*PI*t)+1"
8633 Apply a 3 seconds saturation fade-in effect starting at 0:
8638 The general fade-in expression can be written as:
8640 hue="s=min(0\, max((t-START)/DURATION\, 1))"
8644 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
8646 hue="s=max(0\, min(1\, (8-t)/3))"
8649 The general fade-out expression can be written as:
8651 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
8656 @subsection Commands
8658 This filter supports the following commands:
8664 Modify the hue and/or the saturation and/or brightness of the input video.
8665 The command accepts the same syntax of the corresponding option.
8667 If the specified expression is not valid, it is kept at its current
8673 Detect video interlacing type.
8675 This filter tries to detect if the input frames as interlaced, progressive,
8676 top or bottom field first. It will also try and detect fields that are
8677 repeated between adjacent frames (a sign of telecine).
8679 Single frame detection considers only immediately adjacent frames when classifying each frame.
8680 Multiple frame detection incorporates the classification history of previous frames.
8682 The filter will log these metadata values:
8685 @item single.current_frame
8686 Detected type of current frame using single-frame detection. One of:
8687 ``tff'' (top field first), ``bff'' (bottom field first),
8688 ``progressive'', or ``undetermined''
8691 Cumulative number of frames detected as top field first using single-frame detection.
8694 Cumulative number of frames detected as top field first using multiple-frame detection.
8697 Cumulative number of frames detected as bottom field first using single-frame detection.
8699 @item multiple.current_frame
8700 Detected type of current frame using multiple-frame detection. One of:
8701 ``tff'' (top field first), ``bff'' (bottom field first),
8702 ``progressive'', or ``undetermined''
8705 Cumulative number of frames detected as bottom field first using multiple-frame detection.
8707 @item single.progressive
8708 Cumulative number of frames detected as progressive using single-frame detection.
8710 @item multiple.progressive
8711 Cumulative number of frames detected as progressive using multiple-frame detection.
8713 @item single.undetermined
8714 Cumulative number of frames that could not be classified using single-frame detection.
8716 @item multiple.undetermined
8717 Cumulative number of frames that could not be classified using multiple-frame detection.
8719 @item repeated.current_frame
8720 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
8722 @item repeated.neither
8723 Cumulative number of frames with no repeated field.
8726 Cumulative number of frames with the top field repeated from the previous frame's top field.
8728 @item repeated.bottom
8729 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
8732 The filter accepts the following options:
8736 Set interlacing threshold.
8738 Set progressive threshold.
8740 Threshold for repeated field detection.
8742 Number of frames after which a given frame's contribution to the
8743 statistics is halved (i.e., it contributes only 0.5 to it's
8744 classification). The default of 0 means that all frames seen are given
8745 full weight of 1.0 forever.
8746 @item analyze_interlaced_flag
8747 When this is not 0 then idet will use the specified number of frames to determine
8748 if the interlaced flag is accurate, it will not count undetermined frames.
8749 If the flag is found to be accurate it will be used without any further
8750 computations, if it is found to be inaccurate it will be cleared without any
8751 further computations. This allows inserting the idet filter as a low computational
8752 method to clean up the interlaced flag
8757 Deinterleave or interleave fields.
8759 This filter allows one to process interlaced images fields without
8760 deinterlacing them. Deinterleaving splits the input frame into 2
8761 fields (so called half pictures). Odd lines are moved to the top
8762 half of the output image, even lines to the bottom half.
8763 You can process (filter) them independently and then re-interleave them.
8765 The filter accepts the following options:
8769 @item chroma_mode, c
8771 Available values for @var{luma_mode}, @var{chroma_mode} and
8772 @var{alpha_mode} are:
8778 @item deinterleave, d
8779 Deinterleave fields, placing one above the other.
8782 Interleave fields. Reverse the effect of deinterleaving.
8784 Default value is @code{none}.
8787 @item chroma_swap, cs
8788 @item alpha_swap, as
8789 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
8794 Apply inflate effect to the video.
8796 This filter replaces the pixel by the local(3x3) average by taking into account
8797 only values higher than the pixel.
8799 It accepts the following options:
8806 Limit the maximum change for each plane, default is 65535.
8807 If 0, plane will remain unchanged.
8812 Simple interlacing filter from progressive contents. This interleaves upper (or
8813 lower) lines from odd frames with lower (or upper) lines from even frames,
8814 halving the frame rate and preserving image height.
8817 Original Original New Frame
8818 Frame 'j' Frame 'j+1' (tff)
8819 ========== =========== ==================
8820 Line 0 --------------------> Frame 'j' Line 0
8821 Line 1 Line 1 ----> Frame 'j+1' Line 1
8822 Line 2 ---------------------> Frame 'j' Line 2
8823 Line 3 Line 3 ----> Frame 'j+1' Line 3
8825 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
8828 It accepts the following optional parameters:
8832 This determines whether the interlaced frame is taken from the even
8833 (tff - default) or odd (bff) lines of the progressive frame.
8836 Enable (default) or disable the vertical lowpass filter to avoid twitter
8837 interlacing and reduce moire patterns.
8842 Deinterlace input video by applying Donald Graft's adaptive kernel
8843 deinterling. Work on interlaced parts of a video to produce
8846 The description of the accepted parameters follows.
8850 Set the threshold which affects the filter's tolerance when
8851 determining if a pixel line must be processed. It must be an integer
8852 in the range [0,255] and defaults to 10. A value of 0 will result in
8853 applying the process on every pixels.
8856 Paint pixels exceeding the threshold value to white if set to 1.
8860 Set the fields order. Swap fields if set to 1, leave fields alone if
8864 Enable additional sharpening if set to 1. Default is 0.
8867 Enable twoway sharpening if set to 1. Default is 0.
8870 @subsection Examples
8874 Apply default values:
8876 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
8880 Enable additional sharpening:
8886 Paint processed pixels in white:
8892 @section lenscorrection
8894 Correct radial lens distortion
8896 This filter can be used to correct for radial distortion as can result from the use
8897 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
8898 one can use tools available for example as part of opencv or simply trial-and-error.
8899 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
8900 and extract the k1 and k2 coefficients from the resulting matrix.
8902 Note that effectively the same filter is available in the open-source tools Krita and
8903 Digikam from the KDE project.
8905 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
8906 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
8907 brightness distribution, so you may want to use both filters together in certain
8908 cases, though you will have to take care of ordering, i.e. whether vignetting should
8909 be applied before or after lens correction.
8913 The filter accepts the following options:
8917 Relative x-coordinate of the focal point of the image, and thereby the center of the
8918 distortion. This value has a range [0,1] and is expressed as fractions of the image
8921 Relative y-coordinate of the focal point of the image, and thereby the center of the
8922 distortion. This value has a range [0,1] and is expressed as fractions of the image
8925 Coefficient of the quadratic correction term. 0.5 means no correction.
8927 Coefficient of the double quadratic correction term. 0.5 means no correction.
8930 The formula that generates the correction is:
8932 @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)
8934 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
8935 distances from the focal point in the source and target images, respectively.
8941 The filter accepts the following options:
8945 Set the number of loops.
8948 Set maximal size in number of frames.
8951 Set first frame of loop.
8957 Apply a 3D LUT to an input video.
8959 The filter accepts the following options:
8963 Set the 3D LUT file name.
8965 Currently supported formats:
8977 Select interpolation mode.
8979 Available values are:
8983 Use values from the nearest defined point.
8985 Interpolate values using the 8 points defining a cube.
8987 Interpolate values using a tetrahedron.
8991 @section lut, lutrgb, lutyuv
8993 Compute a look-up table for binding each pixel component input value
8994 to an output value, and apply it to the input video.
8996 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
8997 to an RGB input video.
8999 These filters accept the following parameters:
9002 set first pixel component expression
9004 set second pixel component expression
9006 set third pixel component expression
9008 set fourth pixel component expression, corresponds to the alpha component
9011 set red component expression
9013 set green component expression
9015 set blue component expression
9017 alpha component expression
9020 set Y/luminance component expression
9022 set U/Cb component expression
9024 set V/Cr component expression
9027 Each of them specifies the expression to use for computing the lookup table for
9028 the corresponding pixel component values.
9030 The exact component associated to each of the @var{c*} options depends on the
9033 The @var{lut} filter requires either YUV or RGB pixel formats in input,
9034 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
9036 The expressions can contain the following constants and functions:
9041 The input width and height.
9044 The input value for the pixel component.
9047 The input value, clipped to the @var{minval}-@var{maxval} range.
9050 The maximum value for the pixel component.
9053 The minimum value for the pixel component.
9056 The negated value for the pixel component value, clipped to the
9057 @var{minval}-@var{maxval} range; it corresponds to the expression
9058 "maxval-clipval+minval".
9061 The computed value in @var{val}, clipped to the
9062 @var{minval}-@var{maxval} range.
9064 @item gammaval(gamma)
9065 The computed gamma correction value of the pixel component value,
9066 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
9068 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
9072 All expressions default to "val".
9074 @subsection Examples
9080 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
9081 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
9084 The above is the same as:
9086 lutrgb="r=negval:g=negval:b=negval"
9087 lutyuv="y=negval:u=negval:v=negval"
9097 Remove chroma components, turning the video into a graytone image:
9099 lutyuv="u=128:v=128"
9103 Apply a luma burning effect:
9109 Remove green and blue components:
9115 Set a constant alpha channel value on input:
9117 format=rgba,lutrgb=a="maxval-minval/2"
9121 Correct luminance gamma by a factor of 0.5:
9123 lutyuv=y=gammaval(0.5)
9127 Discard least significant bits of luma:
9129 lutyuv=y='bitand(val, 128+64+32)'
9133 @section maskedmerge
9135 Merge the first input stream with the second input stream using per pixel
9136 weights in the third input stream.
9138 A value of 0 in the third stream pixel component means that pixel component
9139 from first stream is returned unchanged, while maximum value (eg. 255 for
9140 8-bit videos) means that pixel component from second stream is returned
9141 unchanged. Intermediate values define the amount of merging between both
9142 input stream's pixel components.
9144 This filter accepts the following options:
9147 Set which planes will be processed as bitmap, unprocessed planes will be
9148 copied from first stream.
9149 By default value 0xf, all planes will be processed.
9154 Apply motion-compensation deinterlacing.
9156 It needs one field per frame as input and must thus be used together
9157 with yadif=1/3 or equivalent.
9159 This filter accepts the following options:
9162 Set the deinterlacing mode.
9164 It accepts one of the following values:
9169 use iterative motion estimation
9171 like @samp{slow}, but use multiple reference frames.
9173 Default value is @samp{fast}.
9176 Set the picture field parity assumed for the input video. It must be
9177 one of the following values:
9181 assume top field first
9183 assume bottom field first
9186 Default value is @samp{bff}.
9189 Set per-block quantization parameter (QP) used by the internal
9192 Higher values should result in a smoother motion vector field but less
9193 optimal individual vectors. Default value is 1.
9196 @section mergeplanes
9198 Merge color channel components from several video streams.
9200 The filter accepts up to 4 input streams, and merge selected input
9201 planes to the output video.
9203 This filter accepts the following options:
9206 Set input to output plane mapping. Default is @code{0}.
9208 The mappings is specified as a bitmap. It should be specified as a
9209 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
9210 mapping for the first plane of the output stream. 'A' sets the number of
9211 the input stream to use (from 0 to 3), and 'a' the plane number of the
9212 corresponding input to use (from 0 to 3). The rest of the mappings is
9213 similar, 'Bb' describes the mapping for the output stream second
9214 plane, 'Cc' describes the mapping for the output stream third plane and
9215 'Dd' describes the mapping for the output stream fourth plane.
9218 Set output pixel format. Default is @code{yuva444p}.
9221 @subsection Examples
9225 Merge three gray video streams of same width and height into single video stream:
9227 [a0][a1][a2]mergeplanes=0x001020:yuv444p
9231 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
9233 [a0][a1]mergeplanes=0x00010210:yuva444p
9237 Swap Y and A plane in yuva444p stream:
9239 format=yuva444p,mergeplanes=0x03010200:yuva444p
9243 Swap U and V plane in yuv420p stream:
9245 format=yuv420p,mergeplanes=0x000201:yuv420p
9249 Cast a rgb24 clip to yuv444p:
9251 format=rgb24,mergeplanes=0x000102:yuv444p
9257 Drop frames that do not differ greatly from the previous frame in
9258 order to reduce frame rate.
9260 The main use of this filter is for very-low-bitrate encoding
9261 (e.g. streaming over dialup modem), but it could in theory be used for
9262 fixing movies that were inverse-telecined incorrectly.
9264 A description of the accepted options follows.
9268 Set the maximum number of consecutive frames which can be dropped (if
9269 positive), or the minimum interval between dropped frames (if
9270 negative). If the value is 0, the frame is dropped unregarding the
9271 number of previous sequentially dropped frames.
9278 Set the dropping threshold values.
9280 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
9281 represent actual pixel value differences, so a threshold of 64
9282 corresponds to 1 unit of difference for each pixel, or the same spread
9283 out differently over the block.
9285 A frame is a candidate for dropping if no 8x8 blocks differ by more
9286 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
9287 meaning the whole image) differ by more than a threshold of @option{lo}.
9289 Default value for @option{hi} is 64*12, default value for @option{lo} is
9290 64*5, and default value for @option{frac} is 0.33.
9298 It accepts an integer in input; if non-zero it negates the
9299 alpha component (if available). The default value in input is 0.
9303 Deinterlace video using neural network edge directed interpolation.
9305 This filter accepts the following options:
9309 Mandatory option, without binary file filter can not work.
9310 Currently file can be found here:
9311 https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
9314 Set which frames to deinterlace, by default it is @code{all}.
9315 Can be @code{all} or @code{interlaced}.
9318 Set mode of operation.
9320 Can be one of the following:
9324 Use frame flags, both fields.
9326 Use frame flags, single field.
9330 Use bottom field only.
9332 Use both fields, top first.
9334 Use both fields, bottom first.
9338 Set which planes to process, by default filter process all frames.
9341 Set size of local neighborhood around each pixel, used by the predictor neural
9344 Can be one of the following:
9357 Set the number of neurons in predicctor neural network.
9358 Can be one of the following:
9369 Controls the number of different neural network predictions that are blended
9370 together to compute the final output value. Can be @code{fast}, default or
9374 Set which set of weights to use in the predictor.
9375 Can be one of the following:
9379 weights trained to minimize absolute error
9381 weights trained to minimize squared error
9385 Controls whether or not the prescreener neural network is used to decide
9386 which pixels should be processed by the predictor neural network and which
9387 can be handled by simple cubic interpolation.
9388 The prescreener is trained to know whether cubic interpolation will be
9389 sufficient for a pixel or whether it should be predicted by the predictor nn.
9390 The computational complexity of the prescreener nn is much less than that of
9391 the predictor nn. Since most pixels can be handled by cubic interpolation,
9392 using the prescreener generally results in much faster processing.
9393 The prescreener is pretty accurate, so the difference between using it and not
9394 using it is almost always unnoticeable.
9396 Can be one of the following:
9404 Default is @code{new}.
9407 Set various debugging flags.
9412 Force libavfilter not to use any of the specified pixel formats for the
9413 input to the next filter.
9415 It accepts the following parameters:
9419 A '|'-separated list of pixel format names, such as
9420 apix_fmts=yuv420p|monow|rgb24".
9424 @subsection Examples
9428 Force libavfilter to use a format different from @var{yuv420p} for the
9429 input to the vflip filter:
9431 noformat=pix_fmts=yuv420p,vflip
9435 Convert the input video to any of the formats not contained in the list:
9437 noformat=yuv420p|yuv444p|yuv410p
9443 Add noise on video input frame.
9445 The filter accepts the following options:
9453 Set noise seed for specific pixel component or all pixel components in case
9454 of @var{all_seed}. Default value is @code{123457}.
9456 @item all_strength, alls
9457 @item c0_strength, c0s
9458 @item c1_strength, c1s
9459 @item c2_strength, c2s
9460 @item c3_strength, c3s
9461 Set noise strength for specific pixel component or all pixel components in case
9462 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
9464 @item all_flags, allf
9469 Set pixel component flags or set flags for all components if @var{all_flags}.
9470 Available values for component flags are:
9473 averaged temporal noise (smoother)
9475 mix random noise with a (semi)regular pattern
9477 temporal noise (noise pattern changes between frames)
9479 uniform noise (gaussian otherwise)
9483 @subsection Examples
9485 Add temporal and uniform noise to input video:
9487 noise=alls=20:allf=t+u
9492 Pass the video source unchanged to the output.
9495 Optical Character Recognition
9497 This filter uses Tesseract for optical character recognition.
9499 It accepts the following options:
9503 Set datapath to tesseract data. Default is to use whatever was
9504 set at installation.
9507 Set language, default is "eng".
9510 Set character whitelist.
9513 Set character blacklist.
9516 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
9520 Apply a video transform using libopencv.
9522 To enable this filter, install the libopencv library and headers and
9523 configure FFmpeg with @code{--enable-libopencv}.
9525 It accepts the following parameters:
9530 The name of the libopencv filter to apply.
9533 The parameters to pass to the libopencv filter. If not specified, the default
9538 Refer to the official libopencv documentation for more precise
9540 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
9542 Several libopencv filters are supported; see the following subsections.
9547 Dilate an image by using a specific structuring element.
9548 It corresponds to the libopencv function @code{cvDilate}.
9550 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
9552 @var{struct_el} represents a structuring element, and has the syntax:
9553 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
9555 @var{cols} and @var{rows} represent the number of columns and rows of
9556 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
9557 point, and @var{shape} the shape for the structuring element. @var{shape}
9558 must be "rect", "cross", "ellipse", or "custom".
9560 If the value for @var{shape} is "custom", it must be followed by a
9561 string of the form "=@var{filename}". The file with name
9562 @var{filename} is assumed to represent a binary image, with each
9563 printable character corresponding to a bright pixel. When a custom
9564 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
9565 or columns and rows of the read file are assumed instead.
9567 The default value for @var{struct_el} is "3x3+0x0/rect".
9569 @var{nb_iterations} specifies the number of times the transform is
9570 applied to the image, and defaults to 1.
9574 # Use the default values
9577 # Dilate using a structuring element with a 5x5 cross, iterating two times
9578 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
9580 # Read the shape from the file diamond.shape, iterating two times.
9581 # The file diamond.shape may contain a pattern of characters like this
9587 # The specified columns and rows are ignored
9588 # but the anchor point coordinates are not
9589 ocv=dilate:0x0+2x2/custom=diamond.shape|2
9594 Erode an image by using a specific structuring element.
9595 It corresponds to the libopencv function @code{cvErode}.
9597 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
9598 with the same syntax and semantics as the @ref{dilate} filter.
9602 Smooth the input video.
9604 The filter takes the following parameters:
9605 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
9607 @var{type} is the type of smooth filter to apply, and must be one of
9608 the following values: "blur", "blur_no_scale", "median", "gaussian",
9609 or "bilateral". The default value is "gaussian".
9611 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
9612 depend on the smooth type. @var{param1} and
9613 @var{param2} accept integer positive values or 0. @var{param3} and
9614 @var{param4} accept floating point values.
9616 The default value for @var{param1} is 3. The default value for the
9617 other parameters is 0.
9619 These parameters correspond to the parameters assigned to the
9620 libopencv function @code{cvSmooth}.
9625 Overlay one video on top of another.
9627 It takes two inputs and has one output. The first input is the "main"
9628 video on which the second input is overlaid.
9630 It accepts the following parameters:
9632 A description of the accepted options follows.
9637 Set the expression for the x and y coordinates of the overlaid video
9638 on the main video. Default value is "0" for both expressions. In case
9639 the expression is invalid, it is set to a huge value (meaning that the
9640 overlay will not be displayed within the output visible area).
9643 The action to take when EOF is encountered on the secondary input; it accepts
9644 one of the following values:
9648 Repeat the last frame (the default).
9652 Pass the main input through.
9656 Set when the expressions for @option{x}, and @option{y} are evaluated.
9658 It accepts the following values:
9661 only evaluate expressions once during the filter initialization or
9662 when a command is processed
9665 evaluate expressions for each incoming frame
9668 Default value is @samp{frame}.
9671 If set to 1, force the output to terminate when the shortest input
9672 terminates. Default value is 0.
9675 Set the format for the output video.
9677 It accepts the following values:
9692 Default value is @samp{yuv420}.
9694 @item rgb @emph{(deprecated)}
9695 If set to 1, force the filter to accept inputs in the RGB
9696 color space. Default value is 0. This option is deprecated, use
9697 @option{format} instead.
9700 If set to 1, force the filter to draw the last overlay frame over the
9701 main input until the end of the stream. A value of 0 disables this
9702 behavior. Default value is 1.
9705 The @option{x}, and @option{y} expressions can contain the following
9711 The main input width and height.
9715 The overlay input width and height.
9719 The computed values for @var{x} and @var{y}. They are evaluated for
9724 horizontal and vertical chroma subsample values of the output
9725 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
9729 the number of input frame, starting from 0
9732 the position in the file of the input frame, NAN if unknown
9735 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
9739 Note that the @var{n}, @var{pos}, @var{t} variables are available only
9740 when evaluation is done @emph{per frame}, and will evaluate to NAN
9741 when @option{eval} is set to @samp{init}.
9743 Be aware that frames are taken from each input video in timestamp
9744 order, hence, if their initial timestamps differ, it is a good idea
9745 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
9746 have them begin in the same zero timestamp, as the example for
9747 the @var{movie} filter does.
9749 You can chain together more overlays but you should test the
9750 efficiency of such approach.
9752 @subsection Commands
9754 This filter supports the following commands:
9758 Modify the x and y of the overlay input.
9759 The command accepts the same syntax of the corresponding option.
9761 If the specified expression is not valid, it is kept at its current
9765 @subsection Examples
9769 Draw the overlay at 10 pixels from the bottom right corner of the main
9772 overlay=main_w-overlay_w-10:main_h-overlay_h-10
9775 Using named options the example above becomes:
9777 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
9781 Insert a transparent PNG logo in the bottom left corner of the input,
9782 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
9784 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
9788 Insert 2 different transparent PNG logos (second logo on bottom
9789 right corner) using the @command{ffmpeg} tool:
9791 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
9795 Add a transparent color layer on top of the main video; @code{WxH}
9796 must specify the size of the main input to the overlay filter:
9798 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
9802 Play an original video and a filtered version (here with the deshake
9803 filter) side by side using the @command{ffplay} tool:
9805 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
9808 The above command is the same as:
9810 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
9814 Make a sliding overlay appearing from the left to the right top part of the
9815 screen starting since time 2:
9817 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
9821 Compose output by putting two input videos side to side:
9823 ffmpeg -i left.avi -i right.avi -filter_complex "
9824 nullsrc=size=200x100 [background];
9825 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
9826 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
9827 [background][left] overlay=shortest=1 [background+left];
9828 [background+left][right] overlay=shortest=1:x=100 [left+right]
9833 Mask 10-20 seconds of a video by applying the delogo filter to a section
9835 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
9836 -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]'
9841 Chain several overlays in cascade:
9843 nullsrc=s=200x200 [bg];
9844 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
9845 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
9846 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
9847 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
9848 [in3] null, [mid2] overlay=100:100 [out0]
9855 Apply Overcomplete Wavelet denoiser.
9857 The filter accepts the following options:
9863 Larger depth values will denoise lower frequency components more, but
9864 slow down filtering.
9866 Must be an int in the range 8-16, default is @code{8}.
9868 @item luma_strength, ls
9871 Must be a double value in the range 0-1000, default is @code{1.0}.
9873 @item chroma_strength, cs
9874 Set chroma strength.
9876 Must be a double value in the range 0-1000, default is @code{1.0}.
9882 Add paddings to the input image, and place the original input at the
9883 provided @var{x}, @var{y} coordinates.
9885 It accepts the following parameters:
9890 Specify an expression for the size of the output image with the
9891 paddings added. If the value for @var{width} or @var{height} is 0, the
9892 corresponding input size is used for the output.
9894 The @var{width} expression can reference the value set by the
9895 @var{height} expression, and vice versa.
9897 The default value of @var{width} and @var{height} is 0.
9901 Specify the offsets to place the input image at within the padded area,
9902 with respect to the top/left border of the output image.
9904 The @var{x} expression can reference the value set by the @var{y}
9905 expression, and vice versa.
9907 The default value of @var{x} and @var{y} is 0.
9910 Specify the color of the padded area. For the syntax of this option,
9911 check the "Color" section in the ffmpeg-utils manual.
9913 The default value of @var{color} is "black".
9916 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
9917 options are expressions containing the following constants:
9922 The input video width and height.
9926 These are the same as @var{in_w} and @var{in_h}.
9930 The output width and height (the size of the padded area), as
9931 specified by the @var{width} and @var{height} expressions.
9935 These are the same as @var{out_w} and @var{out_h}.
9939 The x and y offsets as specified by the @var{x} and @var{y}
9940 expressions, or NAN if not yet specified.
9943 same as @var{iw} / @var{ih}
9946 input sample aspect ratio
9949 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
9953 The horizontal and vertical chroma subsample values. For example for the
9954 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9957 @subsection Examples
9961 Add paddings with the color "violet" to the input video. The output video
9962 size is 640x480, and the top-left corner of the input video is placed at
9965 pad=640:480:0:40:violet
9968 The example above is equivalent to the following command:
9970 pad=width=640:height=480:x=0:y=40:color=violet
9974 Pad the input to get an output with dimensions increased by 3/2,
9975 and put the input video at the center of the padded area:
9977 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
9981 Pad the input to get a squared output with size equal to the maximum
9982 value between the input width and height, and put the input video at
9983 the center of the padded area:
9985 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
9989 Pad the input to get a final w/h ratio of 16:9:
9991 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
9995 In case of anamorphic video, in order to set the output display aspect
9996 correctly, it is necessary to use @var{sar} in the expression,
9997 according to the relation:
9999 (ih * X / ih) * sar = output_dar
10000 X = output_dar / sar
10003 Thus the previous example needs to be modified to:
10005 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
10009 Double the output size and put the input video in the bottom-right
10010 corner of the output padded area:
10012 pad="2*iw:2*ih:ow-iw:oh-ih"
10016 @anchor{palettegen}
10017 @section palettegen
10019 Generate one palette for a whole video stream.
10021 It accepts the following options:
10025 Set the maximum number of colors to quantize in the palette.
10026 Note: the palette will still contain 256 colors; the unused palette entries
10029 @item reserve_transparent
10030 Create a palette of 255 colors maximum and reserve the last one for
10031 transparency. Reserving the transparency color is useful for GIF optimization.
10032 If not set, the maximum of colors in the palette will be 256. You probably want
10033 to disable this option for a standalone image.
10037 Set statistics mode.
10039 It accepts the following values:
10042 Compute full frame histograms.
10044 Compute histograms only for the part that differs from previous frame. This
10045 might be relevant to give more importance to the moving part of your input if
10046 the background is static.
10049 Default value is @var{full}.
10052 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10053 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10054 color quantization of the palette. This information is also visible at
10055 @var{info} logging level.
10057 @subsection Examples
10061 Generate a representative palette of a given video using @command{ffmpeg}:
10063 ffmpeg -i input.mkv -vf palettegen palette.png
10067 @section paletteuse
10069 Use a palette to downsample an input video stream.
10071 The filter takes two inputs: one video stream and a palette. The palette must
10072 be a 256 pixels image.
10074 It accepts the following options:
10078 Select dithering mode. Available algorithms are:
10081 Ordered 8x8 bayer dithering (deterministic)
10083 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10084 Note: this dithering is sometimes considered "wrong" and is included as a
10086 @item floyd_steinberg
10087 Floyd and Steingberg dithering (error diffusion)
10089 Frankie Sierra dithering v2 (error diffusion)
10091 Frankie Sierra dithering v2 "Lite" (error diffusion)
10094 Default is @var{sierra2_4a}.
10097 When @var{bayer} dithering is selected, this option defines the scale of the
10098 pattern (how much the crosshatch pattern is visible). A low value means more
10099 visible pattern for less banding, and higher value means less visible pattern
10100 at the cost of more banding.
10102 The option must be an integer value in the range [0,5]. Default is @var{2}.
10105 If set, define the zone to process
10109 Only the changing rectangle will be reprocessed. This is similar to GIF
10110 cropping/offsetting compression mechanism. This option can be useful for speed
10111 if only a part of the image is changing, and has use cases such as limiting the
10112 scope of the error diffusal @option{dither} to the rectangle that bounds the
10113 moving scene (it leads to more deterministic output if the scene doesn't change
10114 much, and as a result less moving noise and better GIF compression).
10117 Default is @var{none}.
10120 @subsection Examples
10124 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10125 using @command{ffmpeg}:
10127 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10131 @section perspective
10133 Correct perspective of video not recorded perpendicular to the screen.
10135 A description of the accepted parameters follows.
10146 Set coordinates expression for top left, top right, bottom left and bottom right corners.
10147 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10148 If the @code{sense} option is set to @code{source}, then the specified points will be sent
10149 to the corners of the destination. If the @code{sense} option is set to @code{destination},
10150 then the corners of the source will be sent to the specified coordinates.
10152 The expressions can use the following variables:
10157 the width and height of video frame.
10161 Output frame count.
10164 @item interpolation
10165 Set interpolation for perspective correction.
10167 It accepts the following values:
10173 Default value is @samp{linear}.
10176 Set interpretation of coordinate options.
10178 It accepts the following values:
10182 Send point in the source specified by the given coordinates to
10183 the corners of the destination.
10185 @item 1, destination
10187 Send the corners of the source to the point in the destination specified
10188 by the given coordinates.
10190 Default value is @samp{source}.
10194 Set when the expressions for coordinates @option{x0,y0,...x3,y3} are evaluated.
10196 It accepts the following values:
10199 only evaluate expressions once during the filter initialization or
10200 when a command is processed
10203 evaluate expressions for each incoming frame
10206 Default value is @samp{init}.
10211 Delay interlaced video by one field time so that the field order changes.
10213 The intended use is to fix PAL movies that have been captured with the
10214 opposite field order to the film-to-video transfer.
10216 A description of the accepted parameters follows.
10222 It accepts the following values:
10225 Capture field order top-first, transfer bottom-first.
10226 Filter will delay the bottom field.
10229 Capture field order bottom-first, transfer top-first.
10230 Filter will delay the top field.
10233 Capture and transfer with the same field order. This mode only exists
10234 for the documentation of the other options to refer to, but if you
10235 actually select it, the filter will faithfully do nothing.
10238 Capture field order determined automatically by field flags, transfer
10240 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10241 basis using field flags. If no field information is available,
10242 then this works just like @samp{u}.
10245 Capture unknown or varying, transfer opposite.
10246 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10247 analyzing the images and selecting the alternative that produces best
10248 match between the fields.
10251 Capture top-first, transfer unknown or varying.
10252 Filter selects among @samp{t} and @samp{p} using image analysis.
10255 Capture bottom-first, transfer unknown or varying.
10256 Filter selects among @samp{b} and @samp{p} using image analysis.
10259 Capture determined by field flags, transfer unknown or varying.
10260 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10261 image analysis. If no field information is available, then this works just
10262 like @samp{U}. This is the default mode.
10265 Both capture and transfer unknown or varying.
10266 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10270 @section pixdesctest
10272 Pixel format descriptor test filter, mainly useful for internal
10273 testing. The output video should be equal to the input video.
10277 format=monow, pixdesctest
10280 can be used to test the monowhite pixel format descriptor definition.
10284 Enable the specified chain of postprocessing subfilters using libpostproc. This
10285 library should be automatically selected with a GPL build (@code{--enable-gpl}).
10286 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
10287 Each subfilter and some options have a short and a long name that can be used
10288 interchangeably, i.e. dr/dering are the same.
10290 The filters accept the following options:
10294 Set postprocessing subfilters string.
10297 All subfilters share common options to determine their scope:
10301 Honor the quality commands for this subfilter.
10304 Do chrominance filtering, too (default).
10307 Do luminance filtering only (no chrominance).
10310 Do chrominance filtering only (no luminance).
10313 These options can be appended after the subfilter name, separated by a '|'.
10315 Available subfilters are:
10318 @item hb/hdeblock[|difference[|flatness]]
10319 Horizontal deblocking filter
10322 Difference factor where higher values mean more deblocking (default: @code{32}).
10324 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10327 @item vb/vdeblock[|difference[|flatness]]
10328 Vertical deblocking filter
10331 Difference factor where higher values mean more deblocking (default: @code{32}).
10333 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10336 @item ha/hadeblock[|difference[|flatness]]
10337 Accurate horizontal deblocking filter
10340 Difference factor where higher values mean more deblocking (default: @code{32}).
10342 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10345 @item va/vadeblock[|difference[|flatness]]
10346 Accurate vertical deblocking filter
10349 Difference factor where higher values mean more deblocking (default: @code{32}).
10351 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10355 The horizontal and vertical deblocking filters share the difference and
10356 flatness values so you cannot set different horizontal and vertical
10360 @item h1/x1hdeblock
10361 Experimental horizontal deblocking filter
10363 @item v1/x1vdeblock
10364 Experimental vertical deblocking filter
10369 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
10372 larger -> stronger filtering
10374 larger -> stronger filtering
10376 larger -> stronger filtering
10379 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
10382 Stretch luminance to @code{0-255}.
10385 @item lb/linblenddeint
10386 Linear blend deinterlacing filter that deinterlaces the given block by
10387 filtering all lines with a @code{(1 2 1)} filter.
10389 @item li/linipoldeint
10390 Linear interpolating deinterlacing filter that deinterlaces the given block by
10391 linearly interpolating every second line.
10393 @item ci/cubicipoldeint
10394 Cubic interpolating deinterlacing filter deinterlaces the given block by
10395 cubically interpolating every second line.
10397 @item md/mediandeint
10398 Median deinterlacing filter that deinterlaces the given block by applying a
10399 median filter to every second line.
10401 @item fd/ffmpegdeint
10402 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
10403 second line with a @code{(-1 4 2 4 -1)} filter.
10406 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
10407 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
10409 @item fq/forceQuant[|quantizer]
10410 Overrides the quantizer table from the input with the constant quantizer you
10418 Default pp filter combination (@code{hb|a,vb|a,dr|a})
10421 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
10424 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
10427 @subsection Examples
10431 Apply horizontal and vertical deblocking, deringing and automatic
10432 brightness/contrast:
10438 Apply default filters without brightness/contrast correction:
10444 Apply default filters and temporal denoiser:
10446 pp=default/tmpnoise|1|2|3
10450 Apply deblocking on luminance only, and switch vertical deblocking on or off
10451 automatically depending on available CPU time:
10458 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
10459 similar to spp = 6 with 7 point DCT, where only the center sample is
10462 The filter accepts the following options:
10466 Force a constant quantization parameter. It accepts an integer in range
10467 0 to 63. If not set, the filter will use the QP from the video stream
10471 Set thresholding mode. Available modes are:
10475 Set hard thresholding.
10477 Set soft thresholding (better de-ringing effect, but likely blurrier).
10479 Set medium thresholding (good results, default).
10485 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
10486 Ratio) between two input videos.
10488 This filter takes in input two input videos, the first input is
10489 considered the "main" source and is passed unchanged to the
10490 output. The second input is used as a "reference" video for computing
10493 Both video inputs must have the same resolution and pixel format for
10494 this filter to work correctly. Also it assumes that both inputs
10495 have the same number of frames, which are compared one by one.
10497 The obtained average PSNR is printed through the logging system.
10499 The filter stores the accumulated MSE (mean squared error) of each
10500 frame, and at the end of the processing it is averaged across all frames
10501 equally, and the following formula is applied to obtain the PSNR:
10504 PSNR = 10*log10(MAX^2/MSE)
10507 Where MAX is the average of the maximum values of each component of the
10510 The description of the accepted parameters follows.
10513 @item stats_file, f
10514 If specified the filter will use the named file to save the PSNR of
10515 each individual frame. When filename equals "-" the data is sent to
10518 @item stats_version
10519 Specifies which version of the stats file format to use. Details of
10520 each format are written below.
10521 Default value is 1.
10524 The file printed if @var{stats_file} is selected, contains a sequence of
10525 key/value pairs of the form @var{key}:@var{value} for each compared
10528 If a @var{stats_version} greater than 1 is specified, a header line precedes
10529 the list of per-frame-pair stats, with key value pairs following the frame
10530 format with the following parameters:
10533 @item psnr_log_version
10534 The version of the log file format. Will match @var{stats_version}.
10537 A comma separated list of the per-frame-pair parameters included in
10541 A description of each shown per-frame-pair parameter follows:
10545 sequential number of the input frame, starting from 1
10548 Mean Square Error pixel-by-pixel average difference of the compared
10549 frames, averaged over all the image components.
10551 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
10552 Mean Square Error pixel-by-pixel average difference of the compared
10553 frames for the component specified by the suffix.
10555 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
10556 Peak Signal to Noise ratio of the compared frames for the component
10557 specified by the suffix.
10562 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
10563 [main][ref] psnr="stats_file=stats.log" [out]
10566 On this example the input file being processed is compared with the
10567 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
10568 is stored in @file{stats.log}.
10573 Pulldown reversal (inverse telecine) filter, capable of handling mixed
10574 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
10577 The pullup filter is designed to take advantage of future context in making
10578 its decisions. This filter is stateless in the sense that it does not lock
10579 onto a pattern to follow, but it instead looks forward to the following
10580 fields in order to identify matches and rebuild progressive frames.
10582 To produce content with an even framerate, insert the fps filter after
10583 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
10584 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
10586 The filter accepts the following options:
10593 These options set the amount of "junk" to ignore at the left, right, top, and
10594 bottom of the image, respectively. Left and right are in units of 8 pixels,
10595 while top and bottom are in units of 2 lines.
10596 The default is 8 pixels on each side.
10599 Set the strict breaks. Setting this option to 1 will reduce the chances of
10600 filter generating an occasional mismatched frame, but it may also cause an
10601 excessive number of frames to be dropped during high motion sequences.
10602 Conversely, setting it to -1 will make filter match fields more easily.
10603 This may help processing of video where there is slight blurring between
10604 the fields, but may also cause there to be interlaced frames in the output.
10605 Default value is @code{0}.
10608 Set the metric plane to use. It accepts the following values:
10614 Use chroma blue plane.
10617 Use chroma red plane.
10620 This option may be set to use chroma plane instead of the default luma plane
10621 for doing filter's computations. This may improve accuracy on very clean
10622 source material, but more likely will decrease accuracy, especially if there
10623 is chroma noise (rainbow effect) or any grayscale video.
10624 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
10625 load and make pullup usable in realtime on slow machines.
10628 For best results (without duplicated frames in the output file) it is
10629 necessary to change the output frame rate. For example, to inverse
10630 telecine NTSC input:
10632 ffmpeg -i input -vf pullup -r 24000/1001 ...
10637 Change video quantization parameters (QP).
10639 The filter accepts the following option:
10643 Set expression for quantization parameter.
10646 The expression is evaluated through the eval API and can contain, among others,
10647 the following constants:
10651 1 if index is not 129, 0 otherwise.
10654 Sequentional index starting from -129 to 128.
10657 @subsection Examples
10661 Some equation like:
10669 Flush video frames from internal cache of frames into a random order.
10670 No frame is discarded.
10671 Inspired by @ref{frei0r} nervous filter.
10675 Set size in number of frames of internal cache, in range from @code{2} to
10676 @code{512}. Default is @code{30}.
10679 Set seed for random number generator, must be an integer included between
10680 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
10681 less than @code{0}, the filter will try to use a good random seed on a
10687 Read vertical interval timecode (VITC) information from the top lines of a
10690 The filter adds frame metadata key @code{lavfi.readvitc.tc_str} with the
10691 timecode value, if a valid timecode has been detected. Further metadata key
10692 @code{lavfi.readvitc.found} is set to 0/1 depending on whether
10693 timecode data has been found or not.
10695 This filter accepts the following options:
10699 Set the maximum number of lines to scan for VITC data. If the value is set to
10700 @code{-1} the full video frame is scanned. Default is @code{45}.
10703 Set the luma threshold for black. Accepts float numbers in the range [0.0,1.0],
10704 default value is @code{0.2}. The value must be equal or less than @code{thr_w}.
10707 Set the luma threshold for white. Accepts float numbers in the range [0.0,1.0],
10708 default value is @code{0.6}. The value must be equal or greater than @code{thr_b}.
10711 @subsection Examples
10715 Detect and draw VITC data onto the video frame; if no valid VITC is detected,
10716 draw @code{--:--:--:--} as a placeholder:
10718 ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%@{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--@}:x=(w-tw)/2:y=400-ascent'
10724 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
10726 Destination pixel at position (X, Y) will be picked from source (x, y) position
10727 where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
10728 value for pixel will be used for destination pixel.
10730 Xmap and Ymap input video streams must be of same dimensions. Output video stream
10731 will have Xmap/Ymap video stream dimensions.
10732 Xmap and Ymap input video streams are 16bit depth, single channel.
10734 @section removegrain
10736 The removegrain filter is a spatial denoiser for progressive video.
10740 Set mode for the first plane.
10743 Set mode for the second plane.
10746 Set mode for the third plane.
10749 Set mode for the fourth plane.
10752 Range of mode is from 0 to 24. Description of each mode follows:
10756 Leave input plane unchanged. Default.
10759 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
10762 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
10765 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
10768 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
10769 This is equivalent to a median filter.
10772 Line-sensitive clipping giving the minimal change.
10775 Line-sensitive clipping, intermediate.
10778 Line-sensitive clipping, intermediate.
10781 Line-sensitive clipping, intermediate.
10784 Line-sensitive clipping on a line where the neighbours pixels are the closest.
10787 Replaces the target pixel with the closest neighbour.
10790 [1 2 1] horizontal and vertical kernel blur.
10796 Bob mode, interpolates top field from the line where the neighbours
10797 pixels are the closest.
10800 Bob mode, interpolates bottom field from the line where the neighbours
10801 pixels are the closest.
10804 Bob mode, interpolates top field. Same as 13 but with a more complicated
10805 interpolation formula.
10808 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
10809 interpolation formula.
10812 Clips the pixel with the minimum and maximum of respectively the maximum and
10813 minimum of each pair of opposite neighbour pixels.
10816 Line-sensitive clipping using opposite neighbours whose greatest distance from
10817 the current pixel is minimal.
10820 Replaces the pixel with the average of its 8 neighbours.
10823 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
10826 Clips pixels using the averages of opposite neighbour.
10829 Same as mode 21 but simpler and faster.
10832 Small edge and halo removal, but reputed useless.
10838 @section removelogo
10840 Suppress a TV station logo, using an image file to determine which
10841 pixels comprise the logo. It works by filling in the pixels that
10842 comprise the logo with neighboring pixels.
10844 The filter accepts the following options:
10848 Set the filter bitmap file, which can be any image format supported by
10849 libavformat. The width and height of the image file must match those of the
10850 video stream being processed.
10853 Pixels in the provided bitmap image with a value of zero are not
10854 considered part of the logo, non-zero pixels are considered part of
10855 the logo. If you use white (255) for the logo and black (0) for the
10856 rest, you will be safe. For making the filter bitmap, it is
10857 recommended to take a screen capture of a black frame with the logo
10858 visible, and then using a threshold filter followed by the erode
10859 filter once or twice.
10861 If needed, little splotches can be fixed manually. Remember that if
10862 logo pixels are not covered, the filter quality will be much
10863 reduced. Marking too many pixels as part of the logo does not hurt as
10864 much, but it will increase the amount of blurring needed to cover over
10865 the image and will destroy more information than necessary, and extra
10866 pixels will slow things down on a large logo.
10868 @section repeatfields
10870 This filter uses the repeat_field flag from the Video ES headers and hard repeats
10871 fields based on its value.
10875 Reverse a video clip.
10877 Warning: This filter requires memory to buffer the entire clip, so trimming
10880 @subsection Examples
10884 Take the first 5 seconds of a clip, and reverse it.
10892 Rotate video by an arbitrary angle expressed in radians.
10894 The filter accepts the following options:
10896 A description of the optional parameters follows.
10899 Set an expression for the angle by which to rotate the input video
10900 clockwise, expressed as a number of radians. A negative value will
10901 result in a counter-clockwise rotation. By default it is set to "0".
10903 This expression is evaluated for each frame.
10906 Set the output width expression, default value is "iw".
10907 This expression is evaluated just once during configuration.
10910 Set the output height expression, default value is "ih".
10911 This expression is evaluated just once during configuration.
10914 Enable bilinear interpolation if set to 1, a value of 0 disables
10915 it. Default value is 1.
10918 Set the color used to fill the output area not covered by the rotated
10919 image. For the general syntax of this option, check the "Color" section in the
10920 ffmpeg-utils manual. If the special value "none" is selected then no
10921 background is printed (useful for example if the background is never shown).
10923 Default value is "black".
10926 The expressions for the angle and the output size can contain the
10927 following constants and functions:
10931 sequential number of the input frame, starting from 0. It is always NAN
10932 before the first frame is filtered.
10935 time in seconds of the input frame, it is set to 0 when the filter is
10936 configured. It is always NAN before the first frame is filtered.
10940 horizontal and vertical chroma subsample values. For example for the
10941 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10945 the input video width and height
10949 the output width and height, that is the size of the padded area as
10950 specified by the @var{width} and @var{height} expressions
10954 the minimal width/height required for completely containing the input
10955 video rotated by @var{a} radians.
10957 These are only available when computing the @option{out_w} and
10958 @option{out_h} expressions.
10961 @subsection Examples
10965 Rotate the input by PI/6 radians clockwise:
10971 Rotate the input by PI/6 radians counter-clockwise:
10977 Rotate the input by 45 degrees clockwise:
10983 Apply a constant rotation with period T, starting from an angle of PI/3:
10985 rotate=PI/3+2*PI*t/T
10989 Make the input video rotation oscillating with a period of T
10990 seconds and an amplitude of A radians:
10992 rotate=A*sin(2*PI/T*t)
10996 Rotate the video, output size is chosen so that the whole rotating
10997 input video is always completely contained in the output:
10999 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
11003 Rotate the video, reduce the output size so that no background is ever
11006 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
11010 @subsection Commands
11012 The filter supports the following commands:
11016 Set the angle expression.
11017 The command accepts the same syntax of the corresponding option.
11019 If the specified expression is not valid, it is kept at its current
11025 Apply Shape Adaptive Blur.
11027 The filter accepts the following options:
11030 @item luma_radius, lr
11031 Set luma blur filter strength, must be a value in range 0.1-4.0, default
11032 value is 1.0. A greater value will result in a more blurred image, and
11033 in slower processing.
11035 @item luma_pre_filter_radius, lpfr
11036 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
11039 @item luma_strength, ls
11040 Set luma maximum difference between pixels to still be considered, must
11041 be a value in the 0.1-100.0 range, default value is 1.0.
11043 @item chroma_radius, cr
11044 Set chroma blur filter strength, must be a value in range -0.9-4.0. A
11045 greater value will result in a more blurred image, and in slower
11048 @item chroma_pre_filter_radius, cpfr
11049 Set chroma pre-filter radius, must be a value in the -0.9-2.0 range.
11051 @item chroma_strength, cs
11052 Set chroma maximum difference between pixels to still be considered,
11053 must be a value in the -0.9-100.0 range.
11056 Each chroma option value, if not explicitly specified, is set to the
11057 corresponding luma option value.
11062 Scale (resize) the input video, using the libswscale library.
11064 The scale filter forces the output display aspect ratio to be the same
11065 of the input, by changing the output sample aspect ratio.
11067 If the input image format is different from the format requested by
11068 the next filter, the scale filter will convert the input to the
11071 @subsection Options
11072 The filter accepts the following options, or any of the options
11073 supported by the libswscale scaler.
11075 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
11076 the complete list of scaler options.
11081 Set the output video dimension expression. Default value is the input
11084 If the value is 0, the input width is used for the output.
11086 If one of the values is -1, the scale filter will use a value that
11087 maintains the aspect ratio of the input image, calculated from the
11088 other specified dimension. If both of them are -1, the input size is
11091 If one of the values is -n with n > 1, the scale filter will also use a value
11092 that maintains the aspect ratio of the input image, calculated from the other
11093 specified dimension. After that it will, however, make sure that the calculated
11094 dimension is divisible by n and adjust the value if necessary.
11096 See below for the list of accepted constants for use in the dimension
11100 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
11104 Only evaluate expressions once during the filter initialization or when a command is processed.
11107 Evaluate expressions for each incoming frame.
11111 Default value is @samp{init}.
11115 Set the interlacing mode. It accepts the following values:
11119 Force interlaced aware scaling.
11122 Do not apply interlaced scaling.
11125 Select interlaced aware scaling depending on whether the source frames
11126 are flagged as interlaced or not.
11129 Default value is @samp{0}.
11132 Set libswscale scaling flags. See
11133 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11134 complete list of values. If not explicitly specified the filter applies
11138 @item param0, param1
11139 Set libswscale input parameters for scaling algorithms that need them. See
11140 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11141 complete documentation. If not explicitly specified the filter applies
11147 Set the video size. For the syntax of this option, check the
11148 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11150 @item in_color_matrix
11151 @item out_color_matrix
11152 Set in/output YCbCr color space type.
11154 This allows the autodetected value to be overridden as well as allows forcing
11155 a specific value used for the output and encoder.
11157 If not specified, the color space type depends on the pixel format.
11163 Choose automatically.
11166 Format conforming to International Telecommunication Union (ITU)
11167 Recommendation BT.709.
11170 Set color space conforming to the United States Federal Communications
11171 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
11174 Set color space conforming to:
11178 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
11181 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
11184 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
11189 Set color space conforming to SMPTE ST 240:1999.
11194 Set in/output YCbCr sample range.
11196 This allows the autodetected value to be overridden as well as allows forcing
11197 a specific value used for the output and encoder. If not specified, the
11198 range depends on the pixel format. Possible values:
11202 Choose automatically.
11205 Set full range (0-255 in case of 8-bit luma).
11208 Set "MPEG" range (16-235 in case of 8-bit luma).
11211 @item force_original_aspect_ratio
11212 Enable decreasing or increasing output video width or height if necessary to
11213 keep the original aspect ratio. Possible values:
11217 Scale the video as specified and disable this feature.
11220 The output video dimensions will automatically be decreased if needed.
11223 The output video dimensions will automatically be increased if needed.
11227 One useful instance of this option is that when you know a specific device's
11228 maximum allowed resolution, you can use this to limit the output video to
11229 that, while retaining the aspect ratio. For example, device A allows
11230 1280x720 playback, and your video is 1920x800. Using this option (set it to
11231 decrease) and specifying 1280x720 to the command line makes the output
11234 Please note that this is a different thing than specifying -1 for @option{w}
11235 or @option{h}, you still need to specify the output resolution for this option
11240 The values of the @option{w} and @option{h} options are expressions
11241 containing the following constants:
11246 The input width and height
11250 These are the same as @var{in_w} and @var{in_h}.
11254 The output (scaled) width and height
11258 These are the same as @var{out_w} and @var{out_h}
11261 The same as @var{iw} / @var{ih}
11264 input sample aspect ratio
11267 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
11271 horizontal and vertical input chroma subsample values. For example for the
11272 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11276 horizontal and vertical output chroma subsample values. For example for the
11277 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11280 @subsection Examples
11284 Scale the input video to a size of 200x100
11289 This is equivalent to:
11300 Specify a size abbreviation for the output size:
11305 which can also be written as:
11311 Scale the input to 2x:
11313 scale=w=2*iw:h=2*ih
11317 The above is the same as:
11319 scale=2*in_w:2*in_h
11323 Scale the input to 2x with forced interlaced scaling:
11325 scale=2*iw:2*ih:interl=1
11329 Scale the input to half size:
11331 scale=w=iw/2:h=ih/2
11335 Increase the width, and set the height to the same size:
11341 Seek Greek harmony:
11348 Increase the height, and set the width to 3/2 of the height:
11350 scale=w=3/2*oh:h=3/5*ih
11354 Increase the size, making the size a multiple of the chroma
11357 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
11361 Increase the width to a maximum of 500 pixels,
11362 keeping the same aspect ratio as the input:
11364 scale=w='min(500\, iw*3/2):h=-1'
11368 @subsection Commands
11370 This filter supports the following commands:
11374 Set the output video dimension expression.
11375 The command accepts the same syntax of the corresponding option.
11377 If the specified expression is not valid, it is kept at its current
11383 Use the NVIDIA Performance Primitives (libnpp) to perform scaling and/or pixel
11384 format conversion on CUDA video frames. Setting the output width and height
11385 works in the same way as for the @var{scale} filter.
11387 The following additional options are accepted:
11390 The pixel format of the output CUDA frames. If set to the string "same" (the
11391 default), the input format will be kept. Note that automatic format negotiation
11392 and conversion is not yet supported for hardware frames
11395 The interpolation algorithm used for resizing. One of the following:
11402 @item cubic2p_bspline
11403 2-parameter cubic (B=1, C=0)
11405 @item cubic2p_catmullrom
11406 2-parameter cubic (B=0, C=1/2)
11408 @item cubic2p_b05c03
11409 2-parameter cubic (B=1/2, C=3/10)
11421 Scale (resize) the input video, based on a reference video.
11423 See the scale filter for available options, scale2ref supports the same but
11424 uses the reference video instead of the main input as basis.
11426 @subsection Examples
11430 Scale a subtitle stream to match the main video in size before overlaying
11432 'scale2ref[b][a];[a][b]overlay'
11436 @anchor{selectivecolor}
11437 @section selectivecolor
11439 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
11440 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
11441 by the "purity" of the color (that is, how saturated it already is).
11443 This filter is similar to the Adobe Photoshop Selective Color tool.
11445 The filter accepts the following options:
11448 @item correction_method
11449 Select color correction method.
11451 Available values are:
11454 Specified adjustments are applied "as-is" (added/subtracted to original pixel
11457 Specified adjustments are relative to the original component value.
11459 Default is @code{absolute}.
11461 Adjustments for red pixels (pixels where the red component is the maximum)
11463 Adjustments for yellow pixels (pixels where the blue component is the minimum)
11465 Adjustments for green pixels (pixels where the green component is the maximum)
11467 Adjustments for cyan pixels (pixels where the red component is the minimum)
11469 Adjustments for blue pixels (pixels where the blue component is the maximum)
11471 Adjustments for magenta pixels (pixels where the green component is the minimum)
11473 Adjustments for white pixels (pixels where all components are greater than 128)
11475 Adjustments for all pixels except pure black and pure white
11477 Adjustments for black pixels (pixels where all components are lesser than 128)
11479 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
11482 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
11483 4 space separated floating point adjustment values in the [-1,1] range,
11484 respectively to adjust the amount of cyan, magenta, yellow and black for the
11485 pixels of its range.
11487 @subsection Examples
11491 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
11492 increase magenta by 27% in blue areas:
11494 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
11498 Use a Photoshop selective color preset:
11500 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
11504 @section separatefields
11506 The @code{separatefields} takes a frame-based video input and splits
11507 each frame into its components fields, producing a new half height clip
11508 with twice the frame rate and twice the frame count.
11510 This filter use field-dominance information in frame to decide which
11511 of each pair of fields to place first in the output.
11512 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
11514 @section setdar, setsar
11516 The @code{setdar} filter sets the Display Aspect Ratio for the filter
11519 This is done by changing the specified Sample (aka Pixel) Aspect
11520 Ratio, according to the following equation:
11522 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
11525 Keep in mind that the @code{setdar} filter does not modify the pixel
11526 dimensions of the video frame. Also, the display aspect ratio set by
11527 this filter may be changed by later filters in the filterchain,
11528 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
11531 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
11532 the filter output video.
11534 Note that as a consequence of the application of this filter, the
11535 output display aspect ratio will change according to the equation
11538 Keep in mind that the sample aspect ratio set by the @code{setsar}
11539 filter may be changed by later filters in the filterchain, e.g. if
11540 another "setsar" or a "setdar" filter is applied.
11542 It accepts the following parameters:
11545 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
11546 Set the aspect ratio used by the filter.
11548 The parameter can be a floating point number string, an expression, or
11549 a string of the form @var{num}:@var{den}, where @var{num} and
11550 @var{den} are the numerator and denominator of the aspect ratio. If
11551 the parameter is not specified, it is assumed the value "0".
11552 In case the form "@var{num}:@var{den}" is used, the @code{:} character
11556 Set the maximum integer value to use for expressing numerator and
11557 denominator when reducing the expressed aspect ratio to a rational.
11558 Default value is @code{100}.
11562 The parameter @var{sar} is an expression containing
11563 the following constants:
11567 These are approximated values for the mathematical constants e
11568 (Euler's number), pi (Greek pi), and phi (the golden ratio).
11571 The input width and height.
11574 These are the same as @var{w} / @var{h}.
11577 The input sample aspect ratio.
11580 The input display aspect ratio. It is the same as
11581 (@var{w} / @var{h}) * @var{sar}.
11584 Horizontal and vertical chroma subsample values. For example, for the
11585 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11588 @subsection Examples
11593 To change the display aspect ratio to 16:9, specify one of the following:
11600 To change the sample aspect ratio to 10:11, specify:
11606 To set a display aspect ratio of 16:9, and specify a maximum integer value of
11607 1000 in the aspect ratio reduction, use the command:
11609 setdar=ratio=16/9:max=1000
11617 Force field for the output video frame.
11619 The @code{setfield} filter marks the interlace type field for the
11620 output frames. It does not change the input frame, but only sets the
11621 corresponding property, which affects how the frame is treated by
11622 following filters (e.g. @code{fieldorder} or @code{yadif}).
11624 The filter accepts the following options:
11629 Available values are:
11633 Keep the same field property.
11636 Mark the frame as bottom-field-first.
11639 Mark the frame as top-field-first.
11642 Mark the frame as progressive.
11648 Show a line containing various information for each input video frame.
11649 The input video is not modified.
11651 The shown line contains a sequence of key/value pairs of the form
11652 @var{key}:@var{value}.
11654 The following values are shown in the output:
11658 The (sequential) number of the input frame, starting from 0.
11661 The Presentation TimeStamp of the input frame, expressed as a number of
11662 time base units. The time base unit depends on the filter input pad.
11665 The Presentation TimeStamp of the input frame, expressed as a number of
11669 The position of the frame in the input stream, or -1 if this information is
11670 unavailable and/or meaningless (for example in case of synthetic video).
11673 The pixel format name.
11676 The sample aspect ratio of the input frame, expressed in the form
11677 @var{num}/@var{den}.
11680 The size of the input frame. For the syntax of this option, check the
11681 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11684 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
11685 for bottom field first).
11688 This is 1 if the frame is a key frame, 0 otherwise.
11691 The picture type of the input frame ("I" for an I-frame, "P" for a
11692 P-frame, "B" for a B-frame, or "?" for an unknown type).
11693 Also refer to the documentation of the @code{AVPictureType} enum and of
11694 the @code{av_get_picture_type_char} function defined in
11695 @file{libavutil/avutil.h}.
11698 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
11700 @item plane_checksum
11701 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
11702 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
11705 @section showpalette
11707 Displays the 256 colors palette of each frame. This filter is only relevant for
11708 @var{pal8} pixel format frames.
11710 It accepts the following option:
11714 Set the size of the box used to represent one palette color entry. Default is
11715 @code{30} (for a @code{30x30} pixel box).
11718 @section shuffleframes
11720 Reorder and/or duplicate video frames.
11722 It accepts the following parameters:
11726 Set the destination indexes of input frames.
11727 This is space or '|' separated list of indexes that maps input frames to output
11728 frames. Number of indexes also sets maximal value that each index may have.
11731 The first frame has the index 0. The default is to keep the input unchanged.
11733 Swap second and third frame of every three frames of the input:
11735 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
11738 @section shuffleplanes
11740 Reorder and/or duplicate video planes.
11742 It accepts the following parameters:
11747 The index of the input plane to be used as the first output plane.
11750 The index of the input plane to be used as the second output plane.
11753 The index of the input plane to be used as the third output plane.
11756 The index of the input plane to be used as the fourth output plane.
11760 The first plane has the index 0. The default is to keep the input unchanged.
11762 Swap the second and third planes of the input:
11764 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
11767 @anchor{signalstats}
11768 @section signalstats
11769 Evaluate various visual metrics that assist in determining issues associated
11770 with the digitization of analog video media.
11772 By default the filter will log these metadata values:
11776 Display the minimal Y value contained within the input frame. Expressed in
11780 Display the Y value at the 10% percentile within the input frame. Expressed in
11784 Display the average Y value within the input frame. Expressed in range of
11788 Display the Y value at the 90% percentile within the input frame. Expressed in
11792 Display the maximum Y value contained within the input frame. Expressed in
11796 Display the minimal U value contained within the input frame. Expressed in
11800 Display the U value at the 10% percentile within the input frame. Expressed in
11804 Display the average U value within the input frame. Expressed in range of
11808 Display the U value at the 90% percentile within the input frame. Expressed in
11812 Display the maximum U value contained within the input frame. Expressed in
11816 Display the minimal V value contained within the input frame. Expressed in
11820 Display the V value at the 10% percentile within the input frame. Expressed in
11824 Display the average V value within the input frame. Expressed in range of
11828 Display the V value at the 90% percentile within the input frame. Expressed in
11832 Display the maximum V value contained within the input frame. Expressed in
11836 Display the minimal saturation value contained within the input frame.
11837 Expressed in range of [0-~181.02].
11840 Display the saturation value at the 10% percentile within the input frame.
11841 Expressed in range of [0-~181.02].
11844 Display the average saturation value within the input frame. Expressed in range
11848 Display the saturation value at the 90% percentile within the input frame.
11849 Expressed in range of [0-~181.02].
11852 Display the maximum saturation value contained within the input frame.
11853 Expressed in range of [0-~181.02].
11856 Display the median value for hue within the input frame. Expressed in range of
11860 Display the average value for hue within the input frame. Expressed in range of
11864 Display the average of sample value difference between all values of the Y
11865 plane in the current frame and corresponding values of the previous input frame.
11866 Expressed in range of [0-255].
11869 Display the average of sample value difference between all values of the U
11870 plane in the current frame and corresponding values of the previous input frame.
11871 Expressed in range of [0-255].
11874 Display the average of sample value difference between all values of the V
11875 plane in the current frame and corresponding values of the previous input frame.
11876 Expressed in range of [0-255].
11879 The filter accepts the following options:
11885 @option{stat} specify an additional form of image analysis.
11886 @option{out} output video with the specified type of pixel highlighted.
11888 Both options accept the following values:
11892 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
11893 unlike the neighboring pixels of the same field. Examples of temporal outliers
11894 include the results of video dropouts, head clogs, or tape tracking issues.
11897 Identify @var{vertical line repetition}. Vertical line repetition includes
11898 similar rows of pixels within a frame. In born-digital video vertical line
11899 repetition is common, but this pattern is uncommon in video digitized from an
11900 analog source. When it occurs in video that results from the digitization of an
11901 analog source it can indicate concealment from a dropout compensator.
11904 Identify pixels that fall outside of legal broadcast range.
11908 Set the highlight color for the @option{out} option. The default color is
11912 @subsection Examples
11916 Output data of various video metrics:
11918 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
11922 Output specific data about the minimum and maximum values of the Y plane per frame:
11924 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
11928 Playback video while highlighting pixels that are outside of broadcast range in red.
11930 ffplay example.mov -vf signalstats="out=brng:color=red"
11934 Playback video with signalstats metadata drawn over the frame.
11936 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
11939 The contents of signalstat_drawtext.txt used in the command are:
11942 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
11943 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
11944 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
11945 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
11953 Blur the input video without impacting the outlines.
11955 It accepts the following options:
11958 @item luma_radius, lr
11959 Set the luma radius. The option value must be a float number in
11960 the range [0.1,5.0] that specifies the variance of the gaussian filter
11961 used to blur the image (slower if larger). Default value is 1.0.
11963 @item luma_strength, ls
11964 Set the luma strength. The option value must be a float number
11965 in the range [-1.0,1.0] that configures the blurring. A value included
11966 in [0.0,1.0] will blur the image whereas a value included in
11967 [-1.0,0.0] will sharpen the image. Default value is 1.0.
11969 @item luma_threshold, lt
11970 Set the luma threshold used as a coefficient to determine
11971 whether a pixel should be blurred or not. The option value must be an
11972 integer in the range [-30,30]. A value of 0 will filter all the image,
11973 a value included in [0,30] will filter flat areas and a value included
11974 in [-30,0] will filter edges. Default value is 0.
11976 @item chroma_radius, cr
11977 Set the chroma radius. The option value must be a float number in
11978 the range [0.1,5.0] that specifies the variance of the gaussian filter
11979 used to blur the image (slower if larger). Default value is 1.0.
11981 @item chroma_strength, cs
11982 Set the chroma strength. The option value must be a float number
11983 in the range [-1.0,1.0] that configures the blurring. A value included
11984 in [0.0,1.0] will blur the image whereas a value included in
11985 [-1.0,0.0] will sharpen the image. Default value is 1.0.
11987 @item chroma_threshold, ct
11988 Set the chroma threshold used as a coefficient to determine
11989 whether a pixel should be blurred or not. The option value must be an
11990 integer in the range [-30,30]. A value of 0 will filter all the image,
11991 a value included in [0,30] will filter flat areas and a value included
11992 in [-30,0] will filter edges. Default value is 0.
11995 If a chroma option is not explicitly set, the corresponding luma value
12000 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
12002 This filter takes in input two input videos, the first input is
12003 considered the "main" source and is passed unchanged to the
12004 output. The second input is used as a "reference" video for computing
12007 Both video inputs must have the same resolution and pixel format for
12008 this filter to work correctly. Also it assumes that both inputs
12009 have the same number of frames, which are compared one by one.
12011 The filter stores the calculated SSIM of each frame.
12013 The description of the accepted parameters follows.
12016 @item stats_file, f
12017 If specified the filter will use the named file to save the SSIM of
12018 each individual frame. When filename equals "-" the data is sent to
12022 The file printed if @var{stats_file} is selected, contains a sequence of
12023 key/value pairs of the form @var{key}:@var{value} for each compared
12026 A description of each shown parameter follows:
12030 sequential number of the input frame, starting from 1
12032 @item Y, U, V, R, G, B
12033 SSIM of the compared frames for the component specified by the suffix.
12036 SSIM of the compared frames for the whole frame.
12039 Same as above but in dB representation.
12044 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
12045 [main][ref] ssim="stats_file=stats.log" [out]
12048 On this example the input file being processed is compared with the
12049 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
12050 is stored in @file{stats.log}.
12052 Another example with both psnr and ssim at same time:
12054 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
12059 Convert between different stereoscopic image formats.
12061 The filters accept the following options:
12065 Set stereoscopic image format of input.
12067 Available values for input image formats are:
12070 side by side parallel (left eye left, right eye right)
12073 side by side crosseye (right eye left, left eye right)
12076 side by side parallel with half width resolution
12077 (left eye left, right eye right)
12080 side by side crosseye with half width resolution
12081 (right eye left, left eye right)
12084 above-below (left eye above, right eye below)
12087 above-below (right eye above, left eye below)
12090 above-below with half height resolution
12091 (left eye above, right eye below)
12094 above-below with half height resolution
12095 (right eye above, left eye below)
12098 alternating frames (left eye first, right eye second)
12101 alternating frames (right eye first, left eye second)
12104 interleaved rows (left eye has top row, right eye starts on next row)
12107 interleaved rows (right eye has top row, left eye starts on next row)
12110 interleaved columns, left eye first
12113 interleaved columns, right eye first
12115 Default value is @samp{sbsl}.
12119 Set stereoscopic image format of output.
12123 side by side parallel (left eye left, right eye right)
12126 side by side crosseye (right eye left, left eye right)
12129 side by side parallel with half width resolution
12130 (left eye left, right eye right)
12133 side by side crosseye with half width resolution
12134 (right eye left, left eye right)
12137 above-below (left eye above, right eye below)
12140 above-below (right eye above, left eye below)
12143 above-below with half height resolution
12144 (left eye above, right eye below)
12147 above-below with half height resolution
12148 (right eye above, left eye below)
12151 alternating frames (left eye first, right eye second)
12154 alternating frames (right eye first, left eye second)
12157 interleaved rows (left eye has top row, right eye starts on next row)
12160 interleaved rows (right eye has top row, left eye starts on next row)
12163 anaglyph red/blue gray
12164 (red filter on left eye, blue filter on right eye)
12167 anaglyph red/green gray
12168 (red filter on left eye, green filter on right eye)
12171 anaglyph red/cyan gray
12172 (red filter on left eye, cyan filter on right eye)
12175 anaglyph red/cyan half colored
12176 (red filter on left eye, cyan filter on right eye)
12179 anaglyph red/cyan color
12180 (red filter on left eye, cyan filter on right eye)
12183 anaglyph red/cyan color optimized with the least squares projection of dubois
12184 (red filter on left eye, cyan filter on right eye)
12187 anaglyph green/magenta gray
12188 (green filter on left eye, magenta filter on right eye)
12191 anaglyph green/magenta half colored
12192 (green filter on left eye, magenta filter on right eye)
12195 anaglyph green/magenta colored
12196 (green filter on left eye, magenta filter on right eye)
12199 anaglyph green/magenta color optimized with the least squares projection of dubois
12200 (green filter on left eye, magenta filter on right eye)
12203 anaglyph yellow/blue gray
12204 (yellow filter on left eye, blue filter on right eye)
12207 anaglyph yellow/blue half colored
12208 (yellow filter on left eye, blue filter on right eye)
12211 anaglyph yellow/blue colored
12212 (yellow filter on left eye, blue filter on right eye)
12215 anaglyph yellow/blue color optimized with the least squares projection of dubois
12216 (yellow filter on left eye, blue filter on right eye)
12219 mono output (left eye only)
12222 mono output (right eye only)
12225 checkerboard, left eye first
12228 checkerboard, right eye first
12231 interleaved columns, left eye first
12234 interleaved columns, right eye first
12240 Default value is @samp{arcd}.
12243 @subsection Examples
12247 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
12253 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
12259 @section streamselect, astreamselect
12260 Select video or audio streams.
12262 The filter accepts the following options:
12266 Set number of inputs. Default is 2.
12269 Set input indexes to remap to outputs.
12272 @subsection Commands
12274 The @code{streamselect} and @code{astreamselect} filter supports the following
12279 Set input indexes to remap to outputs.
12282 @subsection Examples
12286 Select first 5 seconds 1st stream and rest of time 2nd stream:
12288 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
12292 Same as above, but for audio:
12294 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
12301 Apply a simple postprocessing filter that compresses and decompresses the image
12302 at several (or - in the case of @option{quality} level @code{6} - all) shifts
12303 and average the results.
12305 The filter accepts the following options:
12309 Set quality. This option defines the number of levels for averaging. It accepts
12310 an integer in the range 0-6. If set to @code{0}, the filter will have no
12311 effect. A value of @code{6} means the higher quality. For each increment of
12312 that value the speed drops by a factor of approximately 2. Default value is
12316 Force a constant quantization parameter. If not set, the filter will use the QP
12317 from the video stream (if available).
12320 Set thresholding mode. Available modes are:
12324 Set hard thresholding (default).
12326 Set soft thresholding (better de-ringing effect, but likely blurrier).
12329 @item use_bframe_qp
12330 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
12331 option may cause flicker since the B-Frames have often larger QP. Default is
12332 @code{0} (not enabled).
12338 Draw subtitles on top of input video using the libass library.
12340 To enable compilation of this filter you need to configure FFmpeg with
12341 @code{--enable-libass}. This filter also requires a build with libavcodec and
12342 libavformat to convert the passed subtitles file to ASS (Advanced Substation
12343 Alpha) subtitles format.
12345 The filter accepts the following options:
12349 Set the filename of the subtitle file to read. It must be specified.
12351 @item original_size
12352 Specify the size of the original video, the video for which the ASS file
12353 was composed. For the syntax of this option, check the
12354 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12355 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
12356 correctly scale the fonts if the aspect ratio has been changed.
12359 Set a directory path containing fonts that can be used by the filter.
12360 These fonts will be used in addition to whatever the font provider uses.
12363 Set subtitles input character encoding. @code{subtitles} filter only. Only
12364 useful if not UTF-8.
12366 @item stream_index, si
12367 Set subtitles stream index. @code{subtitles} filter only.
12370 Override default style or script info parameters of the subtitles. It accepts a
12371 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
12374 If the first key is not specified, it is assumed that the first value
12375 specifies the @option{filename}.
12377 For example, to render the file @file{sub.srt} on top of the input
12378 video, use the command:
12383 which is equivalent to:
12385 subtitles=filename=sub.srt
12388 To render the default subtitles stream from file @file{video.mkv}, use:
12390 subtitles=video.mkv
12393 To render the second subtitles stream from that file, use:
12395 subtitles=video.mkv:si=1
12398 To make the subtitles stream from @file{sub.srt} appear in transparent green
12399 @code{DejaVu Serif}, use:
12401 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
12404 @section super2xsai
12406 Scale the input by 2x and smooth using the Super2xSaI (Scale and
12407 Interpolate) pixel art scaling algorithm.
12409 Useful for enlarging pixel art images without reducing sharpness.
12413 Swap two rectangular objects in video.
12415 This filter accepts the following options:
12425 Set 1st rect x coordinate.
12428 Set 1st rect y coordinate.
12431 Set 2nd rect x coordinate.
12434 Set 2nd rect y coordinate.
12436 All expressions are evaluated once for each frame.
12439 The all options are expressions containing the following constants:
12444 The input width and height.
12447 same as @var{w} / @var{h}
12450 input sample aspect ratio
12453 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
12456 The number of the input frame, starting from 0.
12459 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
12462 the position in the file of the input frame, NAN if unknown
12470 Apply telecine process to the video.
12472 This filter accepts the following options:
12481 The default value is @code{top}.
12485 A string of numbers representing the pulldown pattern you wish to apply.
12486 The default value is @code{23}.
12490 Some typical patterns:
12495 24p: 2332 (preferred)
12502 24p: 222222222223 ("Euro pulldown")
12508 Select the most representative frame in a given sequence of consecutive frames.
12510 The filter accepts the following options:
12514 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
12515 will pick one of them, and then handle the next batch of @var{n} frames until
12516 the end. Default is @code{100}.
12519 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
12520 value will result in a higher memory usage, so a high value is not recommended.
12522 @subsection Examples
12526 Extract one picture each 50 frames:
12532 Complete example of a thumbnail creation with @command{ffmpeg}:
12534 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
12540 Tile several successive frames together.
12542 The filter accepts the following options:
12547 Set the grid size (i.e. the number of lines and columns). For the syntax of
12548 this option, check the
12549 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12552 Set the maximum number of frames to render in the given area. It must be less
12553 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
12554 the area will be used.
12557 Set the outer border margin in pixels.
12560 Set the inner border thickness (i.e. the number of pixels between frames). For
12561 more advanced padding options (such as having different values for the edges),
12562 refer to the pad video filter.
12565 Specify the color of the unused area. For the syntax of this option, check the
12566 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
12570 @subsection Examples
12574 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
12576 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
12578 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
12579 duplicating each output frame to accommodate the originally detected frame
12583 Display @code{5} pictures in an area of @code{3x2} frames,
12584 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
12585 mixed flat and named options:
12587 tile=3x2:nb_frames=5:padding=7:margin=2
12591 @section tinterlace
12593 Perform various types of temporal field interlacing.
12595 Frames are counted starting from 1, so the first input frame is
12598 The filter accepts the following options:
12603 Specify the mode of the interlacing. This option can also be specified
12604 as a value alone. See below for a list of values for this option.
12606 Available values are:
12610 Move odd frames into the upper field, even into the lower field,
12611 generating a double height frame at half frame rate.
12615 Frame 1 Frame 2 Frame 3 Frame 4
12617 11111 22222 33333 44444
12618 11111 22222 33333 44444
12619 11111 22222 33333 44444
12620 11111 22222 33333 44444
12634 Only output odd frames, even frames are dropped, generating a frame with
12635 unchanged height at half frame rate.
12640 Frame 1 Frame 2 Frame 3 Frame 4
12642 11111 22222 33333 44444
12643 11111 22222 33333 44444
12644 11111 22222 33333 44444
12645 11111 22222 33333 44444
12655 Only output even frames, odd frames are dropped, generating a frame with
12656 unchanged height at half frame rate.
12661 Frame 1 Frame 2 Frame 3 Frame 4
12663 11111 22222 33333 44444
12664 11111 22222 33333 44444
12665 11111 22222 33333 44444
12666 11111 22222 33333 44444
12676 Expand each frame to full height, but pad alternate lines with black,
12677 generating a frame with double height at the same input frame rate.
12682 Frame 1 Frame 2 Frame 3 Frame 4
12684 11111 22222 33333 44444
12685 11111 22222 33333 44444
12686 11111 22222 33333 44444
12687 11111 22222 33333 44444
12690 11111 ..... 33333 .....
12691 ..... 22222 ..... 44444
12692 11111 ..... 33333 .....
12693 ..... 22222 ..... 44444
12694 11111 ..... 33333 .....
12695 ..... 22222 ..... 44444
12696 11111 ..... 33333 .....
12697 ..... 22222 ..... 44444
12701 @item interleave_top, 4
12702 Interleave the upper field from odd frames with the lower field from
12703 even frames, generating a frame with unchanged height at half frame rate.
12708 Frame 1 Frame 2 Frame 3 Frame 4
12710 11111<- 22222 33333<- 44444
12711 11111 22222<- 33333 44444<-
12712 11111<- 22222 33333<- 44444
12713 11111 22222<- 33333 44444<-
12723 @item interleave_bottom, 5
12724 Interleave the lower field from odd frames with the upper field from
12725 even frames, generating a frame with unchanged height at half frame rate.
12730 Frame 1 Frame 2 Frame 3 Frame 4
12732 11111 22222<- 33333 44444<-
12733 11111<- 22222 33333<- 44444
12734 11111 22222<- 33333 44444<-
12735 11111<- 22222 33333<- 44444
12745 @item interlacex2, 6
12746 Double frame rate with unchanged height. Frames are inserted each
12747 containing the second temporal field from the previous input frame and
12748 the first temporal field from the next input frame. This mode relies on
12749 the top_field_first flag. Useful for interlaced video displays with no
12750 field synchronisation.
12755 Frame 1 Frame 2 Frame 3 Frame 4
12757 11111 22222 33333 44444
12758 11111 22222 33333 44444
12759 11111 22222 33333 44444
12760 11111 22222 33333 44444
12763 11111 22222 22222 33333 33333 44444 44444
12764 11111 11111 22222 22222 33333 33333 44444
12765 11111 22222 22222 33333 33333 44444 44444
12766 11111 11111 22222 22222 33333 33333 44444
12771 Move odd frames into the upper field, even into the lower field,
12772 generating a double height frame at same frame rate.
12777 Frame 1 Frame 2 Frame 3 Frame 4
12779 11111 22222 33333 44444
12780 11111 22222 33333 44444
12781 11111 22222 33333 44444
12782 11111 22222 33333 44444
12785 11111 33333 33333 55555
12786 22222 22222 44444 44444
12787 11111 33333 33333 55555
12788 22222 22222 44444 44444
12789 11111 33333 33333 55555
12790 22222 22222 44444 44444
12791 11111 33333 33333 55555
12792 22222 22222 44444 44444
12797 Numeric values are deprecated but are accepted for backward
12798 compatibility reasons.
12800 Default mode is @code{merge}.
12803 Specify flags influencing the filter process.
12805 Available value for @var{flags} is:
12808 @item low_pass_filter, vlfp
12809 Enable vertical low-pass filtering in the filter.
12810 Vertical low-pass filtering is required when creating an interlaced
12811 destination from a progressive source which contains high-frequency
12812 vertical detail. Filtering will reduce interlace 'twitter' and Moire
12815 Vertical low-pass filtering can only be enabled for @option{mode}
12816 @var{interleave_top} and @var{interleave_bottom}.
12823 Transpose rows with columns in the input video and optionally flip it.
12825 It accepts the following parameters:
12830 Specify the transposition direction.
12832 Can assume the following values:
12834 @item 0, 4, cclock_flip
12835 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
12843 Rotate by 90 degrees clockwise, that is:
12851 Rotate by 90 degrees counterclockwise, that is:
12858 @item 3, 7, clock_flip
12859 Rotate by 90 degrees clockwise and vertically flip, that is:
12867 For values between 4-7, the transposition is only done if the input
12868 video geometry is portrait and not landscape. These values are
12869 deprecated, the @code{passthrough} option should be used instead.
12871 Numerical values are deprecated, and should be dropped in favor of
12872 symbolic constants.
12875 Do not apply the transposition if the input geometry matches the one
12876 specified by the specified value. It accepts the following values:
12879 Always apply transposition.
12881 Preserve portrait geometry (when @var{height} >= @var{width}).
12883 Preserve landscape geometry (when @var{width} >= @var{height}).
12886 Default value is @code{none}.
12889 For example to rotate by 90 degrees clockwise and preserve portrait
12892 transpose=dir=1:passthrough=portrait
12895 The command above can also be specified as:
12897 transpose=1:portrait
12901 Trim the input so that the output contains one continuous subpart of the input.
12903 It accepts the following parameters:
12906 Specify the time of the start of the kept section, i.e. the frame with the
12907 timestamp @var{start} will be the first frame in the output.
12910 Specify the time of the first frame that will be dropped, i.e. the frame
12911 immediately preceding the one with the timestamp @var{end} will be the last
12912 frame in the output.
12915 This is the same as @var{start}, except this option sets the start timestamp
12916 in timebase units instead of seconds.
12919 This is the same as @var{end}, except this option sets the end timestamp
12920 in timebase units instead of seconds.
12923 The maximum duration of the output in seconds.
12926 The number of the first frame that should be passed to the output.
12929 The number of the first frame that should be dropped.
12932 @option{start}, @option{end}, and @option{duration} are expressed as time
12933 duration specifications; see
12934 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
12935 for the accepted syntax.
12937 Note that the first two sets of the start/end options and the @option{duration}
12938 option look at the frame timestamp, while the _frame variants simply count the
12939 frames that pass through the filter. Also note that this filter does not modify
12940 the timestamps. If you wish for the output timestamps to start at zero, insert a
12941 setpts filter after the trim filter.
12943 If multiple start or end options are set, this filter tries to be greedy and
12944 keep all the frames that match at least one of the specified constraints. To keep
12945 only the part that matches all the constraints at once, chain multiple trim
12948 The defaults are such that all the input is kept. So it is possible to set e.g.
12949 just the end values to keep everything before the specified time.
12954 Drop everything except the second minute of input:
12956 ffmpeg -i INPUT -vf trim=60:120
12960 Keep only the first second:
12962 ffmpeg -i INPUT -vf trim=duration=1
12971 Sharpen or blur the input video.
12973 It accepts the following parameters:
12976 @item luma_msize_x, lx
12977 Set the luma matrix horizontal size. It must be an odd integer between
12978 3 and 63. The default value is 5.
12980 @item luma_msize_y, ly
12981 Set the luma matrix vertical size. It must be an odd integer between 3
12982 and 63. The default value is 5.
12984 @item luma_amount, la
12985 Set the luma effect strength. It must be a floating point number, reasonable
12986 values lay between -1.5 and 1.5.
12988 Negative values will blur the input video, while positive values will
12989 sharpen it, a value of zero will disable the effect.
12991 Default value is 1.0.
12993 @item chroma_msize_x, cx
12994 Set the chroma matrix horizontal size. It must be an odd integer
12995 between 3 and 63. The default value is 5.
12997 @item chroma_msize_y, cy
12998 Set the chroma matrix vertical size. It must be an odd integer
12999 between 3 and 63. The default value is 5.
13001 @item chroma_amount, ca
13002 Set the chroma effect strength. It must be a floating point number, reasonable
13003 values lay between -1.5 and 1.5.
13005 Negative values will blur the input video, while positive values will
13006 sharpen it, a value of zero will disable the effect.
13008 Default value is 0.0.
13011 If set to 1, specify using OpenCL capabilities, only available if
13012 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
13016 All parameters are optional and default to the equivalent of the
13017 string '5:5:1.0:5:5:0.0'.
13019 @subsection Examples
13023 Apply strong luma sharpen effect:
13025 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
13029 Apply a strong blur of both luma and chroma parameters:
13031 unsharp=7:7:-2:7:7:-2
13037 Apply ultra slow/simple postprocessing filter that compresses and decompresses
13038 the image at several (or - in the case of @option{quality} level @code{8} - all)
13039 shifts and average the results.
13041 The way this differs from the behavior of spp is that uspp actually encodes &
13042 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
13043 DCT similar to MJPEG.
13045 The filter accepts the following options:
13049 Set quality. This option defines the number of levels for averaging. It accepts
13050 an integer in the range 0-8. If set to @code{0}, the filter will have no
13051 effect. A value of @code{8} means the higher quality. For each increment of
13052 that value the speed drops by a factor of approximately 2. Default value is
13056 Force a constant quantization parameter. If not set, the filter will use the QP
13057 from the video stream (if available).
13060 @section vectorscope
13062 Display 2 color component values in the two dimensional graph (which is called
13065 This filter accepts the following options:
13069 Set vectorscope mode.
13071 It accepts the following values:
13074 Gray values are displayed on graph, higher brightness means more pixels have
13075 same component color value on location in graph. This is the default mode.
13078 Gray values are displayed on graph. Surrounding pixels values which are not
13079 present in video frame are drawn in gradient of 2 color components which are
13080 set by option @code{x} and @code{y}. The 3rd color component is static.
13083 Actual color components values present in video frame are displayed on graph.
13086 Similar as color2 but higher frequency of same values @code{x} and @code{y}
13087 on graph increases value of another color component, which is luminance by
13088 default values of @code{x} and @code{y}.
13091 Actual colors present in video frame are displayed on graph. If two different
13092 colors map to same position on graph then color with higher value of component
13093 not present in graph is picked.
13096 Gray values are displayed on graph. Similar to @code{color} but with 3rd color
13097 component picked from radial gradient.
13101 Set which color component will be represented on X-axis. Default is @code{1}.
13104 Set which color component will be represented on Y-axis. Default is @code{2}.
13107 Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
13108 of color component which represents frequency of (X, Y) location in graph.
13113 No envelope, this is default.
13116 Instant envelope, even darkest single pixel will be clearly highlighted.
13119 Hold maximum and minimum values presented in graph over time. This way you
13120 can still spot out of range values without constantly looking at vectorscope.
13123 Peak and instant envelope combined together.
13127 Set what kind of graticule to draw.
13135 Set graticule opacity.
13138 Set graticule flags.
13142 Draw graticule for white point.
13145 Draw graticule for black point.
13148 Draw color points short names.
13152 Set background opacity.
13154 @item lthreshold, l
13155 Set low threshold for color component not represented on X or Y axis.
13156 Values lower than this value will be ignored. Default is 0.
13157 Note this value is multiplied with actual max possible value one pixel component
13158 can have. So for 8-bit input and low threshold value of 0.1 actual threshold
13161 @item hthreshold, h
13162 Set high threshold for color component not represented on X or Y axis.
13163 Values higher than this value will be ignored. Default is 1.
13164 Note this value is multiplied with actual max possible value one pixel component
13165 can have. So for 8-bit input and high threshold value of 0.9 actual threshold
13166 is 0.9 * 255 = 230.
13168 @item colorspace, c
13169 Set what kind of colorspace to use when drawing graticule.
13178 @anchor{vidstabdetect}
13179 @section vidstabdetect
13181 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
13182 @ref{vidstabtransform} for pass 2.
13184 This filter generates a file with relative translation and rotation
13185 transform information about subsequent frames, which is then used by
13186 the @ref{vidstabtransform} filter.
13188 To enable compilation of this filter you need to configure FFmpeg with
13189 @code{--enable-libvidstab}.
13191 This filter accepts the following options:
13195 Set the path to the file used to write the transforms information.
13196 Default value is @file{transforms.trf}.
13199 Set how shaky the video is and how quick the camera is. It accepts an
13200 integer in the range 1-10, a value of 1 means little shakiness, a
13201 value of 10 means strong shakiness. Default value is 5.
13204 Set the accuracy of the detection process. It must be a value in the
13205 range 1-15. A value of 1 means low accuracy, a value of 15 means high
13206 accuracy. Default value is 15.
13209 Set stepsize of the search process. The region around minimum is
13210 scanned with 1 pixel resolution. Default value is 6.
13213 Set minimum contrast. Below this value a local measurement field is
13214 discarded. Must be a floating point value in the range 0-1. Default
13218 Set reference frame number for tripod mode.
13220 If enabled, the motion of the frames is compared to a reference frame
13221 in the filtered stream, identified by the specified number. The idea
13222 is to compensate all movements in a more-or-less static scene and keep
13223 the camera view absolutely still.
13225 If set to 0, it is disabled. The frames are counted starting from 1.
13228 Show fields and transforms in the resulting frames. It accepts an
13229 integer in the range 0-2. Default value is 0, which disables any
13233 @subsection Examples
13237 Use default values:
13243 Analyze strongly shaky movie and put the results in file
13244 @file{mytransforms.trf}:
13246 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
13250 Visualize the result of internal transformations in the resulting
13253 vidstabdetect=show=1
13257 Analyze a video with medium shakiness using @command{ffmpeg}:
13259 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
13263 @anchor{vidstabtransform}
13264 @section vidstabtransform
13266 Video stabilization/deshaking: pass 2 of 2,
13267 see @ref{vidstabdetect} for pass 1.
13269 Read a file with transform information for each frame and
13270 apply/compensate them. Together with the @ref{vidstabdetect}
13271 filter this can be used to deshake videos. See also
13272 @url{http://public.hronopik.de/vid.stab}. It is important to also use
13273 the @ref{unsharp} filter, see below.
13275 To enable compilation of this filter you need to configure FFmpeg with
13276 @code{--enable-libvidstab}.
13278 @subsection Options
13282 Set path to the file used to read the transforms. Default value is
13283 @file{transforms.trf}.
13286 Set the number of frames (value*2 + 1) used for lowpass filtering the
13287 camera movements. Default value is 10.
13289 For example a number of 10 means that 21 frames are used (10 in the
13290 past and 10 in the future) to smoothen the motion in the video. A
13291 larger value leads to a smoother video, but limits the acceleration of
13292 the camera (pan/tilt movements). 0 is a special case where a static
13293 camera is simulated.
13296 Set the camera path optimization algorithm.
13298 Accepted values are:
13301 gaussian kernel low-pass filter on camera motion (default)
13303 averaging on transformations
13307 Set maximal number of pixels to translate frames. Default value is -1,
13311 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
13312 value is -1, meaning no limit.
13315 Specify how to deal with borders that may be visible due to movement
13318 Available values are:
13321 keep image information from previous frame (default)
13323 fill the border black
13327 Invert transforms if set to 1. Default value is 0.
13330 Consider transforms as relative to previous frame if set to 1,
13331 absolute if set to 0. Default value is 0.
13334 Set percentage to zoom. A positive value will result in a zoom-in
13335 effect, a negative value in a zoom-out effect. Default value is 0 (no
13339 Set optimal zooming to avoid borders.
13341 Accepted values are:
13346 optimal static zoom value is determined (only very strong movements
13347 will lead to visible borders) (default)
13349 optimal adaptive zoom value is determined (no borders will be
13350 visible), see @option{zoomspeed}
13353 Note that the value given at zoom is added to the one calculated here.
13356 Set percent to zoom maximally each frame (enabled when
13357 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
13361 Specify type of interpolation.
13363 Available values are:
13368 linear only horizontal
13370 linear in both directions (default)
13372 cubic in both directions (slow)
13376 Enable virtual tripod mode if set to 1, which is equivalent to
13377 @code{relative=0:smoothing=0}. Default value is 0.
13379 Use also @code{tripod} option of @ref{vidstabdetect}.
13382 Increase log verbosity if set to 1. Also the detected global motions
13383 are written to the temporary file @file{global_motions.trf}. Default
13387 @subsection Examples
13391 Use @command{ffmpeg} for a typical stabilization with default values:
13393 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
13396 Note the use of the @ref{unsharp} filter which is always recommended.
13399 Zoom in a bit more and load transform data from a given file:
13401 vidstabtransform=zoom=5:input="mytransforms.trf"
13405 Smoothen the video even more:
13407 vidstabtransform=smoothing=30
13413 Flip the input video vertically.
13415 For example, to vertically flip a video with @command{ffmpeg}:
13417 ffmpeg -i in.avi -vf "vflip" out.avi
13423 Make or reverse a natural vignetting effect.
13425 The filter accepts the following options:
13429 Set lens angle expression as a number of radians.
13431 The value is clipped in the @code{[0,PI/2]} range.
13433 Default value: @code{"PI/5"}
13437 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
13441 Set forward/backward mode.
13443 Available modes are:
13446 The larger the distance from the central point, the darker the image becomes.
13449 The larger the distance from the central point, the brighter the image becomes.
13450 This can be used to reverse a vignette effect, though there is no automatic
13451 detection to extract the lens @option{angle} and other settings (yet). It can
13452 also be used to create a burning effect.
13455 Default value is @samp{forward}.
13458 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
13460 It accepts the following values:
13463 Evaluate expressions only once during the filter initialization.
13466 Evaluate expressions for each incoming frame. This is way slower than the
13467 @samp{init} mode since it requires all the scalers to be re-computed, but it
13468 allows advanced dynamic expressions.
13471 Default value is @samp{init}.
13474 Set dithering to reduce the circular banding effects. Default is @code{1}
13478 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
13479 Setting this value to the SAR of the input will make a rectangular vignetting
13480 following the dimensions of the video.
13482 Default is @code{1/1}.
13485 @subsection Expressions
13487 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
13488 following parameters.
13493 input width and height
13496 the number of input frame, starting from 0
13499 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
13500 @var{TB} units, NAN if undefined
13503 frame rate of the input video, NAN if the input frame rate is unknown
13506 the PTS (Presentation TimeStamp) of the filtered video frame,
13507 expressed in seconds, NAN if undefined
13510 time base of the input video
13514 @subsection Examples
13518 Apply simple strong vignetting effect:
13524 Make a flickering vignetting:
13526 vignette='PI/4+random(1)*PI/50':eval=frame
13532 Stack input videos vertically.
13534 All streams must be of same pixel format and of same width.
13536 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
13537 to create same output.
13539 The filter accept the following option:
13543 Set number of input streams. Default is 2.
13546 If set to 1, force the output to terminate when the shortest input
13547 terminates. Default value is 0.
13552 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
13553 Deinterlacing Filter").
13555 Based on the process described by Martin Weston for BBC R&D, and
13556 implemented based on the de-interlace algorithm written by Jim
13557 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
13558 uses filter coefficients calculated by BBC R&D.
13560 There are two sets of filter coefficients, so called "simple":
13561 and "complex". Which set of filter coefficients is used can
13562 be set by passing an optional parameter:
13566 Set the interlacing filter coefficients. Accepts one of the following values:
13570 Simple filter coefficient set.
13572 More-complex filter coefficient set.
13574 Default value is @samp{complex}.
13577 Specify which frames to deinterlace. Accept one of the following values:
13581 Deinterlace all frames,
13583 Only deinterlace frames marked as interlaced.
13586 Default value is @samp{all}.
13590 Video waveform monitor.
13592 The waveform monitor plots color component intensity. By default luminance
13593 only. Each column of the waveform corresponds to a column of pixels in the
13596 It accepts the following options:
13600 Can be either @code{row}, or @code{column}. Default is @code{column}.
13601 In row mode, the graph on the left side represents color component value 0 and
13602 the right side represents value = 255. In column mode, the top side represents
13603 color component value = 0 and bottom side represents value = 255.
13606 Set intensity. Smaller values are useful to find out how many values of the same
13607 luminance are distributed across input rows/columns.
13608 Default value is @code{0.04}. Allowed range is [0, 1].
13611 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
13612 In mirrored mode, higher values will be represented on the left
13613 side for @code{row} mode and at the top for @code{column} mode. Default is
13614 @code{1} (mirrored).
13618 It accepts the following values:
13621 Presents information identical to that in the @code{parade}, except
13622 that the graphs representing color components are superimposed directly
13625 This display mode makes it easier to spot relative differences or similarities
13626 in overlapping areas of the color components that are supposed to be identical,
13627 such as neutral whites, grays, or blacks.
13630 Display separate graph for the color components side by side in
13631 @code{row} mode or one below the other in @code{column} mode.
13634 Display separate graph for the color components side by side in
13635 @code{column} mode or one below the other in @code{row} mode.
13637 Using this display mode makes it easy to spot color casts in the highlights
13638 and shadows of an image, by comparing the contours of the top and the bottom
13639 graphs of each waveform. Since whites, grays, and blacks are characterized
13640 by exactly equal amounts of red, green, and blue, neutral areas of the picture
13641 should display three waveforms of roughly equal width/height. If not, the
13642 correction is easy to perform by making level adjustments the three waveforms.
13644 Default is @code{stack}.
13646 @item components, c
13647 Set which color components to display. Default is 1, which means only luminance
13648 or red color component if input is in RGB colorspace. If is set for example to
13649 7 it will display all 3 (if) available color components.
13654 No envelope, this is default.
13657 Instant envelope, minimum and maximum values presented in graph will be easily
13658 visible even with small @code{step} value.
13661 Hold minimum and maximum values presented in graph across time. This way you
13662 can still spot out of range values without constantly looking at waveforms.
13665 Peak and instant envelope combined together.
13671 No filtering, this is default.
13674 Luma and chroma combined together.
13677 Similar as above, but shows difference between blue and red chroma.
13680 Displays only chroma.
13683 Displays actual color value on waveform.
13686 Similar as above, but with luma showing frequency of chroma values.
13690 Set which graticule to display.
13694 Do not display graticule.
13697 Display green graticule showing legal broadcast ranges.
13701 Set graticule opacity.
13704 Set graticule flags.
13708 Draw numbers above lines. By default enabled.
13711 Draw dots instead of lines.
13715 Set scale used for displaying graticule.
13722 Default is digital.
13726 Apply the xBR high-quality magnification filter which is designed for pixel
13727 art. It follows a set of edge-detection rules, see
13728 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
13730 It accepts the following option:
13734 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
13735 @code{3xBR} and @code{4} for @code{4xBR}.
13736 Default is @code{3}.
13742 Deinterlace the input video ("yadif" means "yet another deinterlacing
13745 It accepts the following parameters:
13751 The interlacing mode to adopt. It accepts one of the following values:
13754 @item 0, send_frame
13755 Output one frame for each frame.
13756 @item 1, send_field
13757 Output one frame for each field.
13758 @item 2, send_frame_nospatial
13759 Like @code{send_frame}, but it skips the spatial interlacing check.
13760 @item 3, send_field_nospatial
13761 Like @code{send_field}, but it skips the spatial interlacing check.
13764 The default value is @code{send_frame}.
13767 The picture field parity assumed for the input interlaced video. It accepts one
13768 of the following values:
13772 Assume the top field is first.
13774 Assume the bottom field is first.
13776 Enable automatic detection of field parity.
13779 The default value is @code{auto}.
13780 If the interlacing is unknown or the decoder does not export this information,
13781 top field first will be assumed.
13784 Specify which frames to deinterlace. Accept one of the following
13789 Deinterlace all frames.
13790 @item 1, interlaced
13791 Only deinterlace frames marked as interlaced.
13794 The default value is @code{all}.
13799 Apply Zoom & Pan effect.
13801 This filter accepts the following options:
13805 Set the zoom expression. Default is 1.
13809 Set the x and y expression. Default is 0.
13812 Set the duration expression in number of frames.
13813 This sets for how many number of frames effect will last for
13814 single input image.
13817 Set the output image size, default is 'hd720'.
13820 Set the output frame rate, default is '25'.
13823 Each expression can contain the following constants:
13842 Output frame count.
13846 Last calculated 'x' and 'y' position from 'x' and 'y' expression
13847 for current input frame.
13851 'x' and 'y' of last output frame of previous input frame or 0 when there was
13852 not yet such frame (first input frame).
13855 Last calculated zoom from 'z' expression for current input frame.
13858 Last calculated zoom of last output frame of previous input frame.
13861 Number of output frames for current input frame. Calculated from 'd' expression
13862 for each input frame.
13865 number of output frames created for previous input frame
13868 Rational number: input width / input height
13871 sample aspect ratio
13874 display aspect ratio
13878 @subsection Examples
13882 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
13884 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
13888 Zoom-in up to 1.5 and pan always at center of picture:
13890 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
13895 Scale (resize) the input video, using the z.lib library:
13896 https://github.com/sekrit-twc/zimg.
13898 The zscale filter forces the output display aspect ratio to be the same
13899 as the input, by changing the output sample aspect ratio.
13901 If the input image format is different from the format requested by
13902 the next filter, the zscale filter will convert the input to the
13905 @subsection Options
13906 The filter accepts the following options.
13911 Set the output video dimension expression. Default value is the input
13914 If the @var{width} or @var{w} is 0, the input width is used for the output.
13915 If the @var{height} or @var{h} is 0, the input height is used for the output.
13917 If one of the values is -1, the zscale filter will use a value that
13918 maintains the aspect ratio of the input image, calculated from the
13919 other specified dimension. If both of them are -1, the input size is
13922 If one of the values is -n with n > 1, the zscale filter will also use a value
13923 that maintains the aspect ratio of the input image, calculated from the other
13924 specified dimension. After that it will, however, make sure that the calculated
13925 dimension is divisible by n and adjust the value if necessary.
13927 See below for the list of accepted constants for use in the dimension
13931 Set the video size. For the syntax of this option, check the
13932 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13935 Set the dither type.
13937 Possible values are:
13942 @item error_diffusion
13948 Set the resize filter type.
13950 Possible values are:
13960 Default is bilinear.
13963 Set the color range.
13965 Possible values are:
13972 Default is same as input.
13975 Set the color primaries.
13977 Possible values are:
13987 Default is same as input.
13990 Set the transfer characteristics.
13992 Possible values are:
14003 Default is same as input.
14006 Set the colorspace matrix.
14008 Possible value are:
14019 Default is same as input.
14022 Set the input color range.
14024 Possible values are:
14031 Default is same as input.
14033 @item primariesin, pin
14034 Set the input color primaries.
14036 Possible values are:
14046 Default is same as input.
14048 @item transferin, tin
14049 Set the input transfer characteristics.
14051 Possible values are:
14062 Default is same as input.
14064 @item matrixin, min
14065 Set the input colorspace matrix.
14067 Possible value are:
14079 The values of the @option{w} and @option{h} options are expressions
14080 containing the following constants:
14085 The input width and height
14089 These are the same as @var{in_w} and @var{in_h}.
14093 The output (scaled) width and height
14097 These are the same as @var{out_w} and @var{out_h}
14100 The same as @var{iw} / @var{ih}
14103 input sample aspect ratio
14106 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
14110 horizontal and vertical input chroma subsample values. For example for the
14111 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
14115 horizontal and vertical output chroma subsample values. For example for the
14116 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
14122 @c man end VIDEO FILTERS
14124 @chapter Video Sources
14125 @c man begin VIDEO SOURCES
14127 Below is a description of the currently available video sources.
14131 Buffer video frames, and make them available to the filter chain.
14133 This source is mainly intended for a programmatic use, in particular
14134 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
14136 It accepts the following parameters:
14141 Specify the size (width and height) of the buffered video frames. For the
14142 syntax of this option, check the
14143 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14146 The input video width.
14149 The input video height.
14152 A string representing the pixel format of the buffered video frames.
14153 It may be a number corresponding to a pixel format, or a pixel format
14157 Specify the timebase assumed by the timestamps of the buffered frames.
14160 Specify the frame rate expected for the video stream.
14162 @item pixel_aspect, sar
14163 The sample (pixel) aspect ratio of the input video.
14166 Specify the optional parameters to be used for the scale filter which
14167 is automatically inserted when an input change is detected in the
14168 input size or format.
14170 @item hw_frames_ctx
14171 When using a hardware pixel format, this should be a reference to an
14172 AVHWFramesContext describing input frames.
14177 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
14180 will instruct the source to accept video frames with size 320x240 and
14181 with format "yuv410p", assuming 1/24 as the timestamps timebase and
14182 square pixels (1:1 sample aspect ratio).
14183 Since the pixel format with name "yuv410p" corresponds to the number 6
14184 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
14185 this example corresponds to:
14187 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
14190 Alternatively, the options can be specified as a flat string, but this
14191 syntax is deprecated:
14193 @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}]
14197 Create a pattern generated by an elementary cellular automaton.
14199 The initial state of the cellular automaton can be defined through the
14200 @option{filename}, and @option{pattern} options. If such options are
14201 not specified an initial state is created randomly.
14203 At each new frame a new row in the video is filled with the result of
14204 the cellular automaton next generation. The behavior when the whole
14205 frame is filled is defined by the @option{scroll} option.
14207 This source accepts the following options:
14211 Read the initial cellular automaton state, i.e. the starting row, from
14212 the specified file.
14213 In the file, each non-whitespace character is considered an alive
14214 cell, a newline will terminate the row, and further characters in the
14215 file will be ignored.
14218 Read the initial cellular automaton state, i.e. the starting row, from
14219 the specified string.
14221 Each non-whitespace character in the string is considered an alive
14222 cell, a newline will terminate the row, and further characters in the
14223 string will be ignored.
14226 Set the video rate, that is the number of frames generated per second.
14229 @item random_fill_ratio, ratio
14230 Set the random fill ratio for the initial cellular automaton row. It
14231 is a floating point number value ranging from 0 to 1, defaults to
14234 This option is ignored when a file or a pattern is specified.
14236 @item random_seed, seed
14237 Set the seed for filling randomly the initial row, must be an integer
14238 included between 0 and UINT32_MAX. If not specified, or if explicitly
14239 set to -1, the filter will try to use a good random seed on a best
14243 Set the cellular automaton rule, it is a number ranging from 0 to 255.
14244 Default value is 110.
14247 Set the size of the output video. For the syntax of this option, check the
14248 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14250 If @option{filename} or @option{pattern} is specified, the size is set
14251 by default to the width of the specified initial state row, and the
14252 height is set to @var{width} * PHI.
14254 If @option{size} is set, it must contain the width of the specified
14255 pattern string, and the specified pattern will be centered in the
14258 If a filename or a pattern string is not specified, the size value
14259 defaults to "320x518" (used for a randomly generated initial state).
14262 If set to 1, scroll the output upward when all the rows in the output
14263 have been already filled. If set to 0, the new generated row will be
14264 written over the top row just after the bottom row is filled.
14267 @item start_full, full
14268 If set to 1, completely fill the output with generated rows before
14269 outputting the first frame.
14270 This is the default behavior, for disabling set the value to 0.
14273 If set to 1, stitch the left and right row edges together.
14274 This is the default behavior, for disabling set the value to 0.
14277 @subsection Examples
14281 Read the initial state from @file{pattern}, and specify an output of
14284 cellauto=f=pattern:s=200x400
14288 Generate a random initial row with a width of 200 cells, with a fill
14291 cellauto=ratio=2/3:s=200x200
14295 Create a pattern generated by rule 18 starting by a single alive cell
14296 centered on an initial row with width 100:
14298 cellauto=p=@@:s=100x400:full=0:rule=18
14302 Specify a more elaborated initial pattern:
14304 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
14309 @anchor{coreimagesrc}
14310 @section coreimagesrc
14311 Video source generated on GPU using Apple's CoreImage API on OSX.
14313 This video source is a specialized version of the @ref{coreimage} video filter.
14314 Use a core image generator at the beginning of the applied filterchain to
14315 generate the content.
14317 The coreimagesrc video source accepts the following options:
14319 @item list_generators
14320 List all available generators along with all their respective options as well as
14321 possible minimum and maximum values along with the default values.
14323 list_generators=true
14327 Specify the size of the sourced video. For the syntax of this option, check the
14328 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14329 The default value is @code{320x240}.
14332 Specify the frame rate of the sourced video, as the number of frames
14333 generated per second. It has to be a string in the format
14334 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
14335 number or a valid video frame rate abbreviation. The default value is
14339 Set the sample aspect ratio of the sourced video.
14342 Set the duration of the sourced video. See
14343 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
14344 for the accepted syntax.
14346 If not specified, or the expressed duration is negative, the video is
14347 supposed to be generated forever.
14350 Additionally, all options of the @ref{coreimage} video filter are accepted.
14351 A complete filterchain can be used for further processing of the
14352 generated input without CPU-HOST transfer. See @ref{coreimage} documentation
14353 and examples for details.
14355 @subsection Examples
14360 Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
14361 given as complete and escaped command-line for Apple's standard bash shell:
14363 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
14365 This example is equivalent to the QRCode example of @ref{coreimage} without the
14366 need for a nullsrc video source.
14370 @section mandelbrot
14372 Generate a Mandelbrot set fractal, and progressively zoom towards the
14373 point specified with @var{start_x} and @var{start_y}.
14375 This source accepts the following options:
14380 Set the terminal pts value. Default value is 400.
14383 Set the terminal scale value.
14384 Must be a floating point value. Default value is 0.3.
14387 Set the inner coloring mode, that is the algorithm used to draw the
14388 Mandelbrot fractal internal region.
14390 It shall assume one of the following values:
14395 Show time until convergence.
14397 Set color based on point closest to the origin of the iterations.
14402 Default value is @var{mincol}.
14405 Set the bailout value. Default value is 10.0.
14408 Set the maximum of iterations performed by the rendering
14409 algorithm. Default value is 7189.
14412 Set outer coloring mode.
14413 It shall assume one of following values:
14415 @item iteration_count
14416 Set iteration cound mode.
14417 @item normalized_iteration_count
14418 set normalized iteration count mode.
14420 Default value is @var{normalized_iteration_count}.
14423 Set frame rate, expressed as number of frames per second. Default
14427 Set frame size. For the syntax of this option, check the "Video
14428 size" section in the ffmpeg-utils manual. Default value is "640x480".
14431 Set the initial scale value. Default value is 3.0.
14434 Set the initial x position. Must be a floating point value between
14435 -100 and 100. Default value is -0.743643887037158704752191506114774.
14438 Set the initial y position. Must be a floating point value between
14439 -100 and 100. Default value is -0.131825904205311970493132056385139.
14444 Generate various test patterns, as generated by the MPlayer test filter.
14446 The size of the generated video is fixed, and is 256x256.
14447 This source is useful in particular for testing encoding features.
14449 This source accepts the following options:
14454 Specify the frame rate of the sourced video, as the number of frames
14455 generated per second. It has to be a string in the format
14456 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
14457 number or a valid video frame rate abbreviation. The default value is
14461 Set the duration of the sourced video. See
14462 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
14463 for the accepted syntax.
14465 If not specified, or the expressed duration is negative, the video is
14466 supposed to be generated forever.
14470 Set the number or the name of the test to perform. Supported tests are:
14486 Default value is "all", which will cycle through the list of all tests.
14491 mptestsrc=t=dc_luma
14494 will generate a "dc_luma" test pattern.
14496 @section frei0r_src
14498 Provide a frei0r source.
14500 To enable compilation of this filter you need to install the frei0r
14501 header and configure FFmpeg with @code{--enable-frei0r}.
14503 This source accepts the following parameters:
14508 The size of the video to generate. For the syntax of this option, check the
14509 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14512 The framerate of the generated video. It may be a string of the form
14513 @var{num}/@var{den} or a frame rate abbreviation.
14516 The name to the frei0r source to load. For more information regarding frei0r and
14517 how to set the parameters, read the @ref{frei0r} section in the video filters
14520 @item filter_params
14521 A '|'-separated list of parameters to pass to the frei0r source.
14525 For example, to generate a frei0r partik0l source with size 200x200
14526 and frame rate 10 which is overlaid on the overlay filter main input:
14528 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
14533 Generate a life pattern.
14535 This source is based on a generalization of John Conway's life game.
14537 The sourced input represents a life grid, each pixel represents a cell
14538 which can be in one of two possible states, alive or dead. Every cell
14539 interacts with its eight neighbours, which are the cells that are
14540 horizontally, vertically, or diagonally adjacent.
14542 At each interaction the grid evolves according to the adopted rule,
14543 which specifies the number of neighbor alive cells which will make a
14544 cell stay alive or born. The @option{rule} option allows one to specify
14547 This source accepts the following options:
14551 Set the file from which to read the initial grid state. In the file,
14552 each non-whitespace character is considered an alive cell, and newline
14553 is used to delimit the end of each row.
14555 If this option is not specified, the initial grid is generated
14559 Set the video rate, that is the number of frames generated per second.
14562 @item random_fill_ratio, ratio
14563 Set the random fill ratio for the initial random grid. It is a
14564 floating point number value ranging from 0 to 1, defaults to 1/PHI.
14565 It is ignored when a file is specified.
14567 @item random_seed, seed
14568 Set the seed for filling the initial random grid, must be an integer
14569 included between 0 and UINT32_MAX. If not specified, or if explicitly
14570 set to -1, the filter will try to use a good random seed on a best
14576 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
14577 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
14578 @var{NS} specifies the number of alive neighbor cells which make a
14579 live cell stay alive, and @var{NB} the number of alive neighbor cells
14580 which make a dead cell to become alive (i.e. to "born").
14581 "s" and "b" can be used in place of "S" and "B", respectively.
14583 Alternatively a rule can be specified by an 18-bits integer. The 9
14584 high order bits are used to encode the next cell state if it is alive
14585 for each number of neighbor alive cells, the low order bits specify
14586 the rule for "borning" new cells. Higher order bits encode for an
14587 higher number of neighbor cells.
14588 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
14589 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
14591 Default value is "S23/B3", which is the original Conway's game of life
14592 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
14593 cells, and will born a new cell if there are three alive cells around
14597 Set the size of the output video. For the syntax of this option, check the
14598 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14600 If @option{filename} is specified, the size is set by default to the
14601 same size of the input file. If @option{size} is set, it must contain
14602 the size specified in the input file, and the initial grid defined in
14603 that file is centered in the larger resulting area.
14605 If a filename is not specified, the size value defaults to "320x240"
14606 (used for a randomly generated initial grid).
14609 If set to 1, stitch the left and right grid edges together, and the
14610 top and bottom edges also. Defaults to 1.
14613 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
14614 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
14615 value from 0 to 255.
14618 Set the color of living (or new born) cells.
14621 Set the color of dead cells. If @option{mold} is set, this is the first color
14622 used to represent a dead cell.
14625 Set mold color, for definitely dead and moldy cells.
14627 For the syntax of these 3 color options, check the "Color" section in the
14628 ffmpeg-utils manual.
14631 @subsection Examples
14635 Read a grid from @file{pattern}, and center it on a grid of size
14638 life=f=pattern:s=300x300
14642 Generate a random grid of size 200x200, with a fill ratio of 2/3:
14644 life=ratio=2/3:s=200x200
14648 Specify a custom rule for evolving a randomly generated grid:
14654 Full example with slow death effect (mold) using @command{ffplay}:
14656 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
14663 @anchor{haldclutsrc}
14665 @anchor{rgbtestsrc}
14667 @anchor{smptehdbars}
14670 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2
14672 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
14674 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
14676 The @code{color} source provides an uniformly colored input.
14678 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
14679 @ref{haldclut} filter.
14681 The @code{nullsrc} source returns unprocessed video frames. It is
14682 mainly useful to be employed in analysis / debugging tools, or as the
14683 source for filters which ignore the input data.
14685 The @code{rgbtestsrc} source generates an RGB test pattern useful for
14686 detecting RGB vs BGR issues. You should see a red, green and blue
14687 stripe from top to bottom.
14689 The @code{smptebars} source generates a color bars pattern, based on
14690 the SMPTE Engineering Guideline EG 1-1990.
14692 The @code{smptehdbars} source generates a color bars pattern, based on
14693 the SMPTE RP 219-2002.
14695 The @code{testsrc} source generates a test video pattern, showing a
14696 color pattern, a scrolling gradient and a timestamp. This is mainly
14697 intended for testing purposes.
14699 The @code{testsrc2} source is similar to testsrc, but supports more
14700 pixel formats instead of just @code{rgb24}. This allows using it as an
14701 input for other tests without requiring a format conversion.
14703 The sources accept the following parameters:
14708 Specify the color of the source, only available in the @code{color}
14709 source. For the syntax of this option, check the "Color" section in the
14710 ffmpeg-utils manual.
14713 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
14714 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
14715 pixels to be used as identity matrix for 3D lookup tables. Each component is
14716 coded on a @code{1/(N*N)} scale.
14719 Specify the size of the sourced video. For the syntax of this option, check the
14720 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14721 The default value is @code{320x240}.
14723 This option is not available with the @code{haldclutsrc} filter.
14726 Specify the frame rate of the sourced video, as the number of frames
14727 generated per second. It has to be a string in the format
14728 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
14729 number or a valid video frame rate abbreviation. The default value is
14733 Set the sample aspect ratio of the sourced video.
14736 Set the duration of the sourced video. See
14737 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
14738 for the accepted syntax.
14740 If not specified, or the expressed duration is negative, the video is
14741 supposed to be generated forever.
14744 Set the number of decimals to show in the timestamp, only available in the
14745 @code{testsrc} source.
14747 The displayed timestamp value will correspond to the original
14748 timestamp value multiplied by the power of 10 of the specified
14749 value. Default value is 0.
14752 For example the following:
14754 testsrc=duration=5.3:size=qcif:rate=10
14757 will generate a video with a duration of 5.3 seconds, with size
14758 176x144 and a frame rate of 10 frames per second.
14760 The following graph description will generate a red source
14761 with an opacity of 0.2, with size "qcif" and a frame rate of 10
14764 color=c=red@@0.2:s=qcif:r=10
14767 If the input content is to be ignored, @code{nullsrc} can be used. The
14768 following command generates noise in the luminance plane by employing
14769 the @code{geq} filter:
14771 nullsrc=s=256x256, geq=random(1)*255:128:128
14774 @subsection Commands
14776 The @code{color} source supports the following commands:
14780 Set the color of the created image. Accepts the same syntax of the
14781 corresponding @option{color} option.
14784 @c man end VIDEO SOURCES
14786 @chapter Video Sinks
14787 @c man begin VIDEO SINKS
14789 Below is a description of the currently available video sinks.
14791 @section buffersink
14793 Buffer video frames, and make them available to the end of the filter
14796 This sink is mainly intended for programmatic use, in particular
14797 through the interface defined in @file{libavfilter/buffersink.h}
14798 or the options system.
14800 It accepts a pointer to an AVBufferSinkContext structure, which
14801 defines the incoming buffers' formats, to be passed as the opaque
14802 parameter to @code{avfilter_init_filter} for initialization.
14806 Null video sink: do absolutely nothing with the input video. It is
14807 mainly useful as a template and for use in analysis / debugging
14810 @c man end VIDEO SINKS
14812 @chapter Multimedia Filters
14813 @c man begin MULTIMEDIA FILTERS
14815 Below is a description of the currently available multimedia filters.
14817 @section ahistogram
14819 Convert input audio to a video output, displaying the volume histogram.
14821 The filter accepts the following options:
14825 Specify how histogram is calculated.
14827 It accepts the following values:
14830 Use single histogram for all channels.
14832 Use separate histogram for each channel.
14834 Default is @code{single}.
14837 Set frame rate, expressed as number of frames per second. Default
14841 Specify the video size for the output. For the syntax of this option, check the
14842 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14843 Default value is @code{hd720}.
14848 It accepts the following values:
14859 reverse logarithmic
14861 Default is @code{log}.
14864 Set amplitude scale.
14866 It accepts the following values:
14873 Default is @code{log}.
14876 Set how much frames to accumulate in histogram.
14877 Defauls is 1. Setting this to -1 accumulates all frames.
14880 Set histogram ratio of window height.
14883 Set sonogram sliding.
14885 It accepts the following values:
14888 replace old rows with new ones.
14890 scroll from top to bottom.
14892 Default is @code{replace}.
14895 @section aphasemeter
14897 Convert input audio to a video output, displaying the audio phase.
14899 The filter accepts the following options:
14903 Set the output frame rate. Default value is @code{25}.
14906 Set the video size for the output. For the syntax of this option, check the
14907 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14908 Default value is @code{800x400}.
14913 Specify the red, green, blue contrast. Default values are @code{2},
14914 @code{7} and @code{1}.
14915 Allowed range is @code{[0, 255]}.
14918 Set color which will be used for drawing median phase. If color is
14919 @code{none} which is default, no median phase value will be drawn.
14922 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
14923 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
14924 The @code{-1} means left and right channels are completely out of phase and
14925 @code{1} means channels are in phase.
14927 @section avectorscope
14929 Convert input audio to a video output, representing the audio vector
14932 The filter is used to measure the difference between channels of stereo
14933 audio stream. A monoaural signal, consisting of identical left and right
14934 signal, results in straight vertical line. Any stereo separation is visible
14935 as a deviation from this line, creating a Lissajous figure.
14936 If the straight (or deviation from it) but horizontal line appears this
14937 indicates that the left and right channels are out of phase.
14939 The filter accepts the following options:
14943 Set the vectorscope mode.
14945 Available values are:
14948 Lissajous rotated by 45 degrees.
14951 Same as above but not rotated.
14954 Shape resembling half of circle.
14957 Default value is @samp{lissajous}.
14960 Set the video size for the output. For the syntax of this option, check the
14961 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14962 Default value is @code{400x400}.
14965 Set the output frame rate. Default value is @code{25}.
14971 Specify the red, green, blue and alpha contrast. Default values are @code{40},
14972 @code{160}, @code{80} and @code{255}.
14973 Allowed range is @code{[0, 255]}.
14979 Specify the red, green, blue and alpha fade. Default values are @code{15},
14980 @code{10}, @code{5} and @code{5}.
14981 Allowed range is @code{[0, 255]}.
14984 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
14987 Set the vectorscope drawing mode.
14989 Available values are:
14992 Draw dot for each sample.
14995 Draw line between previous and current sample.
14998 Default value is @samp{dot}.
15001 @subsection Examples
15005 Complete example using @command{ffplay}:
15007 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
15008 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
15012 @section bench, abench
15014 Benchmark part of a filtergraph.
15016 The filter accepts the following options:
15020 Start or stop a timer.
15022 Available values are:
15025 Get the current time, set it as frame metadata (using the key
15026 @code{lavfi.bench.start_time}), and forward the frame to the next filter.
15029 Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
15030 the input frame metadata to get the time difference. Time difference, average,
15031 maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
15032 @code{min}) are then printed. The timestamps are expressed in seconds.
15036 @subsection Examples
15040 Benchmark @ref{selectivecolor} filter:
15042 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
15048 Concatenate audio and video streams, joining them together one after the
15051 The filter works on segments of synchronized video and audio streams. All
15052 segments must have the same number of streams of each type, and that will
15053 also be the number of streams at output.
15055 The filter accepts the following options:
15060 Set the number of segments. Default is 2.
15063 Set the number of output video streams, that is also the number of video
15064 streams in each segment. Default is 1.
15067 Set the number of output audio streams, that is also the number of audio
15068 streams in each segment. Default is 0.
15071 Activate unsafe mode: do not fail if segments have a different format.
15075 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
15076 @var{a} audio outputs.
15078 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
15079 segment, in the same order as the outputs, then the inputs for the second
15082 Related streams do not always have exactly the same duration, for various
15083 reasons including codec frame size or sloppy authoring. For that reason,
15084 related synchronized streams (e.g. a video and its audio track) should be
15085 concatenated at once. The concat filter will use the duration of the longest
15086 stream in each segment (except the last one), and if necessary pad shorter
15087 audio streams with silence.
15089 For this filter to work correctly, all segments must start at timestamp 0.
15091 All corresponding streams must have the same parameters in all segments; the
15092 filtering system will automatically select a common pixel format for video
15093 streams, and a common sample format, sample rate and channel layout for
15094 audio streams, but other settings, such as resolution, must be converted
15095 explicitly by the user.
15097 Different frame rates are acceptable but will result in variable frame rate
15098 at output; be sure to configure the output file to handle it.
15100 @subsection Examples
15104 Concatenate an opening, an episode and an ending, all in bilingual version
15105 (video in stream 0, audio in streams 1 and 2):
15107 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
15108 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
15109 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
15110 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
15114 Concatenate two parts, handling audio and video separately, using the
15115 (a)movie sources, and adjusting the resolution:
15117 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
15118 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
15119 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
15121 Note that a desync will happen at the stitch if the audio and video streams
15122 do not have exactly the same duration in the first file.
15126 @section drawgraph, adrawgraph
15128 Draw a graph using input video or audio metadata.
15130 It accepts the following parameters:
15134 Set 1st frame metadata key from which metadata values will be used to draw a graph.
15137 Set 1st foreground color expression.
15140 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
15143 Set 2nd foreground color expression.
15146 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
15149 Set 3rd foreground color expression.
15152 Set 4th frame metadata key from which metadata values will be used to draw a graph.
15155 Set 4th foreground color expression.
15158 Set minimal value of metadata value.
15161 Set maximal value of metadata value.
15164 Set graph background color. Default is white.
15169 Available values for mode is:
15176 Default is @code{line}.
15181 Available values for slide is:
15184 Draw new frame when right border is reached.
15187 Replace old columns with new ones.
15190 Scroll from right to left.
15193 Scroll from left to right.
15196 Draw single picture.
15199 Default is @code{frame}.
15202 Set size of graph video. For the syntax of this option, check the
15203 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15204 The default value is @code{900x256}.
15206 The foreground color expressions can use the following variables:
15209 Minimal value of metadata value.
15212 Maximal value of metadata value.
15215 Current metadata key value.
15218 The color is defined as 0xAABBGGRR.
15221 Example using metadata from @ref{signalstats} filter:
15223 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
15226 Example using metadata from @ref{ebur128} filter:
15228 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
15234 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
15235 it unchanged. By default, it logs a message at a frequency of 10Hz with the
15236 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
15237 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
15239 The filter also has a video output (see the @var{video} option) with a real
15240 time graph to observe the loudness evolution. The graphic contains the logged
15241 message mentioned above, so it is not printed anymore when this option is set,
15242 unless the verbose logging is set. The main graphing area contains the
15243 short-term loudness (3 seconds of analysis), and the gauge on the right is for
15244 the momentary loudness (400 milliseconds).
15246 More information about the Loudness Recommendation EBU R128 on
15247 @url{http://tech.ebu.ch/loudness}.
15249 The filter accepts the following options:
15254 Activate the video output. The audio stream is passed unchanged whether this
15255 option is set or no. The video stream will be the first output stream if
15256 activated. Default is @code{0}.
15259 Set the video size. This option is for video only. For the syntax of this
15261 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15262 Default and minimum resolution is @code{640x480}.
15265 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
15266 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
15267 other integer value between this range is allowed.
15270 Set metadata injection. If set to @code{1}, the audio input will be segmented
15271 into 100ms output frames, each of them containing various loudness information
15272 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
15274 Default is @code{0}.
15277 Force the frame logging level.
15279 Available values are:
15282 information logging level
15284 verbose logging level
15287 By default, the logging level is set to @var{info}. If the @option{video} or
15288 the @option{metadata} options are set, it switches to @var{verbose}.
15293 Available modes can be cumulated (the option is a @code{flag} type). Possible
15297 Disable any peak mode (default).
15299 Enable sample-peak mode.
15301 Simple peak mode looking for the higher sample value. It logs a message
15302 for sample-peak (identified by @code{SPK}).
15304 Enable true-peak mode.
15306 If enabled, the peak lookup is done on an over-sampled version of the input
15307 stream for better peak accuracy. It logs a message for true-peak.
15308 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
15309 This mode requires a build with @code{libswresample}.
15313 Treat mono input files as "dual mono". If a mono file is intended for playback
15314 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
15315 If set to @code{true}, this option will compensate for this effect.
15316 Multi-channel input files are not affected by this option.
15319 Set a specific pan law to be used for the measurement of dual mono files.
15320 This parameter is optional, and has a default value of -3.01dB.
15323 @subsection Examples
15327 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
15329 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
15333 Run an analysis with @command{ffmpeg}:
15335 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
15339 @section interleave, ainterleave
15341 Temporally interleave frames from several inputs.
15343 @code{interleave} works with video inputs, @code{ainterleave} with audio.
15345 These filters read frames from several inputs and send the oldest
15346 queued frame to the output.
15348 Input streams must have a well defined, monotonically increasing frame
15351 In order to submit one frame to output, these filters need to enqueue
15352 at least one frame for each input, so they cannot work in case one
15353 input is not yet terminated and will not receive incoming frames.
15355 For example consider the case when one input is a @code{select} filter
15356 which always drop input frames. The @code{interleave} filter will keep
15357 reading from that input, but it will never be able to send new frames
15358 to output until the input will send an end-of-stream signal.
15360 Also, depending on inputs synchronization, the filters will drop
15361 frames in case one input receives more frames than the other ones, and
15362 the queue is already filled.
15364 These filters accept the following options:
15368 Set the number of different inputs, it is 2 by default.
15371 @subsection Examples
15375 Interleave frames belonging to different streams using @command{ffmpeg}:
15377 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
15381 Add flickering blur effect:
15383 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
15387 @section metadata, ametadata
15389 Manipulate frame metadata.
15391 This filter accepts the following options:
15395 Set mode of operation of the filter.
15397 Can be one of the following:
15401 If both @code{value} and @code{key} is set, select frames
15402 which have such metadata. If only @code{key} is set, select
15403 every frame that has such key in metadata.
15406 Add new metadata @code{key} and @code{value}. If key is already available
15410 Modify value of already present key.
15413 If @code{value} is set, delete only keys that have such value.
15414 Otherwise, delete key.
15417 Print key and its value if metadata was found. If @code{key} is not set print all
15418 metadata values available in frame.
15422 Set key used with all modes. Must be set for all modes except @code{print}.
15425 Set metadata value which will be used. This option is mandatory for
15426 @code{modify} and @code{add} mode.
15429 Which function to use when comparing metadata value and @code{value}.
15431 Can be one of following:
15435 Values are interpreted as strings, returns true if metadata value is same as @code{value}.
15438 Values are interpreted as strings, returns true if metadata value starts with
15439 the @code{value} option string.
15442 Values are interpreted as floats, returns true if metadata value is less than @code{value}.
15445 Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
15448 Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
15451 Values are interpreted as floats, returns true if expression from option @code{expr}
15456 Set expression which is used when @code{function} is set to @code{expr}.
15457 The expression is evaluated through the eval API and can contain the following
15462 Float representation of @code{value} from metadata key.
15465 Float representation of @code{value} as supplied by user in @code{value} option.
15468 If specified in @code{print} mode, output is written to the named file. Instead of
15469 plain filename any writable url can be specified. Filename ``-'' is a shorthand
15470 for standard output. If @code{file} option is not set, output is written to the log
15471 with AV_LOG_INFO loglevel.
15476 @subsection Examples
15480 Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
15483 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
15486 Print silencedetect output to file @file{metadata.txt}.
15488 silencedetect,ametadata=mode=print:file=metadata.txt
15491 Direct all metadata to a pipe with file descriptor 4.
15493 metadata=mode=print:file='pipe\:4'
15497 @section perms, aperms
15499 Set read/write permissions for the output frames.
15501 These filters are mainly aimed at developers to test direct path in the
15502 following filter in the filtergraph.
15504 The filters accept the following options:
15508 Select the permissions mode.
15510 It accepts the following values:
15513 Do nothing. This is the default.
15515 Set all the output frames read-only.
15517 Set all the output frames directly writable.
15519 Make the frame read-only if writable, and writable if read-only.
15521 Set each output frame read-only or writable randomly.
15525 Set the seed for the @var{random} mode, must be an integer included between
15526 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
15527 @code{-1}, the filter will try to use a good random seed on a best effort
15531 Note: in case of auto-inserted filter between the permission filter and the
15532 following one, the permission might not be received as expected in that
15533 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
15534 perms/aperms filter can avoid this problem.
15536 @section realtime, arealtime
15538 Slow down filtering to match real time approximatively.
15540 These filters will pause the filtering for a variable amount of time to
15541 match the output rate with the input timestamps.
15542 They are similar to the @option{re} option to @code{ffmpeg}.
15544 They accept the following options:
15548 Time limit for the pauses. Any pause longer than that will be considered
15549 a timestamp discontinuity and reset the timer. Default is 2 seconds.
15552 @section select, aselect
15554 Select frames to pass in output.
15556 This filter accepts the following options:
15561 Set expression, which is evaluated for each input frame.
15563 If the expression is evaluated to zero, the frame is discarded.
15565 If the evaluation result is negative or NaN, the frame is sent to the
15566 first output; otherwise it is sent to the output with index
15567 @code{ceil(val)-1}, assuming that the input index starts from 0.
15569 For example a value of @code{1.2} corresponds to the output with index
15570 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
15573 Set the number of outputs. The output to which to send the selected
15574 frame is based on the result of the evaluation. Default value is 1.
15577 The expression can contain the following constants:
15581 The (sequential) number of the filtered frame, starting from 0.
15584 The (sequential) number of the selected frame, starting from 0.
15586 @item prev_selected_n
15587 The sequential number of the last selected frame. It's NAN if undefined.
15590 The timebase of the input timestamps.
15593 The PTS (Presentation TimeStamp) of the filtered video frame,
15594 expressed in @var{TB} units. It's NAN if undefined.
15597 The PTS of the filtered video frame,
15598 expressed in seconds. It's NAN if undefined.
15601 The PTS of the previously filtered video frame. It's NAN if undefined.
15603 @item prev_selected_pts
15604 The PTS of the last previously filtered video frame. It's NAN if undefined.
15606 @item prev_selected_t
15607 The PTS of the last previously selected video frame. It's NAN if undefined.
15610 The PTS of the first video frame in the video. It's NAN if undefined.
15613 The time of the first video frame in the video. It's NAN if undefined.
15615 @item pict_type @emph{(video only)}
15616 The type of the filtered frame. It can assume one of the following
15628 @item interlace_type @emph{(video only)}
15629 The frame interlace type. It can assume one of the following values:
15632 The frame is progressive (not interlaced).
15634 The frame is top-field-first.
15636 The frame is bottom-field-first.
15639 @item consumed_sample_n @emph{(audio only)}
15640 the number of selected samples before the current frame
15642 @item samples_n @emph{(audio only)}
15643 the number of samples in the current frame
15645 @item sample_rate @emph{(audio only)}
15646 the input sample rate
15649 This is 1 if the filtered frame is a key-frame, 0 otherwise.
15652 the position in the file of the filtered frame, -1 if the information
15653 is not available (e.g. for synthetic video)
15655 @item scene @emph{(video only)}
15656 value between 0 and 1 to indicate a new scene; a low value reflects a low
15657 probability for the current frame to introduce a new scene, while a higher
15658 value means the current frame is more likely to be one (see the example below)
15660 @item concatdec_select
15661 The concat demuxer can select only part of a concat input file by setting an
15662 inpoint and an outpoint, but the output packets may not be entirely contained
15663 in the selected interval. By using this variable, it is possible to skip frames
15664 generated by the concat demuxer which are not exactly contained in the selected
15667 This works by comparing the frame pts against the @var{lavf.concat.start_time}
15668 and the @var{lavf.concat.duration} packet metadata values which are also
15669 present in the decoded frames.
15671 The @var{concatdec_select} variable is -1 if the frame pts is at least
15672 start_time and either the duration metadata is missing or the frame pts is less
15673 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
15676 That basically means that an input frame is selected if its pts is within the
15677 interval set by the concat demuxer.
15681 The default value of the select expression is "1".
15683 @subsection Examples
15687 Select all frames in input:
15692 The example above is the same as:
15704 Select only I-frames:
15706 select='eq(pict_type\,I)'
15710 Select one frame every 100:
15712 select='not(mod(n\,100))'
15716 Select only frames contained in the 10-20 time interval:
15718 select=between(t\,10\,20)
15722 Select only I-frames contained in the 10-20 time interval:
15724 select=between(t\,10\,20)*eq(pict_type\,I)
15728 Select frames with a minimum distance of 10 seconds:
15730 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
15734 Use aselect to select only audio frames with samples number > 100:
15736 aselect='gt(samples_n\,100)'
15740 Create a mosaic of the first scenes:
15742 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
15745 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
15749 Send even and odd frames to separate outputs, and compose them:
15751 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
15755 Select useful frames from an ffconcat file which is using inpoints and
15756 outpoints but where the source files are not intra frame only.
15758 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
15762 @section sendcmd, asendcmd
15764 Send commands to filters in the filtergraph.
15766 These filters read commands to be sent to other filters in the
15769 @code{sendcmd} must be inserted between two video filters,
15770 @code{asendcmd} must be inserted between two audio filters, but apart
15771 from that they act the same way.
15773 The specification of commands can be provided in the filter arguments
15774 with the @var{commands} option, or in a file specified by the
15775 @var{filename} option.
15777 These filters accept the following options:
15780 Set the commands to be read and sent to the other filters.
15782 Set the filename of the commands to be read and sent to the other
15786 @subsection Commands syntax
15788 A commands description consists of a sequence of interval
15789 specifications, comprising a list of commands to be executed when a
15790 particular event related to that interval occurs. The occurring event
15791 is typically the current frame time entering or leaving a given time
15794 An interval is specified by the following syntax:
15796 @var{START}[-@var{END}] @var{COMMANDS};
15799 The time interval is specified by the @var{START} and @var{END} times.
15800 @var{END} is optional and defaults to the maximum time.
15802 The current frame time is considered within the specified interval if
15803 it is included in the interval [@var{START}, @var{END}), that is when
15804 the time is greater or equal to @var{START} and is lesser than
15807 @var{COMMANDS} consists of a sequence of one or more command
15808 specifications, separated by ",", relating to that interval. The
15809 syntax of a command specification is given by:
15811 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
15814 @var{FLAGS} is optional and specifies the type of events relating to
15815 the time interval which enable sending the specified command, and must
15816 be a non-null sequence of identifier flags separated by "+" or "|" and
15817 enclosed between "[" and "]".
15819 The following flags are recognized:
15822 The command is sent when the current frame timestamp enters the
15823 specified interval. In other words, the command is sent when the
15824 previous frame timestamp was not in the given interval, and the
15828 The command is sent when the current frame timestamp leaves the
15829 specified interval. In other words, the command is sent when the
15830 previous frame timestamp was in the given interval, and the
15834 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
15837 @var{TARGET} specifies the target of the command, usually the name of
15838 the filter class or a specific filter instance name.
15840 @var{COMMAND} specifies the name of the command for the target filter.
15842 @var{ARG} is optional and specifies the optional list of argument for
15843 the given @var{COMMAND}.
15845 Between one interval specification and another, whitespaces, or
15846 sequences of characters starting with @code{#} until the end of line,
15847 are ignored and can be used to annotate comments.
15849 A simplified BNF description of the commands specification syntax
15852 @var{COMMAND_FLAG} ::= "enter" | "leave"
15853 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
15854 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
15855 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
15856 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
15857 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
15860 @subsection Examples
15864 Specify audio tempo change at second 4:
15866 asendcmd=c='4.0 atempo tempo 1.5',atempo
15870 Specify a list of drawtext and hue commands in a file.
15872 # show text in the interval 5-10
15873 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
15874 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
15876 # desaturate the image in the interval 15-20
15877 15.0-20.0 [enter] hue s 0,
15878 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
15880 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
15882 # apply an exponential saturation fade-out effect, starting from time 25
15883 25 [enter] hue s exp(25-t)
15886 A filtergraph allowing to read and process the above command list
15887 stored in a file @file{test.cmd}, can be specified with:
15889 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
15894 @section setpts, asetpts
15896 Change the PTS (presentation timestamp) of the input frames.
15898 @code{setpts} works on video frames, @code{asetpts} on audio frames.
15900 This filter accepts the following options:
15905 The expression which is evaluated for each frame to construct its timestamp.
15909 The expression is evaluated through the eval API and can contain the following
15914 frame rate, only defined for constant frame-rate video
15917 The presentation timestamp in input
15920 The count of the input frame for video or the number of consumed samples,
15921 not including the current frame for audio, starting from 0.
15923 @item NB_CONSUMED_SAMPLES
15924 The number of consumed samples, not including the current frame (only
15927 @item NB_SAMPLES, S
15928 The number of samples in the current frame (only audio)
15930 @item SAMPLE_RATE, SR
15931 The audio sample rate.
15934 The PTS of the first frame.
15937 the time in seconds of the first frame
15940 State whether the current frame is interlaced.
15943 the time in seconds of the current frame
15946 original position in the file of the frame, or undefined if undefined
15947 for the current frame
15950 The previous input PTS.
15953 previous input time in seconds
15956 The previous output PTS.
15959 previous output time in seconds
15962 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
15966 The wallclock (RTC) time at the start of the movie in microseconds.
15969 The timebase of the input timestamps.
15973 @subsection Examples
15977 Start counting PTS from zero
15979 setpts=PTS-STARTPTS
15983 Apply fast motion effect:
15989 Apply slow motion effect:
15995 Set fixed rate of 25 frames per second:
16001 Set fixed rate 25 fps with some jitter:
16003 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
16007 Apply an offset of 10 seconds to the input PTS:
16013 Generate timestamps from a "live source" and rebase onto the current timebase:
16015 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
16019 Generate timestamps by counting samples:
16026 @section settb, asettb
16028 Set the timebase to use for the output frames timestamps.
16029 It is mainly useful for testing timebase configuration.
16031 It accepts the following parameters:
16036 The expression which is evaluated into the output timebase.
16040 The value for @option{tb} is an arithmetic expression representing a
16041 rational. The expression can contain the constants "AVTB" (the default
16042 timebase), "intb" (the input timebase) and "sr" (the sample rate,
16043 audio only). Default value is "intb".
16045 @subsection Examples
16049 Set the timebase to 1/25:
16055 Set the timebase to 1/10:
16061 Set the timebase to 1001/1000:
16067 Set the timebase to 2*intb:
16073 Set the default timebase value:
16080 Convert input audio to a video output representing frequency spectrum
16081 logarithmically using Brown-Puckette constant Q transform algorithm with
16082 direct frequency domain coefficient calculation (but the transform itself
16083 is not really constant Q, instead the Q factor is actually variable/clamped),
16084 with musical tone scale, from E0 to D#10.
16086 The filter accepts the following options:
16090 Specify the video size for the output. It must be even. For the syntax of this option,
16091 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16092 Default value is @code{1920x1080}.
16095 Set the output frame rate. Default value is @code{25}.
16098 Set the bargraph height. It must be even. Default value is @code{-1} which
16099 computes the bargraph height automatically.
16102 Set the axis height. It must be even. Default value is @code{-1} which computes
16103 the axis height automatically.
16106 Set the sonogram height. It must be even. Default value is @code{-1} which
16107 computes the sonogram height automatically.
16110 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
16111 instead. Default value is @code{1}.
16113 @item sono_v, volume
16114 Specify the sonogram volume expression. It can contain variables:
16117 the @var{bar_v} evaluated expression
16118 @item frequency, freq, f
16119 the frequency where it is evaluated
16120 @item timeclamp, tc
16121 the value of @var{timeclamp} option
16125 @item a_weighting(f)
16126 A-weighting of equal loudness
16127 @item b_weighting(f)
16128 B-weighting of equal loudness
16129 @item c_weighting(f)
16130 C-weighting of equal loudness.
16132 Default value is @code{16}.
16134 @item bar_v, volume2
16135 Specify the bargraph volume expression. It can contain variables:
16138 the @var{sono_v} evaluated expression
16139 @item frequency, freq, f
16140 the frequency where it is evaluated
16141 @item timeclamp, tc
16142 the value of @var{timeclamp} option
16146 @item a_weighting(f)
16147 A-weighting of equal loudness
16148 @item b_weighting(f)
16149 B-weighting of equal loudness
16150 @item c_weighting(f)
16151 C-weighting of equal loudness.
16153 Default value is @code{sono_v}.
16155 @item sono_g, gamma
16156 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
16157 higher gamma makes the spectrum having more range. Default value is @code{3}.
16158 Acceptable range is @code{[1, 7]}.
16160 @item bar_g, gamma2
16161 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
16164 @item timeclamp, tc
16165 Specify the transform timeclamp. At low frequency, there is trade-off between
16166 accuracy in time domain and frequency domain. If timeclamp is lower,
16167 event in time domain is represented more accurately (such as fast bass drum),
16168 otherwise event in frequency domain is represented more accurately
16169 (such as bass guitar). Acceptable range is @code{[0.1, 1]}. Default value is @code{0.17}.
16172 Specify the transform base frequency. Default value is @code{20.01523126408007475},
16173 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
16176 Specify the transform end frequency. Default value is @code{20495.59681441799654},
16177 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
16180 This option is deprecated and ignored.
16183 Specify the transform length in time domain. Use this option to control accuracy
16184 trade-off between time domain and frequency domain at every frequency sample.
16185 It can contain variables:
16187 @item frequency, freq, f
16188 the frequency where it is evaluated
16189 @item timeclamp, tc
16190 the value of @var{timeclamp} option.
16192 Default value is @code{384*tc/(384+tc*f)}.
16195 Specify the transform count for every video frame. Default value is @code{6}.
16196 Acceptable range is @code{[1, 30]}.
16199 Specify the transform count for every single pixel. Default value is @code{0},
16200 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
16203 Specify font file for use with freetype to draw the axis. If not specified,
16204 use embedded font. Note that drawing with font file or embedded font is not
16205 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
16209 Specify font color expression. This is arithmetic expression that should return
16210 integer value 0xRRGGBB. It can contain variables:
16212 @item frequency, freq, f
16213 the frequency where it is evaluated
16214 @item timeclamp, tc
16215 the value of @var{timeclamp} option
16220 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
16221 @item r(x), g(x), b(x)
16222 red, green, and blue value of intensity x.
16224 Default value is @code{st(0, (midi(f)-59.5)/12);
16225 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
16226 r(1-ld(1)) + b(ld(1))}.
16229 Specify image file to draw the axis. This option override @var{fontfile} and
16230 @var{fontcolor} option.
16233 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
16234 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
16235 Default value is @code{1}.
16239 @subsection Examples
16243 Playing audio while showing the spectrum:
16245 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
16249 Same as above, but with frame rate 30 fps:
16251 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
16255 Playing at 1280x720:
16257 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
16261 Disable sonogram display:
16267 A1 and its harmonics: A1, A2, (near)E3, A3:
16269 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),
16270 asplit[a][out1]; [a] showcqt [out0]'
16274 Same as above, but with more accuracy in frequency domain:
16276 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),
16277 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
16283 bar_v=10:sono_v=bar_v*a_weighting(f)
16287 Custom gamma, now spectrum is linear to the amplitude.
16293 Custom tlength equation:
16295 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)))'
16299 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
16301 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
16305 Custom frequency range with custom axis using image file:
16307 axisfile=myaxis.png:basefreq=40:endfreq=10000
16313 Convert input audio to video output representing the audio power spectrum.
16314 Audio amplitude is on Y-axis while frequency is on X-axis.
16316 The filter accepts the following options:
16320 Specify size of video. For the syntax of this option, check the
16321 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16322 Default is @code{1024x512}.
16326 This set how each frequency bin will be represented.
16328 It accepts the following values:
16334 Default is @code{bar}.
16337 Set amplitude scale.
16339 It accepts the following values:
16353 Default is @code{log}.
16356 Set frequency scale.
16358 It accepts the following values:
16367 Reverse logarithmic scale.
16369 Default is @code{lin}.
16374 It accepts the following values:
16390 Default is @code{w2048}
16393 Set windowing function.
16395 It accepts the following values:
16413 Default is @code{hanning}.
16416 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
16417 which means optimal overlap for selected window function will be picked.
16420 Set time averaging. Setting this to 0 will display current maximal peaks.
16421 Default is @code{1}, which means time averaging is disabled.
16424 Specify list of colors separated by space or by '|' which will be used to
16425 draw channel frequencies. Unrecognized or missing colors will be replaced
16429 Set channel display mode.
16431 It accepts the following values:
16436 Default is @code{combined}.
16440 @anchor{showspectrum}
16441 @section showspectrum
16443 Convert input audio to a video output, representing the audio frequency
16446 The filter accepts the following options:
16450 Specify the video size for the output. For the syntax of this option, check the
16451 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16452 Default value is @code{640x512}.
16455 Specify how the spectrum should slide along the window.
16457 It accepts the following values:
16460 the samples start again on the left when they reach the right
16462 the samples scroll from right to left
16464 the samples scroll from left to right
16466 frames are only produced when the samples reach the right
16469 Default value is @code{replace}.
16472 Specify display mode.
16474 It accepts the following values:
16477 all channels are displayed in the same row
16479 all channels are displayed in separate rows
16482 Default value is @samp{combined}.
16485 Specify display color mode.
16487 It accepts the following values:
16490 each channel is displayed in a separate color
16492 each channel is displayed using the same color scheme
16494 each channel is displayed using the rainbow color scheme
16496 each channel is displayed using the moreland color scheme
16498 each channel is displayed using the nebulae color scheme
16500 each channel is displayed using the fire color scheme
16502 each channel is displayed using the fiery color scheme
16504 each channel is displayed using the fruit color scheme
16506 each channel is displayed using the cool color scheme
16509 Default value is @samp{channel}.
16512 Specify scale used for calculating intensity color values.
16514 It accepts the following values:
16519 square root, default
16530 Default value is @samp{sqrt}.
16533 Set saturation modifier for displayed colors. Negative values provide
16534 alternative color scheme. @code{0} is no saturation at all.
16535 Saturation must be in [-10.0, 10.0] range.
16536 Default value is @code{1}.
16539 Set window function.
16541 It accepts the following values:
16561 Default value is @code{hann}.
16564 Set orientation of time vs frequency axis. Can be @code{vertical} or
16565 @code{horizontal}. Default is @code{vertical}.
16568 Set ratio of overlap window. Default value is @code{0}.
16569 When value is @code{1} overlap is set to recommended size for specific
16570 window function currently used.
16573 Set scale gain for calculating intensity color values.
16574 Default value is @code{1}.
16577 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
16580 Set color rotation, must be in [-1.0, 1.0] range.
16581 Default value is @code{0}.
16584 The usage is very similar to the showwaves filter; see the examples in that
16587 @subsection Examples
16591 Large window with logarithmic color scaling:
16593 showspectrum=s=1280x480:scale=log
16597 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
16599 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
16600 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
16604 @section showspectrumpic
16606 Convert input audio to a single video frame, representing the audio frequency
16609 The filter accepts the following options:
16613 Specify the video size for the output. For the syntax of this option, check the
16614 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16615 Default value is @code{4096x2048}.
16618 Specify display mode.
16620 It accepts the following values:
16623 all channels are displayed in the same row
16625 all channels are displayed in separate rows
16627 Default value is @samp{combined}.
16630 Specify display color mode.
16632 It accepts the following values:
16635 each channel is displayed in a separate color
16637 each channel is displayed using the same color scheme
16639 each channel is displayed using the rainbow color scheme
16641 each channel is displayed using the moreland color scheme
16643 each channel is displayed using the nebulae color scheme
16645 each channel is displayed using the fire color scheme
16647 each channel is displayed using the fiery color scheme
16649 each channel is displayed using the fruit color scheme
16651 each channel is displayed using the cool color scheme
16653 Default value is @samp{intensity}.
16656 Specify scale used for calculating intensity color values.
16658 It accepts the following values:
16663 square root, default
16673 Default value is @samp{log}.
16676 Set saturation modifier for displayed colors. Negative values provide
16677 alternative color scheme. @code{0} is no saturation at all.
16678 Saturation must be in [-10.0, 10.0] range.
16679 Default value is @code{1}.
16682 Set window function.
16684 It accepts the following values:
16703 Default value is @code{hann}.
16706 Set orientation of time vs frequency axis. Can be @code{vertical} or
16707 @code{horizontal}. Default is @code{vertical}.
16710 Set scale gain for calculating intensity color values.
16711 Default value is @code{1}.
16714 Draw time and frequency axes and legends. Default is enabled.
16717 Set color rotation, must be in [-1.0, 1.0] range.
16718 Default value is @code{0}.
16721 @subsection Examples
16725 Extract an audio spectrogram of a whole audio track
16726 in a 1024x1024 picture using @command{ffmpeg}:
16728 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
16732 @section showvolume
16734 Convert input audio volume to a video output.
16736 The filter accepts the following options:
16743 Set border width, allowed range is [0, 5]. Default is 1.
16746 Set channel width, allowed range is [80, 8192]. Default is 400.
16749 Set channel height, allowed range is [1, 900]. Default is 20.
16752 Set fade, allowed range is [0.001, 1]. Default is 0.95.
16755 Set volume color expression.
16757 The expression can use the following variables:
16761 Current max volume of channel in dB.
16764 Current channel number, starting from 0.
16768 If set, displays channel names. Default is enabled.
16771 If set, displays volume values. Default is enabled.
16774 Set orientation, can be @code{horizontal} or @code{vertical},
16775 default is @code{horizontal}.
16778 Set step size, allowed range s [0, 5]. Default is 0, which means
16784 Convert input audio to a video output, representing the samples waves.
16786 The filter accepts the following options:
16790 Specify the video size for the output. For the syntax of this option, check the
16791 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16792 Default value is @code{600x240}.
16797 Available values are:
16800 Draw a point for each sample.
16803 Draw a vertical line for each sample.
16806 Draw a point for each sample and a line between them.
16809 Draw a centered vertical line for each sample.
16812 Default value is @code{point}.
16815 Set the number of samples which are printed on the same column. A
16816 larger value will decrease the frame rate. Must be a positive
16817 integer. This option can be set only if the value for @var{rate}
16818 is not explicitly specified.
16821 Set the (approximate) output frame rate. This is done by setting the
16822 option @var{n}. Default value is "25".
16824 @item split_channels
16825 Set if channels should be drawn separately or overlap. Default value is 0.
16828 Set colors separated by '|' which are going to be used for drawing of each channel.
16831 Set amplitude scale. Can be linear @code{lin} or logarithmic @code{log}.
16836 @subsection Examples
16840 Output the input file audio and the corresponding video representation
16843 amovie=a.mp3,asplit[out0],showwaves[out1]
16847 Create a synthetic signal and show it with showwaves, forcing a
16848 frame rate of 30 frames per second:
16850 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
16854 @section showwavespic
16856 Convert input audio to a single video frame, representing the samples waves.
16858 The filter accepts the following options:
16862 Specify the video size for the output. For the syntax of this option, check the
16863 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16864 Default value is @code{600x240}.
16866 @item split_channels
16867 Set if channels should be drawn separately or overlap. Default value is 0.
16870 Set colors separated by '|' which are going to be used for drawing of each channel.
16873 Set amplitude scale. Can be linear @code{lin} or logarithmic @code{log}.
16877 @subsection Examples
16881 Extract a channel split representation of the wave form of a whole audio track
16882 in a 1024x800 picture using @command{ffmpeg}:
16884 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
16888 @section spectrumsynth
16890 Sythesize audio from 2 input video spectrums, first input stream represents
16891 magnitude across time and second represents phase across time.
16892 The filter will transform from frequency domain as displayed in videos back
16893 to time domain as presented in audio output.
16895 This filter is primarly created for reversing processed @ref{showspectrum}
16896 filter outputs, but can synthesize sound from other spectrograms too.
16897 But in such case results are going to be poor if the phase data is not
16898 available, because in such cases phase data need to be recreated, usually
16899 its just recreated from random noise.
16900 For best results use gray only output (@code{channel} color mode in
16901 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
16902 @code{lin} scale for phase video. To produce phase, for 2nd video, use
16903 @code{data} option. Inputs videos should generally use @code{fullframe}
16904 slide mode as that saves resources needed for decoding video.
16906 The filter accepts the following options:
16910 Specify sample rate of output audio, the sample rate of audio from which
16911 spectrum was generated may differ.
16914 Set number of channels represented in input video spectrums.
16917 Set scale which was used when generating magnitude input spectrum.
16918 Can be @code{lin} or @code{log}. Default is @code{log}.
16921 Set slide which was used when generating inputs spectrums.
16922 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
16923 Default is @code{fullframe}.
16926 Set window function used for resynthesis.
16929 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
16930 which means optimal overlap for selected window function will be picked.
16933 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
16934 Default is @code{vertical}.
16937 @subsection Examples
16941 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
16942 then resynthesize videos back to audio with spectrumsynth:
16944 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
16945 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
16946 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
16950 @section split, asplit
16952 Split input into several identical outputs.
16954 @code{asplit} works with audio input, @code{split} with video.
16956 The filter accepts a single parameter which specifies the number of outputs. If
16957 unspecified, it defaults to 2.
16959 @subsection Examples
16963 Create two separate outputs from the same input:
16965 [in] split [out0][out1]
16969 To create 3 or more outputs, you need to specify the number of
16972 [in] asplit=3 [out0][out1][out2]
16976 Create two separate outputs from the same input, one cropped and
16979 [in] split [splitout1][splitout2];
16980 [splitout1] crop=100:100:0:0 [cropout];
16981 [splitout2] pad=200:200:100:100 [padout];
16985 Create 5 copies of the input audio with @command{ffmpeg}:
16987 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
16993 Receive commands sent through a libzmq client, and forward them to
16994 filters in the filtergraph.
16996 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
16997 must be inserted between two video filters, @code{azmq} between two
17000 To enable these filters you need to install the libzmq library and
17001 headers and configure FFmpeg with @code{--enable-libzmq}.
17003 For more information about libzmq see:
17004 @url{http://www.zeromq.org/}
17006 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
17007 receives messages sent through a network interface defined by the
17008 @option{bind_address} option.
17010 The received message must be in the form:
17012 @var{TARGET} @var{COMMAND} [@var{ARG}]
17015 @var{TARGET} specifies the target of the command, usually the name of
17016 the filter class or a specific filter instance name.
17018 @var{COMMAND} specifies the name of the command for the target filter.
17020 @var{ARG} is optional and specifies the optional argument list for the
17021 given @var{COMMAND}.
17023 Upon reception, the message is processed and the corresponding command
17024 is injected into the filtergraph. Depending on the result, the filter
17025 will send a reply to the client, adopting the format:
17027 @var{ERROR_CODE} @var{ERROR_REASON}
17031 @var{MESSAGE} is optional.
17033 @subsection Examples
17035 Look at @file{tools/zmqsend} for an example of a zmq client which can
17036 be used to send commands processed by these filters.
17038 Consider the following filtergraph generated by @command{ffplay}
17040 ffplay -dumpgraph 1 -f lavfi "
17041 color=s=100x100:c=red [l];
17042 color=s=100x100:c=blue [r];
17043 nullsrc=s=200x100, zmq [bg];
17044 [bg][l] overlay [bg+l];
17045 [bg+l][r] overlay=x=100 "
17048 To change the color of the left side of the video, the following
17049 command can be used:
17051 echo Parsed_color_0 c yellow | tools/zmqsend
17054 To change the right side:
17056 echo Parsed_color_1 c pink | tools/zmqsend
17059 @c man end MULTIMEDIA FILTERS
17061 @chapter Multimedia Sources
17062 @c man begin MULTIMEDIA SOURCES
17064 Below is a description of the currently available multimedia sources.
17068 This is the same as @ref{movie} source, except it selects an audio
17074 Read audio and/or video stream(s) from a movie container.
17076 It accepts the following parameters:
17080 The name of the resource to read (not necessarily a file; it can also be a
17081 device or a stream accessed through some protocol).
17083 @item format_name, f
17084 Specifies the format assumed for the movie to read, and can be either
17085 the name of a container or an input device. If not specified, the
17086 format is guessed from @var{movie_name} or by probing.
17088 @item seek_point, sp
17089 Specifies the seek point in seconds. The frames will be output
17090 starting from this seek point. The parameter is evaluated with
17091 @code{av_strtod}, so the numerical value may be suffixed by an IS
17092 postfix. The default value is "0".
17095 Specifies the streams to read. Several streams can be specified,
17096 separated by "+". The source will then have as many outputs, in the
17097 same order. The syntax is explained in the ``Stream specifiers''
17098 section in the ffmpeg manual. Two special names, "dv" and "da" specify
17099 respectively the default (best suited) video and audio stream. Default
17100 is "dv", or "da" if the filter is called as "amovie".
17102 @item stream_index, si
17103 Specifies the index of the video stream to read. If the value is -1,
17104 the most suitable video stream will be automatically selected. The default
17105 value is "-1". Deprecated. If the filter is called "amovie", it will select
17106 audio instead of video.
17109 Specifies how many times to read the stream in sequence.
17110 If the value is less than 1, the stream will be read again and again.
17111 Default value is "1".
17113 Note that when the movie is looped the source timestamps are not
17114 changed, so it will generate non monotonically increasing timestamps.
17116 @item discontinuity
17117 Specifies the time difference between frames above which the point is
17118 considered a timestamp discontinuity which is removed by adjusting the later
17122 It allows overlaying a second video on top of the main input of
17123 a filtergraph, as shown in this graph:
17125 input -----------> deltapts0 --> overlay --> output
17128 movie --> scale--> deltapts1 -------+
17130 @subsection Examples
17134 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
17135 on top of the input labelled "in":
17137 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
17138 [in] setpts=PTS-STARTPTS [main];
17139 [main][over] overlay=16:16 [out]
17143 Read from a video4linux2 device, and overlay it on top of the input
17146 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
17147 [in] setpts=PTS-STARTPTS [main];
17148 [main][over] overlay=16:16 [out]
17152 Read the first video stream and the audio stream with id 0x81 from
17153 dvd.vob; the video is connected to the pad named "video" and the audio is
17154 connected to the pad named "audio":
17156 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
17160 @subsection Commands
17162 Both movie and amovie support the following commands:
17165 Perform seek using "av_seek_frame".
17166 The syntax is: seek @var{stream_index}|@var{timestamp}|@var{flags}
17169 @var{stream_index}: If stream_index is -1, a default
17170 stream is selected, and @var{timestamp} is automatically converted
17171 from AV_TIME_BASE units to the stream specific time_base.
17173 @var{timestamp}: Timestamp in AVStream.time_base units
17174 or, if no stream is specified, in AV_TIME_BASE units.
17176 @var{flags}: Flags which select direction and seeking mode.
17180 Get movie duration in AV_TIME_BASE units.
17184 @c man end MULTIMEDIA SOURCES