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, 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 in output the top half of the video is mirrored
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 the 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", a filter with no
112 output pads is called a "sink".
114 @anchor{Filtergraph syntax}
115 @section Filtergraph syntax
117 A filtergraph can be represented using a textual representation, which is
118 recognized by the @option{-filter}/@option{-vf} and @option{-filter_complex}
119 options in @command{ffmpeg} and @option{-vf} in @command{ffplay}, and by the
120 @code{avfilter_graph_parse()}/@code{avfilter_graph_parse2()} function defined in
121 @file{libavfilter/avfilter.h}.
123 A filterchain consists of a sequence of connected filters, each one
124 connected to the previous one in the sequence. A filterchain is
125 represented by a list of ","-separated filter descriptions.
127 A filtergraph consists of a sequence of filterchains. A sequence of
128 filterchains is represented by a list of ";"-separated filterchain
131 A filter is represented by a string of the form:
132 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
134 @var{filter_name} is the name of the filter class of which the
135 described filter is an instance of, and has to be the name of one of
136 the filter classes registered in the program.
137 The name of the filter class is optionally followed by a string
140 @var{arguments} is a string which contains the parameters used to
141 initialize the filter instance. It may have one of the following forms:
145 A ':'-separated list of @var{key=value} pairs.
148 A ':'-separated list of @var{value}. In this case, the keys are assumed to be
149 the option names in the order they are declared. E.g. the @code{fade} filter
150 declares three options in this order -- @option{type}, @option{start_frame} and
151 @option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
152 @var{in} is assigned to the option @option{type}, @var{0} to
153 @option{start_frame} and @var{30} to @option{nb_frames}.
156 A ':'-separated list of mixed direct @var{value} and long @var{key=value}
157 pairs. The direct @var{value} must precede the @var{key=value} pairs, and
158 follow the same constraints order of the previous point. The following
159 @var{key=value} pairs can be set in any preferred order.
163 If the option value itself is a list of items (e.g. the @code{format} filter
164 takes a list of pixel formats), the items in the list are usually separated by
167 The list of arguments can be quoted using the character "'" as initial
168 and ending mark, and the character '\' for escaping the characters
169 within the quoted text; otherwise the argument string is considered
170 terminated when the next special character (belonging to the set
171 "[]=;,") is encountered.
173 The name and arguments of the filter are optionally preceded and
174 followed by a list of link labels.
175 A link label allows to name a link and associate it to a filter output
176 or input pad. The preceding labels @var{in_link_1}
177 ... @var{in_link_N}, are associated to the filter input pads,
178 the following labels @var{out_link_1} ... @var{out_link_M}, are
179 associated to the output pads.
181 When two link labels with the same name are found in the
182 filtergraph, a link between the corresponding input and output pad is
185 If an output pad is not labelled, it is linked by default to the first
186 unlabelled input pad of the next filter in the filterchain.
187 For example in the filterchain:
189 nullsrc, split[L1], [L2]overlay, nullsink
191 the split filter instance has two output pads, and the overlay filter
192 instance two input pads. The first output pad of split is labelled
193 "L1", the first input pad of overlay is labelled "L2", and the second
194 output pad of split is linked to the second input pad of overlay,
195 which are both unlabelled.
197 In a complete filterchain all the unlabelled filter input and output
198 pads must be connected. A filtergraph is considered valid if all the
199 filter input and output pads of all the filterchains are connected.
201 Libavfilter will automatically insert @ref{scale} filters where format
202 conversion is required. It is possible to specify swscale flags
203 for those automatically inserted scalers by prepending
204 @code{sws_flags=@var{flags};}
205 to the filtergraph description.
207 Follows a BNF description for the filtergraph syntax:
209 @var{NAME} ::= sequence of alphanumeric characters and '_'
210 @var{LINKLABEL} ::= "[" @var{NAME} "]"
211 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
212 @var{FILTER_ARGUMENTS} ::= sequence of chars (eventually quoted)
213 @var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
214 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
215 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
218 @section Notes on filtergraph escaping
220 Filtergraph description composition entails several levels of
221 escaping. See @ref{quoting_and_escaping,,the "Quoting and escaping"
222 section in the ffmpeg-utils(1) manual,ffmpeg-utils} for more
223 information about the employed escaping procedure.
225 A first level escaping affects the content of each filter option
226 value, which may contain the special character @code{:} used to
227 separate values, or one of the escaping characters @code{\'}.
229 A second level escaping affects the whole filter description, which
230 may contain the escaping characters @code{\'} or the special
231 characters @code{[],;} used by the filtergraph description.
233 Finally, when you specify a filtergraph on a shell commandline, you
234 need to perform a third level escaping for the shell special
235 characters contained within it.
237 For example, consider the following string to be embedded in
238 the @ref{drawtext} filter description @option{text} value:
240 this is a 'string': may contain one, or more, special characters
243 This string contains the @code{'} special escaping character, and the
244 @code{:} special character, so it needs to be escaped in this way:
246 text=this is a \'string\'\: may contain one, or more, special characters
249 A second level of escaping is required when embedding the filter
250 description in a filtergraph description, in order to escape all the
251 filtergraph special characters. Thus the example above becomes:
253 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
255 (note that in addition to the @code{\'} escaping special characters,
256 also @code{,} needs to be escaped).
258 Finally an additional level of escaping is needed when writing the
259 filtergraph description in a shell command, which depends on the
260 escaping rules of the adopted shell. For example, assuming that
261 @code{\} is special and needs to be escaped with another @code{\}, the
262 previous string will finally result in:
264 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
267 @chapter Timeline editing
269 Some filters support a generic @option{enable} option. For the filters
270 supporting timeline editing, this option can be set to an expression which is
271 evaluated before sending a frame to the filter. If the evaluation is non-zero,
272 the filter will be enabled, otherwise the frame will be sent unchanged to the
273 next filter in the filtergraph.
275 The expression accepts the following values:
278 timestamp expressed in seconds, NAN if the input timestamp is unknown
281 sequential number of the input frame, starting from 0
284 the position in the file of the input frame, NAN if unknown
287 Additionally, these filters support an @option{enable} command that can be used
288 to re-define the expression.
290 Like any other filtering option, the @option{enable} option follows the same
293 For example, to enable a blur filter (@ref{smartblur}) from 10 seconds to 3
294 minutes, and a @ref{curves} filter starting at 3 seconds:
296 smartblur = enable='between(t,10,3*60)',
297 curves = enable='gte(t,3)' : preset=cross_process
300 @c man end FILTERGRAPH DESCRIPTION
302 @chapter Audio Filters
303 @c man begin AUDIO FILTERS
305 When you configure your FFmpeg build, you can disable any of the
306 existing filters using @code{--disable-filters}.
307 The configure output will show the audio filters included in your
310 Below is a description of the currently available audio filters.
314 Convert the input audio format to the specified formats.
316 @emph{This filter is deprecated. Use @ref{aformat} instead.}
318 The filter accepts a string of the form:
319 "@var{sample_format}:@var{channel_layout}".
321 @var{sample_format} specifies the sample format, and can be a string or the
322 corresponding numeric value defined in @file{libavutil/samplefmt.h}. Use 'p'
323 suffix for a planar sample format.
325 @var{channel_layout} specifies the channel layout, and can be a string
326 or the corresponding number value defined in @file{libavutil/channel_layout.h}.
328 The special parameter "auto", signifies that the filter will
329 automatically select the output format depending on the output filter.
335 Convert input to float, planar, stereo:
341 Convert input to unsigned 8-bit, automatically select out channel layout:
349 Delay one or more audio channels.
351 Samples in delayed channel are filled with silence.
353 The filter accepts the following option:
357 Set list of delays in milliseconds for each channel separated by '|'.
358 At least one delay greater than 0 should be provided.
359 Unused delays will be silently ignored. If number of given delays is
360 smaller than number of channels all remaining channels will not be delayed.
367 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
368 the second channel (and any other channels that may be present) unchanged.
376 Apply echoing to the input audio.
378 Echoes are reflected sound and can occur naturally amongst mountains
379 (and sometimes large buildings) when talking or shouting; digital echo
380 effects emulate this behaviour and are often used to help fill out the
381 sound of a single instrument or vocal. The time difference between the
382 original signal and the reflection is the @code{delay}, and the
383 loudness of the reflected signal is the @code{decay}.
384 Multiple echoes can have different delays and decays.
386 A description of the accepted parameters follows.
390 Set input gain of reflected signal. Default is @code{0.6}.
393 Set output gain of reflected signal. Default is @code{0.3}.
396 Set list of time intervals in milliseconds between original signal and reflections
397 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
398 Default is @code{1000}.
401 Set list of loudnesses of reflected signals separated by '|'.
402 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
403 Default is @code{0.5}.
410 Make it sound as if there are twice as many instruments as are actually playing:
412 aecho=0.8:0.88:60:0.4
416 If delay is very short, then it sound like a (metallic) robot playing music:
422 A longer delay will sound like an open air concert in the mountains:
424 aecho=0.8:0.9:1000:0.3
428 Same as above but with one more mountain:
430 aecho=0.8:0.9:1000|1800:0.3|0.25
436 Modify an audio signal according to the specified expressions.
438 This filter accepts one or more expressions (one for each channel),
439 which are evaluated and used to modify a corresponding audio signal.
441 This filter accepts the following options:
445 Set the '|'-separated expressions list for each separate channel. If
446 the number of input channels is greater than the number of
447 expressions, the last specified expression is used for the remaining
450 @item channel_layout, c
451 Set output channel layout. If not specified, the channel layout is
452 specified by the number of expressions. If set to @samp{same}, it will
453 use by default the same input channel layout.
456 Each expression in @var{exprs} can contain the following constants and functions:
460 channel number of the current expression
463 number of the evaluated sample, starting from 0
469 time of the evaluated sample expressed in seconds
472 @item nb_out_channels
473 input and output number of channels
476 the value of input channel with number @var{CH}
479 Note: this filter is slow. For faster processing you should use a
488 aeval=val(ch)/2:c=same
492 Invert phase of the second channel:
500 Apply fade-in/out effect to input audio.
502 A description of the accepted parameters follows.
506 Specify the effect type, can be either @code{in} for fade-in, or
507 @code{out} for a fade-out effect. Default is @code{in}.
509 @item start_sample, ss
510 Specify the number of the start sample for starting to apply the fade
511 effect. Default is 0.
514 Specify the number of samples for which the fade effect has to last. At
515 the end of the fade-in effect the output audio will have the same
516 volume as the input audio, at the end of the fade-out transition
517 the output audio will be silence. Default is 44100.
520 Specify time for starting to apply the fade effect. Default is 0.
521 The accepted syntax is:
523 [-]HH[:MM[:SS[.m...]]]
526 See also the function @code{av_parse_time()}.
527 If set this option is used instead of @var{start_sample} one.
530 Specify the duration for which the fade effect has to last. Default is 0.
531 The accepted syntax is:
533 [-]HH[:MM[:SS[.m...]]]
536 See also the function @code{av_parse_time()}.
537 At the end of the fade-in effect the output audio will have the same
538 volume as the input audio, at the end of the fade-out transition
539 the output audio will be silence.
540 If set this option is used instead of @var{nb_samples} one.
543 Set curve for fade transition.
545 It accepts the following values:
548 select triangular, linear slope (default)
550 select quarter of sine wave
552 select half of sine wave
554 select exponential sine wave
558 select inverted parabola
574 Fade in first 15 seconds of audio:
580 Fade out last 25 seconds of a 900 seconds audio:
582 afade=t=out:st=875:d=25
589 Set output format constraints for the input audio. The framework will
590 negotiate the most appropriate format to minimize conversions.
592 The filter accepts the following named parameters:
596 A '|'-separated list of requested sample formats.
599 A '|'-separated list of requested sample rates.
601 @item channel_layouts
602 A '|'-separated list of requested channel layouts.
604 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
605 for the required syntax.
608 If a parameter is omitted, all values are allowed.
610 For example to force the output to either unsigned 8-bit or signed 16-bit stereo:
612 aformat=sample_fmts=u8|s16:channel_layouts=stereo
617 Apply a two-pole all-pass filter with central frequency (in Hz)
618 @var{frequency}, and filter-width @var{width}.
619 An all-pass filter changes the audio's frequency to phase relationship
620 without changing its frequency to amplitude relationship.
622 The filter accepts the following options:
629 Set method to specify band-width of filter.
642 Specify the band-width of a filter in width_type units.
647 Merge two or more audio streams into a single multi-channel stream.
649 The filter accepts the following options:
654 Set the number of inputs. Default is 2.
658 If the channel layouts of the inputs are disjoint, and therefore compatible,
659 the channel layout of the output will be set accordingly and the channels
660 will be reordered as necessary. If the channel layouts of the inputs are not
661 disjoint, the output will have all the channels of the first input then all
662 the channels of the second input, in that order, and the channel layout of
663 the output will be the default value corresponding to the total number of
666 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
667 is FC+BL+BR, then the output will be in 5.1, with the channels in the
668 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
669 first input, b1 is the first channel of the second input).
671 On the other hand, if both input are in stereo, the output channels will be
672 in the default order: a1, a2, b1, b2, and the channel layout will be
673 arbitrarily set to 4.0, which may or may not be the expected value.
675 All inputs must have the same sample rate, and format.
677 If inputs do not have the same duration, the output will stop with the
684 Merge two mono files into a stereo stream:
686 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
690 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
692 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
698 Mixes multiple audio inputs into a single output.
702 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
704 will mix 3 input audio streams to a single output with the same duration as the
705 first input and a dropout transition time of 3 seconds.
707 The filter accepts the following named parameters:
711 Number of inputs. If unspecified, it defaults to 2.
714 How to determine the end-of-stream.
718 Duration of longest input. (default)
721 Duration of shortest input.
724 Duration of first input.
728 @item dropout_transition
729 Transition time, in seconds, for volume renormalization when an input
730 stream ends. The default value is 2 seconds.
736 Pass the audio source unchanged to the output.
740 Pad the end of a audio stream with silence, this can be used together with
741 -shortest to extend audio streams to the same length as the video stream.
744 Add a phasing effect to the input audio.
746 A phaser filter creates series of peaks and troughs in the frequency spectrum.
747 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
749 A description of the accepted parameters follows.
753 Set input gain. Default is 0.4.
756 Set output gain. Default is 0.74
759 Set delay in milliseconds. Default is 3.0.
762 Set decay. Default is 0.4.
765 Set modulation speed in Hz. Default is 0.5.
768 Set modulation type. Default is triangular.
770 It accepts the following values:
780 Resample the input audio to the specified parameters, using the
781 libswresample library. If none are specified then the filter will
782 automatically convert between its input and output.
784 This filter is also able to stretch/squeeze the audio data to make it match
785 the timestamps or to inject silence / cut out audio to make it match the
786 timestamps, do a combination of both or do neither.
788 The filter accepts the syntax
789 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
790 expresses a sample rate and @var{resampler_options} is a list of
791 @var{key}=@var{value} pairs, separated by ":". See the
792 ffmpeg-resampler manual for the complete list of supported options.
798 Resample the input audio to 44100Hz:
804 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
805 samples per second compensation:
811 @section asetnsamples
813 Set the number of samples per each output audio frame.
815 The last output packet may contain a different number of samples, as
816 the filter will flush all the remaining samples when the input audio
819 The filter accepts the following options:
823 @item nb_out_samples, n
824 Set the number of frames per each output audio frame. The number is
825 intended as the number of samples @emph{per each channel}.
826 Default value is 1024.
829 If set to 1, the filter will pad the last audio frame with zeroes, so
830 that the last frame will contain the same number of samples as the
831 previous ones. Default value is 1.
834 For example, to set the number of per-frame samples to 1234 and
835 disable padding for the last frame, use:
837 asetnsamples=n=1234:p=0
842 Set the sample rate without altering the PCM data.
843 This will result in a change of speed and pitch.
845 The filter accepts the following options:
849 Set the output sample rate. Default is 44100 Hz.
854 Show a line containing various information for each input audio frame.
855 The input audio is not modified.
857 The shown line contains a sequence of key/value pairs of the form
858 @var{key}:@var{value}.
860 A description of each shown parameter follows:
864 sequential number of the input frame, starting from 0
867 Presentation timestamp of the input frame, in time base units; the time base
868 depends on the filter input pad, and is usually 1/@var{sample_rate}.
871 presentation timestamp of the input frame in seconds
874 position of the frame in the input stream, -1 if this information in
875 unavailable and/or meaningless (for example in case of synthetic audio)
884 sample rate for the audio frame
887 number of samples (per channel) in the frame
890 Adler-32 checksum (printed in hexadecimal) of the audio data. For planar audio
891 the data is treated as if all the planes were concatenated.
893 @item plane_checksums
894 A list of Adler-32 checksums for each data plane.
899 Display time domain statistical information about the audio channels.
900 Statistics are calculated and displayed for each audio channel and,
901 where applicable, an overall figure is also given.
903 The filter accepts the following option:
906 Short window length in seconds, used for peak and trough RMS measurement.
907 Default is @code{0.05} (50 miliseconds). Allowed range is @code{[0.1 - 10]}.
910 A description of each shown parameter follows:
914 Mean amplitude displacement from zero.
917 Minimal sample level.
920 Maximal sample level.
924 Standard peak and RMS level measured in dBFS.
928 Peak and trough values for RMS level measured over a short window.
931 Standard ratio of peak to RMS level (note: not in dB).
934 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
935 (i.e. either @var{Min level} or @var{Max level}).
938 Number of occasions (not the number of samples) that the signal attained either
939 @var{Min level} or @var{Max level}.
944 Forward two audio streams and control the order the buffers are forwarded.
946 The filter accepts the following options:
950 Set the expression deciding which stream should be
951 forwarded next: if the result is negative, the first stream is forwarded; if
952 the result is positive or zero, the second stream is forwarded. It can use
953 the following variables:
957 number of buffers forwarded so far on each stream
959 number of samples forwarded so far on each stream
961 current timestamp of each stream
964 The default value is @code{t1-t2}, which means to always forward the stream
965 that has a smaller timestamp.
970 Stress-test @code{amerge} by randomly sending buffers on the wrong
971 input, while avoiding too much of a desynchronization:
973 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
974 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
980 Synchronize audio data with timestamps by squeezing/stretching it and/or
981 dropping samples/adding silence when needed.
983 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
985 The filter accepts the following named parameters:
989 Enable stretching/squeezing the data to make it match the timestamps. Disabled
990 by default. When disabled, time gaps are covered with silence.
993 Minimum difference between timestamps and audio data (in seconds) to trigger
994 adding/dropping samples. Default value is 0.1. If you get non-perfect sync with
995 this filter, try setting this parameter to 0.
998 Maximum compensation in samples per second. Relevant only with compensate=1.
1002 Assume the first pts should be this value. The time base is 1 / sample rate.
1003 This allows for padding/trimming at the start of stream. By default, no
1004 assumption is made about the first frame's expected pts, so no padding or
1005 trimming is done. For example, this could be set to 0 to pad the beginning with
1006 silence if an audio stream starts after the video stream or to trim any samples
1007 with a negative pts due to encoder delay.
1015 The filter accepts exactly one parameter, the audio tempo. If not
1016 specified then the filter will assume nominal 1.0 tempo. Tempo must
1017 be in the [0.5, 2.0] range.
1019 @subsection Examples
1023 Slow down audio to 80% tempo:
1029 To speed up audio to 125% tempo:
1037 Trim the input so that the output contains one continuous subpart of the input.
1039 This filter accepts the following options:
1042 Specify time of the start of the kept section, i.e. the audio sample
1043 with the timestamp @var{start} will be the first sample in the output.
1046 Specify time of the first audio sample that will be dropped, i.e. the
1047 audio sample immediately preceding the one with the timestamp @var{end} will be
1048 the last sample in the output.
1051 Same as @var{start}, except this option sets the start timestamp in samples
1055 Same as @var{end}, except this option sets the end timestamp in samples instead
1059 Specify maximum duration of the output.
1062 Number of the first sample that should be passed to output.
1065 Number of the first sample that should be dropped.
1068 @option{start}, @option{end}, @option{duration} are expressed as time
1069 duration specifications, check the "Time duration" section in the
1070 ffmpeg-utils manual.
1072 Note that the first two sets of the start/end options and the @option{duration}
1073 option look at the frame timestamp, while the _sample options simply count the
1074 samples that pass through the filter. So start/end_pts and start/end_sample will
1075 give different results when the timestamps are wrong, inexact or do not start at
1076 zero. Also note that this filter does not modify the timestamps. If you wish
1077 that the output timestamps start at zero, insert the asetpts filter after the
1080 If multiple start or end options are set, this filter tries to be greedy and
1081 keep all samples that match at least one of the specified constraints. To keep
1082 only the part that matches all the constraints at once, chain multiple atrim
1085 The defaults are such that all the input is kept. So it is possible to set e.g.
1086 just the end values to keep everything before the specified time.
1091 drop everything except the second minute of input
1093 ffmpeg -i INPUT -af atrim=60:120
1097 keep only the first 1000 samples
1099 ffmpeg -i INPUT -af atrim=end_sample=1000
1106 Apply a two-pole Butterworth band-pass filter with central
1107 frequency @var{frequency}, and (3dB-point) band-width width.
1108 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1109 instead of the default: constant 0dB peak gain.
1110 The filter roll off at 6dB per octave (20dB per decade).
1112 The filter accepts the following options:
1116 Set the filter's central frequency. Default is @code{3000}.
1119 Constant skirt gain if set to 1. Defaults to 0.
1122 Set method to specify band-width of filter.
1135 Specify the band-width of a filter in width_type units.
1140 Apply a two-pole Butterworth band-reject filter with central
1141 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1142 The filter roll off at 6dB per octave (20dB per decade).
1144 The filter accepts the following options:
1148 Set the filter's central frequency. Default is @code{3000}.
1151 Set method to specify band-width of filter.
1164 Specify the band-width of a filter in width_type units.
1169 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1170 shelving filter with a response similar to that of a standard
1171 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1173 The filter accepts the following options:
1177 Give the gain at 0 Hz. Its useful range is about -20
1178 (for a large cut) to +20 (for a large boost).
1179 Beware of clipping when using a positive gain.
1182 Set the filter's central frequency and so can be used
1183 to extend or reduce the frequency range to be boosted or cut.
1184 The default value is @code{100} Hz.
1187 Set method to specify band-width of filter.
1200 Determine how steep is the filter's shelf transition.
1205 Apply a biquad IIR filter with the given coefficients.
1206 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1207 are the numerator and denominator coefficients respectively.
1211 Remap input channels to new locations.
1213 This filter accepts the following named parameters:
1215 @item channel_layout
1216 Channel layout of the output stream.
1219 Map channels from input to output. The argument is a '|'-separated list of
1220 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1221 @var{in_channel} form. @var{in_channel} can be either the name of the input
1222 channel (e.g. FL for front left) or its index in the input channel layout.
1223 @var{out_channel} is the name of the output channel or its index in the output
1224 channel layout. If @var{out_channel} is not given then it is implicitly an
1225 index, starting with zero and increasing by one for each mapping.
1228 If no mapping is present, the filter will implicitly map input channels to
1229 output channels preserving index.
1231 For example, assuming a 5.1+downmix input MOV file
1233 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1235 will create an output WAV file tagged as stereo from the downmix channels of
1238 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1240 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:channel_layout=5.1' out.wav
1243 @section channelsplit
1245 Split each channel in input audio stream into a separate output stream.
1247 This filter accepts the following named parameters:
1249 @item channel_layout
1250 Channel layout of the input stream. Default is "stereo".
1253 For example, assuming a stereo input MP3 file
1255 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1257 will create an output Matroska file with two audio streams, one containing only
1258 the left channel and the other the right channel.
1260 To split a 5.1 WAV file into per-channel files
1262 ffmpeg -i in.wav -filter_complex
1263 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1264 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1265 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1271 Compress or expand audio dynamic range.
1273 A description of the accepted options follows.
1278 Set list of times in seconds for each channel over which the instantaneous
1279 level of the input signal is averaged to determine its volume.
1280 @option{attacks} refers to increase of volume and @option{decays} refers
1281 to decrease of volume.
1282 For most situations, the attack time (response to the audio getting louder)
1283 should be shorter than the decay time because the human ear is more sensitive
1284 to sudden loud audio than sudden soft audio.
1285 Typical value for attack is @code{0.3} seconds and for decay @code{0.8}
1289 Set list of points for transfer function, specified in dB relative to maximum
1290 possible signal amplitude.
1291 Each key points list need to be defined using the following syntax:
1292 @code{x0/y0 x1/y1 x2/y2 ...}.
1294 The input values must be in strictly increasing order but the transfer
1295 function does not have to be monotonically rising.
1296 The point @code{0/0} is assumed but may be overridden (by @code{0/out-dBn}).
1297 Typical values for the transfer function are @code{-70/-70 -60/-20}.
1300 Set amount for which the points at where adjacent line segments on the
1301 transfer function meet will be rounded. Defaults is @code{0.01}.
1304 Set additional gain in dB to be applied at all points on the transfer function
1305 and allows easy adjustment of the overall gain.
1306 Default is @code{0}.
1309 Set initial volume in dB to be assumed for each channel when filtering starts.
1310 This permits the user to supply a nominal level initially, so that,
1311 for example, a very large gain is not applied to initial signal levels before
1312 the companding has begun to operate. A typical value for audio which is
1313 initially quiet is -90 dB. Default is @code{0}.
1316 Set delay in seconds. Default is @code{0}. The input audio
1317 is analysed immediately, but audio is delayed before being fed to the
1318 volume adjuster. Specifying a delay approximately equal to the attack/decay
1319 times allows the filter to effectively operate in predictive rather than
1323 @subsection Examples
1326 Make music with both quiet and loud passages suitable for listening
1327 in a noisy environment:
1329 compand=.3 .3:1 1:-90/-60 -60/-40 -40/-30 -20/-20:6:0:-90:0.2
1333 Noise-gate for when the noise is at a lower level than the signal:
1335 compand=.1 .1:.2 .2:-900/-900 -50.1/-900 -50/-50:.01:0:-90:.1
1339 Here is another noise-gate, this time for when the noise is at a higher level
1340 than the signal (making it, in some ways, similar to squelch):
1342 compand=.1 .1:.1 .1:-45.1/-45.1 -45/-900 0/-900:.01:45:-90:.1
1348 Make audio easier to listen to on headphones.
1350 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
1351 so that when listened to on headphones the stereo image is moved from
1352 inside your head (standard for headphones) to outside and in front of
1353 the listener (standard for speakers).
1359 Apply a two-pole peaking equalisation (EQ) filter. With this
1360 filter, the signal-level at and around a selected frequency can
1361 be increased or decreased, whilst (unlike bandpass and bandreject
1362 filters) that at all other frequencies is unchanged.
1364 In order to produce complex equalisation curves, this filter can
1365 be given several times, each with a different central frequency.
1367 The filter accepts the following options:
1371 Set the filter's central frequency in Hz.
1374 Set method to specify band-width of filter.
1387 Specify the band-width of a filter in width_type units.
1390 Set the required gain or attenuation in dB.
1391 Beware of clipping when using a positive gain.
1394 @subsection Examples
1397 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
1399 equalizer=f=1000:width_type=h:width=200:g=-10
1403 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
1405 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
1411 Apply a high-pass filter with 3dB point frequency.
1412 The filter can be either single-pole, or double-pole (the default).
1413 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1415 The filter accepts the following options:
1419 Set frequency in Hz. Default is 3000.
1422 Set number of poles. Default is 2.
1425 Set method to specify band-width of filter.
1438 Specify the band-width of a filter in width_type units.
1439 Applies only to double-pole filter.
1440 The default is 0.707q and gives a Butterworth response.
1445 Join multiple input streams into one multi-channel stream.
1447 The filter accepts the following named parameters:
1451 Number of input streams. Defaults to 2.
1453 @item channel_layout
1454 Desired output channel layout. Defaults to stereo.
1457 Map channels from inputs to output. The argument is a '|'-separated list of
1458 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
1459 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
1460 can be either the name of the input channel (e.g. FL for front left) or its
1461 index in the specified input stream. @var{out_channel} is the name of the output
1465 The filter will attempt to guess the mappings when those are not specified
1466 explicitly. It does so by first trying to find an unused matching input channel
1467 and if that fails it picks the first unused input channel.
1469 E.g. to join 3 inputs (with properly set channel layouts)
1471 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
1474 To build a 5.1 output from 6 single-channel streams:
1476 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
1477 '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'
1483 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
1485 To enable compilation of this filter you need to configure FFmpeg with
1486 @code{--enable-ladspa}.
1490 Specifies the name of LADSPA plugin library to load. If the environment
1491 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
1492 each one of the directories specified by the colon separated list in
1493 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
1494 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
1495 @file{/usr/lib/ladspa/}.
1498 Specifies the plugin within the library. Some libraries contain only
1499 one plugin, but others contain many of them. If this is not set filter
1500 will list all available plugins within the specified library.
1503 Set the '|' separated list of controls which are zero or more floating point
1504 values that determine the behavior of the loaded plugin (for example delay,
1506 Controls need to be defined using the following syntax:
1507 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
1508 @var{valuei} is the value set on the @var{i}-th control.
1509 If @option{controls} is set to @code{help}, all available controls and
1510 their valid ranges are printed.
1512 @item sample_rate, s
1513 Specify the sample rate, default to 44100. Only used if plugin have
1517 Set the number of samples per channel per each output frame, default
1518 is 1024. Only used if plugin have zero inputs.
1521 Set the minimum duration of the sourced audio. See the function
1522 @code{av_parse_time()} for the accepted format, also check the "Time duration"
1523 section in the ffmpeg-utils manual.
1524 Note that the resulting duration may be greater than the specified duration,
1525 as the generated audio is always cut at the end of a complete frame.
1526 If not specified, or the expressed duration is negative, the audio is
1527 supposed to be generated forever.
1528 Only used if plugin have zero inputs.
1532 @subsection Examples
1536 List all available plugins within amp (LADSPA example plugin) library:
1542 List all available controls and their valid ranges for @code{vcf_notch}
1543 plugin from @code{VCF} library:
1545 ladspa=f=vcf:p=vcf_notch:c=help
1549 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
1552 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
1556 Add reverberation to the audio using TAP-plugins
1557 (Tom's Audio Processing plugins):
1559 ladspa=file=tap_reverb:tap_reverb
1563 Generate white noise, with 0.2 amplitude:
1565 ladspa=file=cmt:noise_source_white:c=c0=.2
1569 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
1570 @code{C* Audio Plugin Suite} (CAPS) library:
1572 ladspa=file=caps:Click:c=c1=20'
1576 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
1578 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
1582 @subsection Commands
1584 This filter supports the following commands:
1587 Modify the @var{N}-th control value.
1589 If the specified value is not valid, it is ignored and prior one is kept.
1594 Apply a low-pass filter with 3dB point frequency.
1595 The filter can be either single-pole or double-pole (the default).
1596 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1598 The filter accepts the following options:
1602 Set frequency in Hz. Default is 500.
1605 Set number of poles. Default is 2.
1608 Set method to specify band-width of filter.
1621 Specify the band-width of a filter in width_type units.
1622 Applies only to double-pole filter.
1623 The default is 0.707q and gives a Butterworth response.
1628 Mix channels with specific gain levels. The filter accepts the output
1629 channel layout followed by a set of channels definitions.
1631 This filter is also designed to remap efficiently the channels of an audio
1634 The filter accepts parameters of the form:
1635 "@var{l}:@var{outdef}:@var{outdef}:..."
1639 output channel layout or number of channels
1642 output channel specification, of the form:
1643 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
1646 output channel to define, either a channel name (FL, FR, etc.) or a channel
1647 number (c0, c1, etc.)
1650 multiplicative coefficient for the channel, 1 leaving the volume unchanged
1653 input channel to use, see out_name for details; it is not possible to mix
1654 named and numbered input channels
1657 If the `=' in a channel specification is replaced by `<', then the gains for
1658 that specification will be renormalized so that the total is 1, thus
1659 avoiding clipping noise.
1661 @subsection Mixing examples
1663 For example, if you want to down-mix from stereo to mono, but with a bigger
1664 factor for the left channel:
1666 pan=1:c0=0.9*c0+0.1*c1
1669 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
1670 7-channels surround:
1672 pan=stereo: FL < FL + 0.5*FC + 0.6*BL + 0.6*SL : FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
1675 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
1676 that should be preferred (see "-ac" option) unless you have very specific
1679 @subsection Remapping examples
1681 The channel remapping will be effective if, and only if:
1684 @item gain coefficients are zeroes or ones,
1685 @item only one input per channel output,
1688 If all these conditions are satisfied, the filter will notify the user ("Pure
1689 channel mapping detected"), and use an optimized and lossless method to do the
1692 For example, if you have a 5.1 source and want a stereo audio stream by
1693 dropping the extra channels:
1695 pan="stereo: c0=FL : c1=FR"
1698 Given the same source, you can also switch front left and front right channels
1699 and keep the input channel layout:
1701 pan="5.1: c0=c1 : c1=c0 : c2=c2 : c3=c3 : c4=c4 : c5=c5"
1704 If the input is a stereo audio stream, you can mute the front left channel (and
1705 still keep the stereo channel layout) with:
1710 Still with a stereo audio stream input, you can copy the right channel in both
1711 front left and right:
1713 pan="stereo: c0=FR : c1=FR"
1718 ReplayGain scanner filter. This filter takes an audio stream as an input and
1719 outputs it unchanged.
1720 At end of filtering it displays @code{track_gain} and @code{track_peak}.
1724 Convert the audio sample format, sample rate and channel layout. This filter is
1725 not meant to be used directly.
1727 @section silencedetect
1729 Detect silence in an audio stream.
1731 This filter logs a message when it detects that the input audio volume is less
1732 or equal to a noise tolerance value for a duration greater or equal to the
1733 minimum detected noise duration.
1735 The printed times and duration are expressed in seconds.
1737 The filter accepts the following options:
1741 Set silence duration until notification (default is 2 seconds).
1744 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
1745 specified value) or amplitude ratio. Default is -60dB, or 0.001.
1748 @subsection Examples
1752 Detect 5 seconds of silence with -50dB noise tolerance:
1754 silencedetect=n=-50dB:d=5
1758 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
1759 tolerance in @file{silence.mp3}:
1761 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
1767 Boost or cut treble (upper) frequencies of the audio using a two-pole
1768 shelving filter with a response similar to that of a standard
1769 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1771 The filter accepts the following options:
1775 Give the gain at whichever is the lower of ~22 kHz and the
1776 Nyquist frequency. Its useful range is about -20 (for a large cut)
1777 to +20 (for a large boost). Beware of clipping when using a positive gain.
1780 Set the filter's central frequency and so can be used
1781 to extend or reduce the frequency range to be boosted or cut.
1782 The default value is @code{3000} Hz.
1785 Set method to specify band-width of filter.
1798 Determine how steep is the filter's shelf transition.
1803 Adjust the input audio volume.
1805 The filter accepts the following options:
1810 Set audio volume expression.
1812 Output values are clipped to the maximum value.
1814 The output audio volume is given by the relation:
1816 @var{output_volume} = @var{volume} * @var{input_volume}
1819 Default value for @var{volume} is "1.0".
1822 Set the mathematical precision.
1824 This determines which input sample formats will be allowed, which affects the
1825 precision of the volume scaling.
1829 8-bit fixed-point; limits input sample format to U8, S16, and S32.
1831 32-bit floating-point; limits input sample format to FLT. (default)
1833 64-bit floating-point; limits input sample format to DBL.
1837 Set when the volume expression is evaluated.
1839 It accepts the following values:
1842 only evaluate expression once during the filter initialization, or
1843 when the @samp{volume} command is sent
1846 evaluate expression for each incoming frame
1849 Default value is @samp{once}.
1852 The volume expression can contain the following parameters.
1856 frame number (starting at zero)
1859 @item nb_consumed_samples
1860 number of samples consumed by the filter
1862 number of samples in the current frame
1864 original frame position in the file
1870 PTS at start of stream
1872 time at start of stream
1878 last set volume value
1881 Note that when @option{eval} is set to @samp{once} only the
1882 @var{sample_rate} and @var{tb} variables are available, all other
1883 variables will evaluate to NAN.
1885 @subsection Commands
1887 This filter supports the following commands:
1890 Modify the volume expression.
1891 The command accepts the same syntax of the corresponding option.
1893 If the specified expression is not valid, it is kept at its current
1897 @subsection Examples
1901 Halve the input audio volume:
1905 volume=volume=-6.0206dB
1908 In all the above example the named key for @option{volume} can be
1909 omitted, for example like in:
1915 Increase input audio power by 6 decibels using fixed-point precision:
1917 volume=volume=6dB:precision=fixed
1921 Fade volume after time 10 with an annihilation period of 5 seconds:
1923 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
1927 @section volumedetect
1929 Detect the volume of the input video.
1931 The filter has no parameters. The input is not modified. Statistics about
1932 the volume will be printed in the log when the input stream end is reached.
1934 In particular it will show the mean volume (root mean square), maximum
1935 volume (on a per-sample basis), and the beginning of a histogram of the
1936 registered volume values (from the maximum value to a cumulated 1/1000 of
1939 All volumes are in decibels relative to the maximum PCM value.
1941 @subsection Examples
1943 Here is an excerpt of the output:
1945 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
1946 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
1947 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
1948 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
1949 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
1950 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
1951 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
1952 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
1953 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
1959 The mean square energy is approximately -27 dB, or 10^-2.7.
1961 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
1963 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
1966 In other words, raising the volume by +4 dB does not cause any clipping,
1967 raising it by +5 dB causes clipping for 6 samples, etc.
1969 @c man end AUDIO FILTERS
1971 @chapter Audio Sources
1972 @c man begin AUDIO SOURCES
1974 Below is a description of the currently available audio sources.
1978 Buffer audio frames, and make them available to the filter chain.
1980 This source is mainly intended for a programmatic use, in particular
1981 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
1983 It accepts the following named parameters:
1988 Timebase which will be used for timestamps of submitted frames. It must be
1989 either a floating-point number or in @var{numerator}/@var{denominator} form.
1992 The sample rate of the incoming audio buffers.
1995 The sample format of the incoming audio buffers.
1996 Either a sample format name or its corresponging integer representation from
1997 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
1999 @item channel_layout
2000 The channel layout of the incoming audio buffers.
2001 Either a channel layout name from channel_layout_map in
2002 @file{libavutil/channel_layout.c} or its corresponding integer representation
2003 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
2006 The number of channels of the incoming audio buffers.
2007 If both @var{channels} and @var{channel_layout} are specified, then they
2012 @subsection Examples
2015 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
2018 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
2019 Since the sample format with name "s16p" corresponds to the number
2020 6 and the "stereo" channel layout corresponds to the value 0x3, this is
2023 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
2028 Generate an audio signal specified by an expression.
2030 This source accepts in input one or more expressions (one for each
2031 channel), which are evaluated and used to generate a corresponding
2034 This source accepts the following options:
2038 Set the '|'-separated expressions list for each separate channel. In case the
2039 @option{channel_layout} option is not specified, the selected channel layout
2040 depends on the number of provided expressions. Otherwise the last
2041 specified expression is applied to the remaining output channels.
2043 @item channel_layout, c
2044 Set the channel layout. The number of channels in the specified layout
2045 must be equal to the number of specified expressions.
2048 Set the minimum duration of the sourced audio. See the function
2049 @code{av_parse_time()} for the accepted format.
2050 Note that the resulting duration may be greater than the specified
2051 duration, as the generated audio is always cut at the end of a
2054 If not specified, or the expressed duration is negative, the audio is
2055 supposed to be generated forever.
2058 Set the number of samples per channel per each output frame,
2061 @item sample_rate, s
2062 Specify the sample rate, default to 44100.
2065 Each expression in @var{exprs} can contain the following constants:
2069 number of the evaluated sample, starting from 0
2072 time of the evaluated sample expressed in seconds, starting from 0
2079 @subsection Examples
2089 Generate a sin signal with frequency of 440 Hz, set sample rate to
2092 aevalsrc="sin(440*2*PI*t):s=8000"
2096 Generate a two channels signal, specify the channel layout (Front
2097 Center + Back Center) explicitly:
2099 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
2103 Generate white noise:
2105 aevalsrc="-2+random(0)"
2109 Generate an amplitude modulated signal:
2111 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
2115 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
2117 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
2124 Null audio source, return unprocessed audio frames. It is mainly useful
2125 as a template and to be employed in analysis / debugging tools, or as
2126 the source for filters which ignore the input data (for example the sox
2129 This source accepts the following options:
2133 @item channel_layout, cl
2135 Specify the channel layout, and can be either an integer or a string
2136 representing a channel layout. The default value of @var{channel_layout}
2139 Check the channel_layout_map definition in
2140 @file{libavutil/channel_layout.c} for the mapping between strings and
2141 channel layout values.
2143 @item sample_rate, r
2144 Specify the sample rate, and defaults to 44100.
2147 Set the number of samples per requested frames.
2151 @subsection Examples
2155 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
2157 anullsrc=r=48000:cl=4
2161 Do the same operation with a more obvious syntax:
2163 anullsrc=r=48000:cl=mono
2167 All the parameters need to be explicitly defined.
2171 Synthesize a voice utterance using the libflite library.
2173 To enable compilation of this filter you need to configure FFmpeg with
2174 @code{--enable-libflite}.
2176 Note that the flite library is not thread-safe.
2178 The filter accepts the following options:
2183 If set to 1, list the names of the available voices and exit
2184 immediately. Default value is 0.
2187 Set the maximum number of samples per frame. Default value is 512.
2190 Set the filename containing the text to speak.
2193 Set the text to speak.
2196 Set the voice to use for the speech synthesis. Default value is
2197 @code{kal}. See also the @var{list_voices} option.
2200 @subsection Examples
2204 Read from file @file{speech.txt}, and synthetize the text using the
2205 standard flite voice:
2207 flite=textfile=speech.txt
2211 Read the specified text selecting the @code{slt} voice:
2213 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2217 Input text to ffmpeg:
2219 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2223 Make @file{ffplay} speak the specified text, using @code{flite} and
2224 the @code{lavfi} device:
2226 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
2230 For more information about libflite, check:
2231 @url{http://www.speech.cs.cmu.edu/flite/}
2235 Generate an audio signal made of a sine wave with amplitude 1/8.
2237 The audio signal is bit-exact.
2239 The filter accepts the following options:
2244 Set the carrier frequency. Default is 440 Hz.
2246 @item beep_factor, b
2247 Enable a periodic beep every second with frequency @var{beep_factor} times
2248 the carrier frequency. Default is 0, meaning the beep is disabled.
2250 @item sample_rate, r
2251 Specify the sample rate, default is 44100.
2254 Specify the duration of the generated audio stream.
2256 @item samples_per_frame
2257 Set the number of samples per output frame, default is 1024.
2260 @subsection Examples
2265 Generate a simple 440 Hz sine wave:
2271 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
2275 sine=frequency=220:beep_factor=4:duration=5
2280 @c man end AUDIO SOURCES
2282 @chapter Audio Sinks
2283 @c man begin AUDIO SINKS
2285 Below is a description of the currently available audio sinks.
2287 @section abuffersink
2289 Buffer audio frames, and make them available to the end of filter chain.
2291 This sink is mainly intended for programmatic use, in particular
2292 through the interface defined in @file{libavfilter/buffersink.h}
2293 or the options system.
2295 It accepts a pointer to an AVABufferSinkContext structure, which
2296 defines the incoming buffers' formats, to be passed as the opaque
2297 parameter to @code{avfilter_init_filter} for initialization.
2301 Null audio sink, do absolutely nothing with the input audio. It is
2302 mainly useful as a template and to be employed in analysis / debugging
2305 @c man end AUDIO SINKS
2307 @chapter Video Filters
2308 @c man begin VIDEO FILTERS
2310 When you configure your FFmpeg build, you can disable any of the
2311 existing filters using @code{--disable-filters}.
2312 The configure output will show the video filters included in your
2315 Below is a description of the currently available video filters.
2317 @section alphaextract
2319 Extract the alpha component from the input as a grayscale video. This
2320 is especially useful with the @var{alphamerge} filter.
2324 Add or replace the alpha component of the primary input with the
2325 grayscale value of a second input. This is intended for use with
2326 @var{alphaextract} to allow the transmission or storage of frame
2327 sequences that have alpha in a format that doesn't support an alpha
2330 For example, to reconstruct full frames from a normal YUV-encoded video
2331 and a separate video created with @var{alphaextract}, you might use:
2333 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
2336 Since this filter is designed for reconstruction, it operates on frame
2337 sequences without considering timestamps, and terminates when either
2338 input reaches end of stream. This will cause problems if your encoding
2339 pipeline drops frames. If you're trying to apply an image as an
2340 overlay to a video stream, consider the @var{overlay} filter instead.
2344 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
2345 and libavformat to work. On the other hand, it is limited to ASS (Advanced
2346 Substation Alpha) subtitles files.
2350 Compute the bounding box for the non-black pixels in the input frame
2353 This filter computes the bounding box containing all the pixels with a
2354 luminance value greater than the minimum allowed value.
2355 The parameters describing the bounding box are printed on the filter
2358 The filter accepts the following option:
2362 Set the minimal luminance value. Default is @code{16}.
2365 @section blackdetect
2367 Detect video intervals that are (almost) completely black. Can be
2368 useful to detect chapter transitions, commercials, or invalid
2369 recordings. Output lines contains the time for the start, end and
2370 duration of the detected black interval expressed in seconds.
2372 In order to display the output lines, you need to set the loglevel at
2373 least to the AV_LOG_INFO value.
2375 The filter accepts the following options:
2378 @item black_min_duration, d
2379 Set the minimum detected black duration expressed in seconds. It must
2380 be a non-negative floating point number.
2382 Default value is 2.0.
2384 @item picture_black_ratio_th, pic_th
2385 Set the threshold for considering a picture "black".
2386 Express the minimum value for the ratio:
2388 @var{nb_black_pixels} / @var{nb_pixels}
2391 for which a picture is considered black.
2392 Default value is 0.98.
2394 @item pixel_black_th, pix_th
2395 Set the threshold for considering a pixel "black".
2397 The threshold expresses the maximum pixel luminance value for which a
2398 pixel is considered "black". The provided value is scaled according to
2399 the following equation:
2401 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
2404 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
2405 the input video format, the range is [0-255] for YUV full-range
2406 formats and [16-235] for YUV non full-range formats.
2408 Default value is 0.10.
2411 The following example sets the maximum pixel threshold to the minimum
2412 value, and detects only black intervals of 2 or more seconds:
2414 blackdetect=d=2:pix_th=0.00
2419 Detect frames that are (almost) completely black. Can be useful to
2420 detect chapter transitions or commercials. Output lines consist of
2421 the frame number of the detected frame, the percentage of blackness,
2422 the position in the file if known or -1 and the timestamp in seconds.
2424 In order to display the output lines, you need to set the loglevel at
2425 least to the AV_LOG_INFO value.
2427 The filter accepts the following options:
2432 Set the percentage of the pixels that have to be below the threshold, defaults
2435 @item threshold, thresh
2436 Set the threshold below which a pixel value is considered black, defaults to
2443 Blend two video frames into each other.
2445 It takes two input streams and outputs one stream, the first input is the
2446 "top" layer and second input is "bottom" layer.
2447 Output terminates when shortest input terminates.
2449 A description of the accepted options follows.
2457 Set blend mode for specific pixel component or all pixel components in case
2458 of @var{all_mode}. Default value is @code{normal}.
2460 Available values for component modes are:
2493 Set blend opacity for specific pixel component or all pixel components in case
2494 of @var{all_opacity}. Only used in combination with pixel component blend modes.
2501 Set blend expression for specific pixel component or all pixel components in case
2502 of @var{all_expr}. Note that related mode options will be ignored if those are set.
2504 The expressions can use the following variables:
2508 The sequential number of the filtered frame, starting from @code{0}.
2512 the coordinates of the current sample
2516 the width and height of currently filtered plane
2520 Width and height scale depending on the currently filtered plane. It is the
2521 ratio between the corresponding luma plane number of pixels and the current
2522 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
2523 @code{0.5,0.5} for chroma planes.
2526 Time of the current frame, expressed in seconds.
2529 Value of pixel component at current location for first video frame (top layer).
2532 Value of pixel component at current location for second video frame (bottom layer).
2536 Force termination when the shortest input terminates. Default is @code{0}.
2538 Continue applying the last bottom frame after the end of the stream. A value of
2539 @code{0} disable the filter after the last frame of the bottom layer is reached.
2540 Default is @code{1}.
2543 @subsection Examples
2547 Apply transition from bottom layer to top layer in first 10 seconds:
2549 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
2553 Apply 1x1 checkerboard effect:
2555 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
2559 Apply uncover left effect:
2561 blend=all_expr='if(gte(N*SW+X,W),A,B)'
2565 Apply uncover down effect:
2567 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
2571 Apply uncover up-left effect:
2573 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
2579 Apply boxblur algorithm to the input video.
2581 The filter accepts the following options:
2585 @item luma_radius, lr
2586 @item luma_power, lp
2587 @item chroma_radius, cr
2588 @item chroma_power, cp
2589 @item alpha_radius, ar
2590 @item alpha_power, ap
2594 A description of the accepted options follows.
2597 @item luma_radius, lr
2598 @item chroma_radius, cr
2599 @item alpha_radius, ar
2600 Set an expression for the box radius in pixels used for blurring the
2601 corresponding input plane.
2603 The radius value must be a non-negative number, and must not be
2604 greater than the value of the expression @code{min(w,h)/2} for the
2605 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
2608 Default value for @option{luma_radius} is "2". If not specified,
2609 @option{chroma_radius} and @option{alpha_radius} default to the
2610 corresponding value set for @option{luma_radius}.
2612 The expressions can contain the following constants:
2616 the input width and height in pixels
2620 the input chroma image width and height in pixels
2624 horizontal and vertical chroma subsample values. For example for the
2625 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2628 @item luma_power, lp
2629 @item chroma_power, cp
2630 @item alpha_power, ap
2631 Specify how many times the boxblur filter is applied to the
2632 corresponding plane.
2634 Default value for @option{luma_power} is 2. If not specified,
2635 @option{chroma_power} and @option{alpha_power} default to the
2636 corresponding value set for @option{luma_power}.
2638 A value of 0 will disable the effect.
2641 @subsection Examples
2645 Apply a boxblur filter with luma, chroma, and alpha radius
2648 boxblur=luma_radius=2:luma_power=1
2653 Set luma radius to 2, alpha and chroma radius to 0:
2655 boxblur=2:1:cr=0:ar=0
2659 Set luma and chroma radius to a fraction of the video dimension:
2661 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
2665 @section colorbalance
2666 Modify intensity of primary colors (red, green and blue) of input frames.
2668 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
2669 regions for the red-cyan, green-magenta or blue-yellow balance.
2671 A positive adjustment value shifts the balance towards the primary color, a negative
2672 value towards the complementary color.
2674 The filter accepts the following options:
2680 Adjust red, green and blue shadows (darkest pixels).
2685 Adjust red, green and blue midtones (medium pixels).
2690 Adjust red, green and blue highlights (brightest pixels).
2692 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
2695 @subsection Examples
2699 Add red color cast to shadows:
2705 @section colorchannelmixer
2707 Adjust video input frames by re-mixing color channels.
2709 This filter modifies a color channel by adding the values associated to
2710 the other channels of the same pixels. For example if the value to
2711 modify is red, the output value will be:
2713 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
2716 The filter accepts the following options:
2723 Adjust contribution of input red, green, blue and alpha channels for output red channel.
2724 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
2730 Adjust contribution of input red, green, blue and alpha channels for output green channel.
2731 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
2737 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
2738 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
2744 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
2745 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
2747 Allowed ranges for options are @code{[-2.0, 2.0]}.
2750 @subsection Examples
2754 Convert source to grayscale:
2756 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
2759 Simulate sepia tones:
2761 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
2765 @section colormatrix
2767 Convert color matrix.
2769 The filter accepts the following options:
2774 Specify the source and destination color matrix. Both values must be
2777 The accepted values are:
2793 For example to convert from BT.601 to SMPTE-240M, use the command:
2795 colormatrix=bt601:smpte240m
2800 Copy the input source unchanged to the output. Mainly useful for
2805 Crop the input video to given dimensions.
2807 The filter accepts the following options:
2811 Width of the output video. It defaults to @code{iw}.
2812 This expression is evaluated only once during the filter
2816 Height of the output video. It defaults to @code{ih}.
2817 This expression is evaluated only once during the filter
2821 Horizontal position, in the input video, of the left edge of the output video.
2822 It defaults to @code{(in_w-out_w)/2}.
2823 This expression is evaluated per-frame.
2826 Vertical position, in the input video, of the top edge of the output video.
2827 It defaults to @code{(in_h-out_h)/2}.
2828 This expression is evaluated per-frame.
2831 If set to 1 will force the output display aspect ratio
2832 to be the same of the input, by changing the output sample aspect
2833 ratio. It defaults to 0.
2836 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
2837 expressions containing the following constants:
2842 the computed values for @var{x} and @var{y}. They are evaluated for
2847 the input width and height
2851 same as @var{in_w} and @var{in_h}
2855 the output (cropped) width and height
2859 same as @var{out_w} and @var{out_h}
2862 same as @var{iw} / @var{ih}
2865 input sample aspect ratio
2868 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
2872 horizontal and vertical chroma subsample values. For example for the
2873 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2876 the number of input frame, starting from 0
2879 the position in the file of the input frame, NAN if unknown
2882 timestamp expressed in seconds, NAN if the input timestamp is unknown
2886 The expression for @var{out_w} may depend on the value of @var{out_h},
2887 and the expression for @var{out_h} may depend on @var{out_w}, but they
2888 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
2889 evaluated after @var{out_w} and @var{out_h}.
2891 The @var{x} and @var{y} parameters specify the expressions for the
2892 position of the top-left corner of the output (non-cropped) area. They
2893 are evaluated for each frame. If the evaluated value is not valid, it
2894 is approximated to the nearest valid value.
2896 The expression for @var{x} may depend on @var{y}, and the expression
2897 for @var{y} may depend on @var{x}.
2899 @subsection Examples
2903 Crop area with size 100x100 at position (12,34).
2908 Using named options, the example above becomes:
2910 crop=w=100:h=100:x=12:y=34
2914 Crop the central input area with size 100x100:
2920 Crop the central input area with size 2/3 of the input video:
2922 crop=2/3*in_w:2/3*in_h
2926 Crop the input video central square:
2933 Delimit the rectangle with the top-left corner placed at position
2934 100:100 and the right-bottom corner corresponding to the right-bottom
2935 corner of the input image:
2937 crop=in_w-100:in_h-100:100:100
2941 Crop 10 pixels from the left and right borders, and 20 pixels from
2942 the top and bottom borders
2944 crop=in_w-2*10:in_h-2*20
2948 Keep only the bottom right quarter of the input image:
2950 crop=in_w/2:in_h/2:in_w/2:in_h/2
2954 Crop height for getting Greek harmony:
2956 crop=in_w:1/PHI*in_w
2960 Appply trembling effect:
2962 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)
2966 Apply erratic camera effect depending on timestamp:
2968 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)"
2972 Set x depending on the value of y:
2974 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
2980 Auto-detect crop size.
2982 Calculate necessary cropping parameters and prints the recommended
2983 parameters through the logging system. The detected dimensions
2984 correspond to the non-black area of the input video.
2986 The filter accepts the following options:
2991 Set higher black value threshold, which can be optionally specified
2992 from nothing (0) to everything (255). An intensity value greater
2993 to the set value is considered non-black. Default value is 24.
2996 Set the value for which the width/height should be divisible by. The
2997 offset is automatically adjusted to center the video. Use 2 to get
2998 only even dimensions (needed for 4:2:2 video). 16 is best when
2999 encoding to most video codecs. Default value is 16.
3001 @item reset_count, reset
3002 Set the counter that determines after how many frames cropdetect will
3003 reset the previously detected largest video area and start over to
3004 detect the current optimal crop area. Default value is 0.
3006 This can be useful when channel logos distort the video area. 0
3007 indicates never reset and return the largest area encountered during
3014 Apply color adjustments using curves.
3016 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
3017 component (red, green and blue) has its values defined by @var{N} key points
3018 tied from each other using a smooth curve. The x-axis represents the pixel
3019 values from the input frame, and the y-axis the new pixel values to be set for
3022 By default, a component curve is defined by the two points @var{(0;0)} and
3023 @var{(1;1)}. This creates a straight line where each original pixel value is
3024 "adjusted" to its own value, which means no change to the image.
3026 The filter allows you to redefine these two points and add some more. A new
3027 curve (using a natural cubic spline interpolation) will be define to pass
3028 smoothly through all these new coordinates. The new defined points needs to be
3029 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
3030 be in the @var{[0;1]} interval. If the computed curves happened to go outside
3031 the vector spaces, the values will be clipped accordingly.
3033 If there is no key point defined in @code{x=0}, the filter will automatically
3034 insert a @var{(0;0)} point. In the same way, if there is no key point defined
3035 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
3037 The filter accepts the following options:
3041 Select one of the available color presets. This option can be used in addition
3042 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
3043 options takes priority on the preset values.
3044 Available presets are:
3047 @item color_negative
3050 @item increase_contrast
3052 @item linear_contrast
3053 @item medium_contrast
3055 @item strong_contrast
3058 Default is @code{none}.
3060 Set the master key points. These points will define a second pass mapping. It
3061 is sometimes called a "luminance" or "value" mapping. It can be used with
3062 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
3063 post-processing LUT.
3065 Set the key points for the red component.
3067 Set the key points for the green component.
3069 Set the key points for the blue component.
3071 Set the key points for all components (not including master).
3072 Can be used in addition to the other key points component
3073 options. In this case, the unset component(s) will fallback on this
3074 @option{all} setting.
3076 Specify a Photoshop curves file (@code{.asv}) to import the settings from.
3079 To avoid some filtergraph syntax conflicts, each key points list need to be
3080 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
3082 @subsection Examples
3086 Increase slightly the middle level of blue:
3088 curves=blue='0.5/0.58'
3094 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
3096 Here we obtain the following coordinates for each components:
3099 @code{(0;0.11) (0.42;0.51) (1;0.95)}
3101 @code{(0;0) (0.50;0.48) (1;1)}
3103 @code{(0;0.22) (0.49;0.44) (1;0.80)}
3107 The previous example can also be achieved with the associated built-in preset:
3109 curves=preset=vintage
3119 Use a Photoshop preset and redefine the points of the green component:
3121 curves=psfile='MyCurvesPresets/purple.asv':green='0.45/0.53'
3127 Denoise frames using 2D DCT (frequency domain filtering).
3129 This filter is not designed for real time and can be extremely slow.
3131 The filter accepts the following options:
3135 Set the noise sigma constant.
3137 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
3138 coefficient (absolute value) below this threshold with be dropped.
3140 If you need a more advanced filtering, see @option{expr}.
3142 Default is @code{0}.
3145 Set number overlapping pixels for each block. Each block is of size
3146 @code{16x16}. Since the filter can be slow, you may want to reduce this value,
3147 at the cost of a less effective filter and the risk of various artefacts.
3149 If the overlapping value doesn't allow to process the whole input width or
3150 height, a warning will be displayed and according borders won't be denoised.
3152 Default value is @code{15}.
3155 Set the coefficient factor expression.
3157 For each coefficient of a DCT block, this expression will be evaluated as a
3158 multiplier value for the coefficient.
3160 If this is option is set, the @option{sigma} option will be ignored.
3162 The absolute value of the coefficient can be accessed through the @var{c}
3166 @subsection Examples
3168 Apply a denoise with a @option{sigma} of @code{4.5}:
3173 The same operation can be achieved using the expression system:
3175 dctdnoiz=e='gte(c, 4.5*3)'
3181 Drop duplicated frames at regular intervals.
3183 The filter accepts the following options:
3187 Set the number of frames from which one will be dropped. Setting this to
3188 @var{N} means one frame in every batch of @var{N} frames will be dropped.
3189 Default is @code{5}.
3192 Set the threshold for duplicate detection. If the difference metric for a frame
3193 is less than or equal to this value, then it is declared as duplicate. Default
3197 Set scene change threshold. Default is @code{15}.
3201 Set the size of the x and y-axis blocks used during metric calculations.
3202 Larger blocks give better noise suppression, but also give worse detection of
3203 small movements. Must be a power of two. Default is @code{32}.
3206 Mark main input as a pre-processed input and activate clean source input
3207 stream. This allows the input to be pre-processed with various filters to help
3208 the metrics calculation while keeping the frame selection lossless. When set to
3209 @code{1}, the first stream is for the pre-processed input, and the second
3210 stream is the clean source from where the kept frames are chosen. Default is
3214 Set whether or not chroma is considered in the metric calculations. Default is
3220 Remove judder produced by partially interlaced telecined content.
3222 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
3223 source was partially telecined content then the output of @code{pullup,dejudder}
3224 will have a variable frame rate. May change the recorded frame rate of the
3225 container. Aside from that change, this filter will not affect constant frame
3228 The option available in this filter is:
3232 Specify the length of the window over which the judder repeats.
3234 Accepts any interger greater than 1. Useful values are:
3238 If the original was telecined from 24 to 30 fps (Film to NTSC).
3241 If the original was telecined from 25 to 30 fps (PAL to NTSC).
3244 If a mixture of the two.
3247 The default is @samp{4}.
3252 Suppress a TV station logo by a simple interpolation of the surrounding
3253 pixels. Just set a rectangle covering the logo and watch it disappear
3254 (and sometimes something even uglier appear - your mileage may vary).
3256 This filter accepts the following options:
3261 Specify the top left corner coordinates of the logo. They must be
3266 Specify the width and height of the logo to clear. They must be
3270 Specify the thickness of the fuzzy edge of the rectangle (added to
3271 @var{w} and @var{h}). The default value is 4.
3274 When set to 1, a green rectangle is drawn on the screen to simplify
3275 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
3276 The default value is 0.
3278 The rectangle is drawn on the outermost pixels which will be (partly)
3279 replaced with interpolated values. The values of the next pixels
3280 immediately outside this rectangle in each direction will be used to
3281 compute the interpolated pixel values inside the rectangle.
3285 @subsection Examples
3289 Set a rectangle covering the area with top left corner coordinates 0,0
3290 and size 100x77, setting a band of size 10:
3292 delogo=x=0:y=0:w=100:h=77:band=10
3299 Attempt to fix small changes in horizontal and/or vertical shift. This
3300 filter helps remove camera shake from hand-holding a camera, bumping a
3301 tripod, moving on a vehicle, etc.
3303 The filter accepts the following options:
3311 Specify a rectangular area where to limit the search for motion
3313 If desired the search for motion vectors can be limited to a
3314 rectangular area of the frame defined by its top left corner, width
3315 and height. These parameters have the same meaning as the drawbox
3316 filter which can be used to visualise the position of the bounding
3319 This is useful when simultaneous movement of subjects within the frame
3320 might be confused for camera motion by the motion vector search.
3322 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
3323 then the full frame is used. This allows later options to be set
3324 without specifying the bounding box for the motion vector search.
3326 Default - search the whole frame.
3330 Specify the maximum extent of movement in x and y directions in the
3331 range 0-64 pixels. Default 16.
3334 Specify how to generate pixels to fill blanks at the edge of the
3335 frame. Available values are:
3338 Fill zeroes at blank locations
3340 Original image at blank locations
3342 Extruded edge value at blank locations
3344 Mirrored edge at blank locations
3346 Default value is @samp{mirror}.
3349 Specify the blocksize to use for motion search. Range 4-128 pixels,
3353 Specify the contrast threshold for blocks. Only blocks with more than
3354 the specified contrast (difference between darkest and lightest
3355 pixels) will be considered. Range 1-255, default 125.
3358 Specify the search strategy. Available values are:
3361 Set exhaustive search
3363 Set less exhaustive search.
3365 Default value is @samp{exhaustive}.
3368 If set then a detailed log of the motion search is written to the
3372 If set to 1, specify using OpenCL capabilities, only available if
3373 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
3379 Draw a colored box on the input image.
3381 This filter accepts the following options:
3386 The expressions which specify the top left corner coordinates of the box. Default to 0.
3390 The expressions which specify the width and height of the box, if 0 they are interpreted as
3391 the input width and height. Default to 0.
3394 Specify the color of the box to write. For the general syntax of this option,
3395 check the "Color" section in the ffmpeg-utils manual. If the special
3396 value @code{invert} is used, the box edge color is the same as the
3397 video with inverted luma.
3400 The expression which sets the thickness of the box edge. Default value is @code{3}.
3402 See below for the list of accepted constants.
3405 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3406 following constants:
3410 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3414 horizontal and vertical chroma subsample values. For example for the
3415 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3419 The input width and height.
3422 The input sample aspect ratio.
3426 The x and y offset coordinates where the box is drawn.
3430 The width and height of the drawn box.
3433 The thickness of the drawn box.
3435 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3436 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3440 @subsection Examples
3444 Draw a black box around the edge of the input image:
3450 Draw a box with color red and an opacity of 50%:
3452 drawbox=10:20:200:60:red@@0.5
3455 The previous example can be specified as:
3457 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
3461 Fill the box with pink color:
3463 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
3467 Draw a 2-pixel red 2.40:1 mask:
3469 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
3475 Draw a grid on the input image.
3477 This filter accepts the following options:
3482 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
3486 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
3487 input width and height, respectively, minus @code{thickness}, so image gets
3488 framed. Default to 0.
3491 Specify the color of the grid. For the general syntax of this option,
3492 check the "Color" section in the ffmpeg-utils manual. If the special
3493 value @code{invert} is used, the grid color is the same as the
3494 video with inverted luma.
3497 The expression which sets the thickness of the grid line. Default value is @code{1}.
3499 See below for the list of accepted constants.
3502 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3503 following constants:
3507 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3511 horizontal and vertical chroma subsample values. For example for the
3512 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3516 The input grid cell width and height.
3519 The input sample aspect ratio.
3523 The x and y coordinates of some point of grid intersection (meant to configure offset).
3527 The width and height of the drawn cell.
3530 The thickness of the drawn cell.
3532 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3533 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3537 @subsection Examples
3541 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
3543 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
3547 Draw a white 3x3 grid with an opacity of 50%:
3549 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
3556 Draw text string or text from specified file on top of video using the
3557 libfreetype library.
3559 To enable compilation of this filter you need to configure FFmpeg with
3560 @code{--enable-libfreetype}.
3564 The description of the accepted parameters follows.
3569 Used to draw a box around text using background color.
3570 Value should be either 1 (enable) or 0 (disable).
3571 The default value of @var{box} is 0.
3574 The color to be used for drawing box around text. For the syntax of this
3575 option, check the "Color" section in the ffmpeg-utils manual.
3577 The default value of @var{boxcolor} is "white".
3580 Set the width of the border to be drawn around the text using @var{bordercolor}.
3581 The default value of @var{borderw} is 0.
3584 Set the color to be used for drawing border around text. For the syntax of this
3585 option, check the "Color" section in the ffmpeg-utils manual.
3587 The default value of @var{bordercolor} is "black".
3590 Select how the @var{text} is expanded. Can be either @code{none},
3591 @code{strftime} (deprecated) or
3592 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
3596 If true, check and fix text coords to avoid clipping.
3599 The color to be used for drawing fonts. For the syntax of this option, check
3600 the "Color" section in the ffmpeg-utils manual.
3602 The default value of @var{fontcolor} is "black".
3605 The font file to be used for drawing text. Path must be included.
3606 This parameter is mandatory.
3609 The font size to be used for drawing text.
3610 The default value of @var{fontsize} is 16.
3613 Flags to be used for loading the fonts.
3615 The flags map the corresponding flags supported by libfreetype, and are
3616 a combination of the following values:
3623 @item vertical_layout
3624 @item force_autohint
3627 @item ignore_global_advance_width
3629 @item ignore_transform
3635 Default value is "default".
3637 For more information consult the documentation for the FT_LOAD_*
3641 The color to be used for drawing a shadow behind the drawn text. For the
3642 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
3644 The default value of @var{shadowcolor} is "black".
3648 The x and y offsets for the text shadow position with respect to the
3649 position of the text. They can be either positive or negative
3650 values. Default value for both is "0".
3653 The starting frame number for the n/frame_num variable. The default value
3657 The size in number of spaces to use for rendering the tab.
3661 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
3662 format. It can be used with or without text parameter. @var{timecode_rate}
3663 option must be specified.
3665 @item timecode_rate, rate, r
3666 Set the timecode frame rate (timecode only).
3669 The text string to be drawn. The text must be a sequence of UTF-8
3671 This parameter is mandatory if no file is specified with the parameter
3675 A text file containing text to be drawn. The text must be a sequence
3676 of UTF-8 encoded characters.
3678 This parameter is mandatory if no text string is specified with the
3679 parameter @var{text}.
3681 If both @var{text} and @var{textfile} are specified, an error is thrown.
3684 If set to 1, the @var{textfile} will be reloaded before each frame.
3685 Be sure to update it atomically, or it may be read partially, or even fail.
3689 The expressions which specify the offsets where text will be drawn
3690 within the video frame. They are relative to the top/left border of the
3693 The default value of @var{x} and @var{y} is "0".
3695 See below for the list of accepted constants and functions.
3698 The parameters for @var{x} and @var{y} are expressions containing the
3699 following constants and functions:
3703 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
3707 horizontal and vertical chroma subsample values. For example for the
3708 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3711 the height of each text line
3719 @item max_glyph_a, ascent
3720 the maximum distance from the baseline to the highest/upper grid
3721 coordinate used to place a glyph outline point, for all the rendered
3723 It is a positive value, due to the grid's orientation with the Y axis
3726 @item max_glyph_d, descent
3727 the maximum distance from the baseline to the lowest grid coordinate
3728 used to place a glyph outline point, for all the rendered glyphs.
3729 This is a negative value, due to the grid's orientation, with the Y axis
3733 maximum glyph height, that is the maximum height for all the glyphs
3734 contained in the rendered text, it is equivalent to @var{ascent} -
3738 maximum glyph width, that is the maximum width for all the glyphs
3739 contained in the rendered text
3742 the number of input frame, starting from 0
3744 @item rand(min, max)
3745 return a random number included between @var{min} and @var{max}
3748 input sample aspect ratio
3751 timestamp expressed in seconds, NAN if the input timestamp is unknown
3754 the height of the rendered text
3757 the width of the rendered text
3761 the x and y offset coordinates where the text is drawn.
3763 These parameters allow the @var{x} and @var{y} expressions to refer
3764 each other, so you can for example specify @code{y=x/dar}.
3767 If libavfilter was built with @code{--enable-fontconfig}, then
3768 @option{fontfile} can be a fontconfig pattern or omitted.
3770 @anchor{drawtext_expansion}
3771 @subsection Text expansion
3773 If @option{expansion} is set to @code{strftime},
3774 the filter recognizes strftime() sequences in the provided text and
3775 expands them accordingly. Check the documentation of strftime(). This
3776 feature is deprecated.
3778 If @option{expansion} is set to @code{none}, the text is printed verbatim.
3780 If @option{expansion} is set to @code{normal} (which is the default),
3781 the following expansion mechanism is used.
3783 The backslash character '\', followed by any character, always expands to
3784 the second character.
3786 Sequence of the form @code{%@{...@}} are expanded. The text between the
3787 braces is a function name, possibly followed by arguments separated by ':'.
3788 If the arguments contain special characters or delimiters (':' or '@}'),
3789 they should be escaped.
3791 Note that they probably must also be escaped as the value for the
3792 @option{text} option in the filter argument string and as the filter
3793 argument in the filtergraph description, and possibly also for the shell,
3794 that makes up to four levels of escaping; using a text file avoids these
3797 The following functions are available:
3802 The expression evaluation result.
3804 It must take one argument specifying the expression to be evaluated,
3805 which accepts the same constants and functions as the @var{x} and
3806 @var{y} values. Note that not all constants should be used, for
3807 example the text size is not known when evaluating the expression, so
3808 the constants @var{text_w} and @var{text_h} will have an undefined
3812 The time at which the filter is running, expressed in UTC.
3813 It can accept an argument: a strftime() format string.
3816 The time at which the filter is running, expressed in the local time zone.
3817 It can accept an argument: a strftime() format string.
3820 Frame metadata. It must take one argument specifying metadata key.
3823 The frame number, starting from 0.
3826 A 1 character description of the current picture type.
3829 The timestamp of the current frame, in seconds, with microsecond accuracy.
3833 @subsection Examples
3837 Draw "Test Text" with font FreeSerif, using the default values for the
3838 optional parameters.
3841 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
3845 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
3846 and y=50 (counting from the top-left corner of the screen), text is
3847 yellow with a red box around it. Both the text and the box have an
3851 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
3852 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
3855 Note that the double quotes are not necessary if spaces are not used
3856 within the parameter list.
3859 Show the text at the center of the video frame:
3861 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
3865 Show a text line sliding from right to left in the last row of the video
3866 frame. The file @file{LONG_LINE} is assumed to contain a single line
3869 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
3873 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
3875 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
3879 Draw a single green letter "g", at the center of the input video.
3880 The glyph baseline is placed at half screen height.
3882 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
3886 Show text for 1 second every 3 seconds:
3888 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
3892 Use fontconfig to set the font. Note that the colons need to be escaped.
3894 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
3898 Print the date of a real-time encoding (see strftime(3)):
3900 drawtext='fontfile=FreeSans.ttf:text=%@{localtime:%a %b %d %Y@}'
3905 For more information about libfreetype, check:
3906 @url{http://www.freetype.org/}.
3908 For more information about fontconfig, check:
3909 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
3913 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
3915 The filter accepts the following options:
3920 Set low and high threshold values used by the Canny thresholding
3923 The high threshold selects the "strong" edge pixels, which are then
3924 connected through 8-connectivity with the "weak" edge pixels selected
3925 by the low threshold.
3927 @var{low} and @var{high} threshold values must be choosen in the range
3928 [0,1], and @var{low} should be lesser or equal to @var{high}.
3930 Default value for @var{low} is @code{20/255}, and default value for @var{high}
3936 edgedetect=low=0.1:high=0.4
3939 @section extractplanes
3941 Extract color channel components from input video stream into
3942 separate grayscale video streams.
3944 The filter accepts the following option:
3948 Set plane(s) to extract.
3950 Available values for planes are:
3961 Choosing planes not available in the input will result in an error.
3962 That means you cannot select @code{r}, @code{g}, @code{b} planes
3963 with @code{y}, @code{u}, @code{v} planes at same time.
3966 @subsection Examples
3970 Extract luma, u and v color channel component from input video frame
3971 into 3 grayscale outputs:
3973 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
3979 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
3981 For each input image, the filter will compute the optimal mapping from
3982 the input to the output given the codebook length, that is the number
3983 of distinct output colors.
3985 This filter accepts the following options.
3988 @item codebook_length, l
3989 Set codebook length. The value must be a positive integer, and
3990 represents the number of distinct output colors. Default value is 256.
3993 Set the maximum number of iterations to apply for computing the optimal
3994 mapping. The higher the value the better the result and the higher the
3995 computation time. Default value is 1.
3998 Set a random seed, must be an integer included between 0 and
3999 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
4000 will try to use a good random seed on a best effort basis.
4005 Apply fade-in/out effect to input video.
4007 This filter accepts the following options:
4011 The effect type -- can be either "in" for fade-in, or "out" for a fade-out
4013 Default is @code{in}.
4015 @item start_frame, s
4016 Specify the number of the start frame for starting to apply the fade
4017 effect. Default is 0.
4020 The number of frames for which the fade effect has to last. At the end of the
4021 fade-in effect the output video will have the same intensity as the input video,
4022 at the end of the fade-out transition the output video will be filled with the
4023 selected @option{color}.
4027 If set to 1, fade only alpha channel, if one exists on the input.
4030 @item start_time, st
4031 Specify the timestamp (in seconds) of the frame to start to apply the fade
4032 effect. If both start_frame and start_time are specified, the fade will start at
4033 whichever comes last. Default is 0.
4036 The number of seconds for which the fade effect has to last. At the end of the
4037 fade-in effect the output video will have the same intensity as the input video,
4038 at the end of the fade-out transition the output video will be filled with the
4039 selected @option{color}.
4040 If both duration and nb_frames are specified, duration is used. Default is 0.
4043 Specify the color of the fade. Default is "black".
4046 @subsection Examples
4050 Fade in first 30 frames of video:
4055 The command above is equivalent to:
4061 Fade out last 45 frames of a 200-frame video:
4064 fade=type=out:start_frame=155:nb_frames=45
4068 Fade in first 25 frames and fade out last 25 frames of a 1000-frame video:
4070 fade=in:0:25, fade=out:975:25
4074 Make first 5 frames yellow, then fade in from frame 5-24:
4076 fade=in:5:20:color=yellow
4080 Fade in alpha over first 25 frames of video:
4082 fade=in:0:25:alpha=1
4086 Make first 5.5 seconds black, then fade in for 0.5 seconds:
4088 fade=t=in:st=5.5:d=0.5
4095 Extract a single field from an interlaced image using stride
4096 arithmetic to avoid wasting CPU time. The output frames are marked as
4099 The filter accepts the following options:
4103 Specify whether to extract the top (if the value is @code{0} or
4104 @code{top}) or the bottom field (if the value is @code{1} or
4110 Field matching filter for inverse telecine. It is meant to reconstruct the
4111 progressive frames from a telecined stream. The filter does not drop duplicated
4112 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
4113 followed by a decimation filter such as @ref{decimate} in the filtergraph.
4115 The separation of the field matching and the decimation is notably motivated by
4116 the possibility of inserting a de-interlacing filter fallback between the two.
4117 If the source has mixed telecined and real interlaced content,
4118 @code{fieldmatch} will not be able to match fields for the interlaced parts.
4119 But these remaining combed frames will be marked as interlaced, and thus can be
4120 de-interlaced by a later filter such as @ref{yadif} before decimation.
4122 In addition to the various configuration options, @code{fieldmatch} can take an
4123 optional second stream, activated through the @option{ppsrc} option. If
4124 enabled, the frames reconstruction will be based on the fields and frames from
4125 this second stream. This allows the first input to be pre-processed in order to
4126 help the various algorithms of the filter, while keeping the output lossless
4127 (assuming the fields are matched properly). Typically, a field-aware denoiser,
4128 or brightness/contrast adjustments can help.
4130 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
4131 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
4132 which @code{fieldmatch} is based on. While the semantic and usage are very
4133 close, some behaviour and options names can differ.
4135 The filter accepts the following options:
4139 Specify the assumed field order of the input stream. Available values are:
4143 Auto detect parity (use FFmpeg's internal parity value).
4145 Assume bottom field first.
4147 Assume top field first.
4150 Note that it is sometimes recommended not to trust the parity announced by the
4153 Default value is @var{auto}.
4156 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
4157 sense that it won't risk creating jerkiness due to duplicate frames when
4158 possible, but if there are bad edits or blended fields it will end up
4159 outputting combed frames when a good match might actually exist. On the other
4160 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
4161 but will almost always find a good frame if there is one. The other values are
4162 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
4163 jerkiness and creating duplicate frames versus finding good matches in sections
4164 with bad edits, orphaned fields, blended fields, etc.
4166 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
4168 Available values are:
4172 2-way matching (p/c)
4174 2-way matching, and trying 3rd match if still combed (p/c + n)
4176 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
4178 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
4179 still combed (p/c + n + u/b)
4181 3-way matching (p/c/n)
4183 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
4184 detected as combed (p/c/n + u/b)
4187 The parenthesis at the end indicate the matches that would be used for that
4188 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
4191 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
4194 Default value is @var{pc_n}.
4197 Mark the main input stream as a pre-processed input, and enable the secondary
4198 input stream as the clean source to pick the fields from. See the filter
4199 introduction for more details. It is similar to the @option{clip2} feature from
4202 Default value is @code{0} (disabled).
4205 Set the field to match from. It is recommended to set this to the same value as
4206 @option{order} unless you experience matching failures with that setting. In
4207 certain circumstances changing the field that is used to match from can have a
4208 large impact on matching performance. Available values are:
4212 Automatic (same value as @option{order}).
4214 Match from the bottom field.
4216 Match from the top field.
4219 Default value is @var{auto}.
4222 Set whether or not chroma is included during the match comparisons. In most
4223 cases it is recommended to leave this enabled. You should set this to @code{0}
4224 only if your clip has bad chroma problems such as heavy rainbowing or other
4225 artifacts. Setting this to @code{0} could also be used to speed things up at
4226 the cost of some accuracy.
4228 Default value is @code{1}.
4232 These define an exclusion band which excludes the lines between @option{y0} and
4233 @option{y1} from being included in the field matching decision. An exclusion
4234 band can be used to ignore subtitles, a logo, or other things that may
4235 interfere with the matching. @option{y0} sets the starting scan line and
4236 @option{y1} sets the ending line; all lines in between @option{y0} and
4237 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
4238 @option{y0} and @option{y1} to the same value will disable the feature.
4239 @option{y0} and @option{y1} defaults to @code{0}.
4242 Set the scene change detection threshold as a percentage of maximum change on
4243 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
4244 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
4245 @option{scthresh} is @code{[0.0, 100.0]}.
4247 Default value is @code{12.0}.
4250 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
4251 account the combed scores of matches when deciding what match to use as the
4252 final match. Available values are:
4256 No final matching based on combed scores.
4258 Combed scores are only used when a scene change is detected.
4260 Use combed scores all the time.
4263 Default is @var{sc}.
4266 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
4267 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
4268 Available values are:
4272 No forced calculation.
4274 Force p/c/n calculations.
4276 Force p/c/n/u/b calculations.
4279 Default value is @var{none}.
4282 This is the area combing threshold used for combed frame detection. This
4283 essentially controls how "strong" or "visible" combing must be to be detected.
4284 Larger values mean combing must be more visible and smaller values mean combing
4285 can be less visible or strong and still be detected. Valid settings are from
4286 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
4287 be detected as combed). This is basically a pixel difference value. A good
4288 range is @code{[8, 12]}.
4290 Default value is @code{9}.
4293 Sets whether or not chroma is considered in the combed frame decision. Only
4294 disable this if your source has chroma problems (rainbowing, etc.) that are
4295 causing problems for the combed frame detection with chroma enabled. Actually,
4296 using @option{chroma}=@var{0} is usually more reliable, except for the case
4297 where there is chroma only combing in the source.
4299 Default value is @code{0}.
4303 Respectively set the x-axis and y-axis size of the window used during combed
4304 frame detection. This has to do with the size of the area in which
4305 @option{combpel} pixels are required to be detected as combed for a frame to be
4306 declared combed. See the @option{combpel} parameter description for more info.
4307 Possible values are any number that is a power of 2 starting at 4 and going up
4310 Default value is @code{16}.
4313 The number of combed pixels inside any of the @option{blocky} by
4314 @option{blockx} size blocks on the frame for the frame to be detected as
4315 combed. While @option{cthresh} controls how "visible" the combing must be, this
4316 setting controls "how much" combing there must be in any localized area (a
4317 window defined by the @option{blockx} and @option{blocky} settings) on the
4318 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
4319 which point no frames will ever be detected as combed). This setting is known
4320 as @option{MI} in TFM/VFM vocabulary.
4322 Default value is @code{80}.
4325 @anchor{p/c/n/u/b meaning}
4326 @subsection p/c/n/u/b meaning
4328 @subsubsection p/c/n
4330 We assume the following telecined stream:
4333 Top fields: 1 2 2 3 4
4334 Bottom fields: 1 2 3 4 4
4337 The numbers correspond to the progressive frame the fields relate to. Here, the
4338 first two frames are progressive, the 3rd and 4th are combed, and so on.
4340 When @code{fieldmatch} is configured to run a matching from bottom
4341 (@option{field}=@var{bottom}) this is how this input stream get transformed:
4346 B 1 2 3 4 4 <-- matching reference
4355 As a result of the field matching, we can see that some frames get duplicated.
4356 To perform a complete inverse telecine, you need to rely on a decimation filter
4357 after this operation. See for instance the @ref{decimate} filter.
4359 The same operation now matching from top fields (@option{field}=@var{top})
4364 T 1 2 2 3 4 <-- matching reference
4374 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
4375 basically, they refer to the frame and field of the opposite parity:
4378 @item @var{p} matches the field of the opposite parity in the previous frame
4379 @item @var{c} matches the field of the opposite parity in the current frame
4380 @item @var{n} matches the field of the opposite parity in the next frame
4385 The @var{u} and @var{b} matching are a bit special in the sense that they match
4386 from the opposite parity flag. In the following examples, we assume that we are
4387 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
4388 'x' is placed above and below each matched fields.
4390 With bottom matching (@option{field}=@var{bottom}):
4395 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
4396 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
4404 With top matching (@option{field}=@var{top}):
4409 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
4410 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
4418 @subsection Examples
4420 Simple IVTC of a top field first telecined stream:
4422 fieldmatch=order=tff:combmatch=none, decimate
4425 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
4427 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
4432 Transform the field order of the input video.
4434 This filter accepts the following options:
4439 Output field order. Valid values are @var{tff} for top field first or @var{bff}
4440 for bottom field first.
4443 Default value is @samp{tff}.
4445 Transformation is achieved by shifting the picture content up or down
4446 by one line, and filling the remaining line with appropriate picture content.
4447 This method is consistent with most broadcast field order converters.
4449 If the input video is not flagged as being interlaced, or it is already
4450 flagged as being of the required output field order then this filter does
4451 not alter the incoming video.
4453 This filter is very useful when converting to or from PAL DV material,
4454 which is bottom field first.
4458 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
4463 Buffer input images and send them when they are requested.
4465 This filter is mainly useful when auto-inserted by the libavfilter
4468 The filter does not take parameters.
4473 Convert the input video to one of the specified pixel formats.
4474 Libavfilter will try to pick one that is supported for the input to
4477 This filter accepts the following parameters:
4481 A '|'-separated list of pixel format names, for example
4482 "pix_fmts=yuv420p|monow|rgb24".
4486 @subsection Examples
4490 Convert the input video to the format @var{yuv420p}
4492 format=pix_fmts=yuv420p
4495 Convert the input video to any of the formats in the list
4497 format=pix_fmts=yuv420p|yuv444p|yuv410p
4504 Convert the video to specified constant frame rate by duplicating or dropping
4505 frames as necessary.
4507 This filter accepts the following named parameters:
4511 Desired output frame rate. The default is @code{25}.
4516 Possible values are:
4519 zero round towards 0
4523 round towards -infinity
4525 round towards +infinity
4529 The default is @code{near}.
4532 Assume the first PTS should be the given value, in seconds. This allows for
4533 padding/trimming at the start of stream. By default, no assumption is made
4534 about the first frame's expected PTS, so no padding or trimming is done.
4535 For example, this could be set to 0 to pad the beginning with duplicates of
4536 the first frame if a video stream starts after the audio stream or to trim any
4537 frames with a negative PTS.
4541 Alternatively, the options can be specified as a flat string:
4542 @var{fps}[:@var{round}].
4544 See also the @ref{setpts} filter.
4546 @subsection Examples
4550 A typical usage in order to set the fps to 25:
4556 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
4558 fps=fps=film:round=near
4564 Pack two different video streams into a stereoscopic video, setting proper
4565 metadata on supported codecs. The two views should have the same size and
4566 framerate and processing will stop when the shorter video ends. Please note
4567 that you may conveniently adjust view properties with the @ref{scale} and
4570 This filter accepts the following named parameters:
4574 Desired packing format. Supported values are:
4579 Views are next to each other (default).
4582 Views are on top of each other.
4585 Views are packed by line.
4588 Views are eacked by column.
4591 Views are temporally interleaved.
4597 Some examples follow:
4600 # Convert left and right views into a frame sequential video.
4601 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
4603 # Convert views into a side-by-side video with the same output resolution as the input.
4604 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
4609 Select one frame every N-th frame.
4611 This filter accepts the following option:
4614 Select frame after every @code{step} frames.
4615 Allowed values are positive integers higher than 0. Default value is @code{1}.
4621 Apply a frei0r effect to the input video.
4623 To enable compilation of this filter you need to install the frei0r
4624 header and configure FFmpeg with @code{--enable-frei0r}.
4626 This filter accepts the following options:
4631 The name to the frei0r effect to load. If the environment variable
4632 @env{FREI0R_PATH} is defined, the frei0r effect is searched in each one of the
4633 directories specified by the colon separated list in @env{FREIOR_PATH},
4634 otherwise in the standard frei0r paths, which are in this order:
4635 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
4636 @file{/usr/lib/frei0r-1/}.
4639 A '|'-separated list of parameters to pass to the frei0r effect.
4643 A frei0r effect parameter can be a boolean (whose values are specified
4644 with "y" and "n"), a double, a color (specified by the syntax
4645 @var{R}/@var{G}/@var{B}, (@var{R}, @var{G}, and @var{B} being float
4646 numbers from 0.0 to 1.0) or by a color description specified in the "Color"
4647 section in the ffmpeg-utils manual), a position (specified by the syntax @var{X}/@var{Y},
4648 @var{X} and @var{Y} being float numbers) and a string.
4650 The number and kind of parameters depend on the loaded effect. If an
4651 effect parameter is not specified the default value is set.
4653 @subsection Examples
4657 Apply the distort0r effect, set the first two double parameters:
4659 frei0r=filter_name=distort0r:filter_params=0.5|0.01
4663 Apply the colordistance effect, take a color as first parameter:
4665 frei0r=colordistance:0.2/0.3/0.4
4666 frei0r=colordistance:violet
4667 frei0r=colordistance:0x112233
4671 Apply the perspective effect, specify the top left and top right image
4674 frei0r=perspective:0.2/0.2|0.8/0.2
4678 For more information see:
4679 @url{http://frei0r.dyne.org}
4683 The filter accepts the following options:
4687 Set the luminance expression.
4689 Set the chrominance blue expression.
4691 Set the chrominance red expression.
4693 Set the alpha expression.
4695 Set the red expression.
4697 Set the green expression.
4699 Set the blue expression.
4702 The colorspace is selected according to the specified options. If one
4703 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
4704 options is specified, the filter will automatically select a YCbCr
4705 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
4706 @option{blue_expr} options is specified, it will select an RGB
4709 If one of the chrominance expression is not defined, it falls back on the other
4710 one. If no alpha expression is specified it will evaluate to opaque value.
4711 If none of chrominance expressions are specified, they will evaluate
4712 to the luminance expression.
4714 The expressions can use the following variables and functions:
4718 The sequential number of the filtered frame, starting from @code{0}.
4722 The coordinates of the current sample.
4726 The width and height of the image.
4730 Width and height scale depending on the currently filtered plane. It is the
4731 ratio between the corresponding luma plane number of pixels and the current
4732 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4733 @code{0.5,0.5} for chroma planes.
4736 Time of the current frame, expressed in seconds.
4739 Return the value of the pixel at location (@var{x},@var{y}) of the current
4743 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
4747 Return the value of the pixel at location (@var{x},@var{y}) of the
4748 blue-difference chroma plane. Return 0 if there is no such plane.
4751 Return the value of the pixel at location (@var{x},@var{y}) of the
4752 red-difference chroma plane. Return 0 if there is no such plane.
4757 Return the value of the pixel at location (@var{x},@var{y}) of the
4758 red/green/blue component. Return 0 if there is no such component.
4761 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
4762 plane. Return 0 if there is no such plane.
4765 For functions, if @var{x} and @var{y} are outside the area, the value will be
4766 automatically clipped to the closer edge.
4768 @subsection Examples
4772 Flip the image horizontally:
4778 Generate a bidimensional sine wave, with angle @code{PI/3} and a
4779 wavelength of 100 pixels:
4781 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
4785 Generate a fancy enigmatic moving light:
4787 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
4791 Generate a quick emboss effect:
4793 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
4797 Modify RGB components depending on pixel position:
4799 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
4805 Fix the banding artifacts that are sometimes introduced into nearly flat
4806 regions by truncation to 8bit color depth.
4807 Interpolate the gradients that should go where the bands are, and
4810 This filter is designed for playback only. Do not use it prior to
4811 lossy compression, because compression tends to lose the dither and
4812 bring back the bands.
4814 This filter accepts the following options:
4819 The maximum amount by which the filter will change any one pixel. Also the
4820 threshold for detecting nearly flat regions. Acceptable values range from .51 to
4821 64, default value is 1.2, out-of-range values will be clipped to the valid
4825 The neighborhood to fit the gradient to. A larger radius makes for smoother
4826 gradients, but also prevents the filter from modifying the pixels near detailed
4827 regions. Acceptable values are 8-32, default value is 16, out-of-range values
4828 will be clipped to the valid range.
4832 Alternatively, the options can be specified as a flat string:
4833 @var{strength}[:@var{radius}]
4835 @subsection Examples
4839 Apply the filter with a @code{3.5} strength and radius of @code{8}:
4845 Specify radius, omitting the strength (which will fall-back to the default
4856 Apply a Hald CLUT to a video stream.
4858 First input is the video stream to process, and second one is the Hald CLUT.
4859 The Hald CLUT input can be a simple picture or a complete video stream.
4861 The filter accepts the following options:
4865 Force termination when the shortest input terminates. Default is @code{0}.
4867 Continue applying the last CLUT after the end of the stream. A value of
4868 @code{0} disable the filter after the last frame of the CLUT is reached.
4869 Default is @code{1}.
4872 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
4873 filters share the same internals).
4875 More information about the Hald CLUT can be found on Eskil Steenberg's website
4876 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
4878 @subsection Workflow examples
4880 @subsubsection Hald CLUT video stream
4882 Generate an identity Hald CLUT stream altered with various effects:
4884 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
4887 Note: make sure you use a lossless codec.
4889 Then use it with @code{haldclut} to apply it on some random stream:
4891 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
4894 The Hald CLUT will be applied to the 10 first seconds (duration of
4895 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
4896 to the remaining frames of the @code{mandelbrot} stream.
4898 @subsubsection Hald CLUT with preview
4900 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
4901 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
4902 biggest possible square starting at the top left of the picture. The remaining
4903 padding pixels (bottom or right) will be ignored. This area can be used to add
4904 a preview of the Hald CLUT.
4906 Typically, the following generated Hald CLUT will be supported by the
4907 @code{haldclut} filter:
4910 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
4911 pad=iw+320 [padded_clut];
4912 smptebars=s=320x256, split [a][b];
4913 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
4914 [main][b] overlay=W-320" -frames:v 1 clut.png
4917 It contains the original and a preview of the effect of the CLUT: SMPTE color
4918 bars are displayed on the right-top, and below the same color bars processed by
4921 Then, the effect of this Hald CLUT can be visualized with:
4923 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
4928 Flip the input video horizontally.
4930 For example to horizontally flip the input video with @command{ffmpeg}:
4932 ffmpeg -i in.avi -vf "hflip" out.avi
4936 This filter applies a global color histogram equalization on a
4939 It can be used to correct video that has a compressed range of pixel
4940 intensities. The filter redistributes the pixel intensities to
4941 equalize their distribution across the intensity range. It may be
4942 viewed as an "automatically adjusting contrast filter". This filter is
4943 useful only for correcting degraded or poorly captured source
4946 The filter accepts the following options:
4950 Determine the amount of equalization to be applied. As the strength
4951 is reduced, the distribution of pixel intensities more-and-more
4952 approaches that of the input frame. The value must be a float number
4953 in the range [0,1] and defaults to 0.200.
4956 Set the maximum intensity that can generated and scale the output
4957 values appropriately. The strength should be set as desired and then
4958 the intensity can be limited if needed to avoid washing-out. The value
4959 must be a float number in the range [0,1] and defaults to 0.210.
4962 Set the antibanding level. If enabled the filter will randomly vary
4963 the luminance of output pixels by a small amount to avoid banding of
4964 the histogram. Possible values are @code{none}, @code{weak} or
4965 @code{strong}. It defaults to @code{none}.
4970 Compute and draw a color distribution histogram for the input video.
4972 The computed histogram is a representation of the color component
4973 distribution in an image.
4975 The filter accepts the following options:
4981 It accepts the following values:
4984 Standard histogram that displays the color components distribution in an
4985 image. Displays color graph for each color component. Shows distribution of
4986 the Y, U, V, A or R, G, B components, depending on input format, in the
4987 current frame. Below each graph a color component scale meter is shown.
4990 Displays chroma values (U/V color placement) in a two dimensional
4991 graph (which is called a vectorscope). The brighter a pixel in the
4992 vectorscope, the more pixels of the input frame correspond to that pixel
4993 (i.e., more pixels have this chroma value). The V component is displayed on
4994 the horizontal (X) axis, with the leftmost side being V = 0 and the rightmost
4995 side being V = 255. The U component is displayed on the vertical (Y) axis,
4996 with the top representing U = 0 and the bottom representing U = 255.
4998 The position of a white pixel in the graph corresponds to the chroma value of
4999 a pixel of the input clip. The graph can therefore be used to read the hue
5000 (color flavor) and the saturation (the dominance of the hue in the color). As
5001 the hue of a color changes, it moves around the square. At the center of the
5002 square the saturation is zero, which means that the corresponding pixel has no
5003 color. If the amount of a specific color is increased (while leaving the other
5004 colors unchanged) the saturation increases, and the indicator moves towards
5005 the edge of the square.
5008 Chroma values in vectorscope, similar as @code{color} but actual chroma values
5012 Per row/column color component graph. In row mode, the graph on the left side
5013 represents color component value 0 and the right side represents value = 255.
5014 In column mode, the top side represents color component value = 0 and bottom
5015 side represents value = 255.
5017 Default value is @code{levels}.
5020 Set height of level in @code{levels}. Default value is @code{200}.
5021 Allowed range is [50, 2048].
5024 Set height of color scale in @code{levels}. Default value is @code{12}.
5025 Allowed range is [0, 40].
5028 Set step for @code{waveform} mode. Smaller values are useful to find out how
5029 many values of the same luminance are distributed across input rows/columns.
5030 Default value is @code{10}. Allowed range is [1, 255].
5033 Set mode for @code{waveform}. Can be either @code{row}, or @code{column}.
5034 Default is @code{row}.
5036 @item waveform_mirror
5037 Set mirroring mode for @code{waveform}. @code{0} means unmirrored, @code{1}
5038 means mirrored. In mirrored mode, higher values will be represented on the left
5039 side for @code{row} mode and at the top for @code{column} mode. Default is
5040 @code{0} (unmirrored).
5043 Set display mode for @code{waveform} and @code{levels}.
5044 It accepts the following values:
5047 Display separate graph for the color components side by side in
5048 @code{row} waveform mode or one below the other in @code{column} waveform mode
5049 for @code{waveform} histogram mode. For @code{levels} histogram mode,
5050 per color component graphs are placed below each other.
5052 Using this display mode in @code{waveform} histogram mode makes it easy to
5053 spot color casts in the highlights and shadows of an image, by comparing the
5054 contours of the top and the bottom graphs of each waveform. Since whites,
5055 grays, and blacks are characterized by exactly equal amounts of red, green,
5056 and blue, neutral areas of the picture should display three waveforms of
5057 roughly equal width/height. If not, the correction is easy to perform by
5058 making level adjustments the three waveforms.
5061 Presents information identical to that in the @code{parade}, except
5062 that the graphs representing color components are superimposed directly
5065 This display mode in @code{waveform} histogram mode makes it easier to spot
5066 relative differences or similarities in overlapping areas of the color
5067 components that are supposed to be identical, such as neutral whites, grays,
5070 Default is @code{parade}.
5073 Set mode for @code{levels}. Can be either @code{linear}, or @code{logarithmic}.
5074 Default is @code{linear}.
5077 @subsection Examples
5082 Calculate and draw histogram:
5084 ffplay -i input -vf histogram
5092 High precision/quality 3d denoise filter. This filter aims to reduce
5093 image noise producing smooth images and making still images really
5094 still. It should enhance compressibility.
5096 It accepts the following optional parameters:
5100 a non-negative float number which specifies spatial luma strength,
5103 @item chroma_spatial
5104 a non-negative float number which specifies spatial chroma strength,
5105 defaults to 3.0*@var{luma_spatial}/4.0
5108 a float number which specifies luma temporal strength, defaults to
5109 6.0*@var{luma_spatial}/4.0
5112 a float number which specifies chroma temporal strength, defaults to
5113 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}
5118 Modify the hue and/or the saturation of the input.
5120 This filter accepts the following options:
5124 Specify the hue angle as a number of degrees. It accepts an expression,
5125 and defaults to "0".
5128 Specify the saturation in the [-10,10] range. It accepts an expression and
5132 Specify the hue angle as a number of radians. It accepts an
5133 expression, and defaults to "0".
5136 Specify the brightness in the [-10,10] range. It accepts an expression and
5140 @option{h} and @option{H} are mutually exclusive, and can't be
5141 specified at the same time.
5143 The @option{b}, @option{h}, @option{H} and @option{s} option values are
5144 expressions containing the following constants:
5148 frame count of the input frame starting from 0
5151 presentation timestamp of the input frame expressed in time base units
5154 frame rate of the input video, NAN if the input frame rate is unknown
5157 timestamp expressed in seconds, NAN if the input timestamp is unknown
5160 time base of the input video
5163 @subsection Examples
5167 Set the hue to 90 degrees and the saturation to 1.0:
5173 Same command but expressing the hue in radians:
5179 Rotate hue and make the saturation swing between 0
5180 and 2 over a period of 1 second:
5182 hue="H=2*PI*t: s=sin(2*PI*t)+1"
5186 Apply a 3 seconds saturation fade-in effect starting at 0:
5191 The general fade-in expression can be written as:
5193 hue="s=min(0\, max((t-START)/DURATION\, 1))"
5197 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
5199 hue="s=max(0\, min(1\, (8-t)/3))"
5202 The general fade-out expression can be written as:
5204 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
5209 @subsection Commands
5211 This filter supports the following commands:
5217 Modify the hue and/or the saturation and/or brightness of the input video.
5218 The command accepts the same syntax of the corresponding option.
5220 If the specified expression is not valid, it is kept at its current
5226 Detect video interlacing type.
5228 This filter tries to detect if the input is interlaced or progressive,
5229 top or bottom field first.
5231 The filter accepts the following options:
5235 Set interlacing threshold.
5237 Set progressive threshold.
5242 Deinterleave or interleave fields.
5244 This filter allows to process interlaced images fields without
5245 deinterlacing them. Deinterleaving splits the input frame into 2
5246 fields (so called half pictures). Odd lines are moved to the top
5247 half of the output image, even lines to the bottom half.
5248 You can process (filter) them independently and then re-interleave them.
5250 The filter accepts the following options:
5254 @item chroma_mode, c
5256 Available values for @var{luma_mode}, @var{chroma_mode} and
5257 @var{alpha_mode} are:
5263 @item deinterleave, d
5264 Deinterleave fields, placing one above the other.
5267 Interleave fields. Reverse the effect of deinterleaving.
5269 Default value is @code{none}.
5272 @item chroma_swap, cs
5273 @item alpha_swap, as
5274 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
5279 Simple interlacing filter from progressive contents. This interleaves upper (or
5280 lower) lines from odd frames with lower (or upper) lines from even frames,
5281 halving the frame rate and preserving image height. A vertical lowpass filter
5282 is always applied in order to avoid twitter effects and reduce moiré patterns.
5285 Original Original New Frame
5286 Frame 'j' Frame 'j+1' (tff)
5287 ========== =========== ==================
5288 Line 0 --------------------> Frame 'j' Line 0
5289 Line 1 Line 1 ----> Frame 'j+1' Line 1
5290 Line 2 ---------------------> Frame 'j' Line 2
5291 Line 3 Line 3 ----> Frame 'j+1' Line 3
5293 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
5296 It accepts the following optional parameters:
5300 determines whether the interlaced frame is taken from the even (tff - default)
5301 or odd (bff) lines of the progressive frame.
5306 Deinterlace input video by applying Donald Graft's adaptive kernel
5307 deinterling. Work on interlaced parts of a video to produce
5310 The description of the accepted parameters follows.
5314 Set the threshold which affects the filter's tolerance when
5315 determining if a pixel line must be processed. It must be an integer
5316 in the range [0,255] and defaults to 10. A value of 0 will result in
5317 applying the process on every pixels.
5320 Paint pixels exceeding the threshold value to white if set to 1.
5324 Set the fields order. Swap fields if set to 1, leave fields alone if
5328 Enable additional sharpening if set to 1. Default is 0.
5331 Enable twoway sharpening if set to 1. Default is 0.
5334 @subsection Examples
5338 Apply default values:
5340 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
5344 Enable additional sharpening:
5350 Paint processed pixels in white:
5359 Apply a 3D LUT to an input video.
5361 The filter accepts the following options:
5365 Set the 3D LUT file name.
5367 Currently supported formats:
5379 Select interpolation mode.
5381 Available values are:
5385 Use values from the nearest defined point.
5387 Interpolate values using the 8 points defining a cube.
5389 Interpolate values using a tetrahedron.
5393 @section lut, lutrgb, lutyuv
5395 Compute a look-up table for binding each pixel component input value
5396 to an output value, and apply it to input video.
5398 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
5399 to an RGB input video.
5401 These filters accept the following options:
5404 set first pixel component expression
5406 set second pixel component expression
5408 set third pixel component expression
5410 set fourth pixel component expression, corresponds to the alpha component
5413 set red component expression
5415 set green component expression
5417 set blue component expression
5419 alpha component expression
5422 set Y/luminance component expression
5424 set U/Cb component expression
5426 set V/Cr component expression
5429 Each of them specifies the expression to use for computing the lookup table for
5430 the corresponding pixel component values.
5432 The exact component associated to each of the @var{c*} options depends on the
5435 The @var{lut} filter requires either YUV or RGB pixel formats in input,
5436 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
5438 The expressions can contain the following constants and functions:
5443 the input width and height
5446 input value for the pixel component
5449 the input value clipped in the @var{minval}-@var{maxval} range
5452 maximum value for the pixel component
5455 minimum value for the pixel component
5458 the negated value for the pixel component value clipped in the
5459 @var{minval}-@var{maxval} range , it corresponds to the expression
5460 "maxval-clipval+minval"
5463 the computed value in @var{val} clipped in the
5464 @var{minval}-@var{maxval} range
5466 @item gammaval(gamma)
5467 the computed gamma correction value of the pixel component value
5468 clipped in the @var{minval}-@var{maxval} range, corresponds to the
5470 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
5474 All expressions default to "val".
5476 @subsection Examples
5482 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
5483 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
5486 The above is the same as:
5488 lutrgb="r=negval:g=negval:b=negval"
5489 lutyuv="y=negval:u=negval:v=negval"
5499 Remove chroma components, turns the video into a graytone image:
5501 lutyuv="u=128:v=128"
5505 Apply a luma burning effect:
5511 Remove green and blue components:
5517 Set a constant alpha channel value on input:
5519 format=rgba,lutrgb=a="maxval-minval/2"
5523 Correct luminance gamma by a 0.5 factor:
5525 lutyuv=y=gammaval(0.5)
5529 Discard least significant bits of luma:
5531 lutyuv=y='bitand(val, 128+64+32)'
5535 @section mergeplanes
5537 Merge color channel components from several video streams.
5539 The filter accepts up to 4 input streams, and merge selected input
5540 planes to the output video.
5542 This filter accepts the following options:
5545 Set input to output plane mapping. Default is @code{0}.
5547 The mappings is specified as a bitmap. It should be specified as a
5548 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
5549 mapping for the first plane of the output stream. 'A' sets the number of
5550 the input stream to use (from 0 to 3), and 'a' the plane number of the
5551 corresponding input to use (from 0 to 3). The rest of the mappings is
5552 similar, 'Bb' describes the mapping for the output stream second
5553 plane, 'Cc' describes the mapping for the output stream third plane and
5554 'Dd' describes the mapping for the output stream fourth plane.
5557 Set output pixel format. Default is @code{yuva444p}.
5560 @subsection Examples
5564 Merge three gray video streams of same width and height into single video stream:
5566 [a0][a1][a2]mergeplanes=0x001020:yuv444p
5570 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
5572 [a0][a1]mergeplanes=0x00010210:yuva444p
5576 Swap Y and A plane in yuva444p stream:
5578 format=yuva444p,mergeplanes=0x03010200:yuva444p
5582 Swap U and V plane in yuv420p stream:
5584 format=yuv420p,mergeplanes=0x000201:yuv420p
5588 Cast a rgb24 clip to yuv444p:
5590 format=rgb24,mergeplanes=0x000102:yuv444p
5596 Apply motion-compensation deinterlacing.
5598 It needs one field per frame as input and must thus be used together
5599 with yadif=1/3 or equivalent.
5601 This filter accepts the following options:
5604 Set the deinterlacing mode.
5606 It accepts one of the following values:
5611 use iterative motion estimation
5613 like @samp{slow}, but use multiple reference frames.
5615 Default value is @samp{fast}.
5618 Set the picture field parity assumed for the input video. It must be
5619 one of the following values:
5623 assume top field first
5625 assume bottom field first
5628 Default value is @samp{bff}.
5631 Set per-block quantization parameter (QP) used by the internal
5634 Higher values should result in a smoother motion vector field but less
5635 optimal individual vectors. Default value is 1.
5640 Apply an MPlayer filter to the input video.
5642 This filter provides a wrapper around some of the filters of
5645 This wrapper is considered experimental. Some of the wrapped filters
5646 may not work properly and we may drop support for them, as they will
5647 be implemented natively into FFmpeg. Thus you should avoid
5648 depending on them when writing portable scripts.
5650 The filter accepts the parameters:
5651 @var{filter_name}[:=]@var{filter_params}
5653 @var{filter_name} is the name of a supported MPlayer filter,
5654 @var{filter_params} is a string containing the parameters accepted by
5657 The list of the currently supported filters follows:
5668 The parameter syntax and behavior for the listed filters are the same
5669 of the corresponding MPlayer filters. For detailed instructions check
5670 the "VIDEO FILTERS" section in the MPlayer manual.
5672 @subsection Examples
5676 Adjust gamma, brightness, contrast:
5682 See also mplayer(1), @url{http://www.mplayerhq.hu/}.
5686 Drop frames that do not differ greatly from the previous frame in
5687 order to reduce frame rate.
5689 The main use of this filter is for very-low-bitrate encoding
5690 (e.g. streaming over dialup modem), but it could in theory be used for
5691 fixing movies that were inverse-telecined incorrectly.
5693 A description of the accepted options follows.
5697 Set the maximum number of consecutive frames which can be dropped (if
5698 positive), or the minimum interval between dropped frames (if
5699 negative). If the value is 0, the frame is dropped unregarding the
5700 number of previous sequentially dropped frames.
5707 Set the dropping threshold values.
5709 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
5710 represent actual pixel value differences, so a threshold of 64
5711 corresponds to 1 unit of difference for each pixel, or the same spread
5712 out differently over the block.
5714 A frame is a candidate for dropping if no 8x8 blocks differ by more
5715 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
5716 meaning the whole image) differ by more than a threshold of @option{lo}.
5718 Default value for @option{hi} is 64*12, default value for @option{lo} is
5719 64*5, and default value for @option{frac} is 0.33.
5727 This filter accepts an integer in input, if non-zero it negates the
5728 alpha component (if available). The default value in input is 0.
5732 Force libavfilter not to use any of the specified pixel formats for the
5733 input to the next filter.
5735 This filter accepts the following parameters:
5739 A '|'-separated list of pixel format names, for example
5740 "pix_fmts=yuv420p|monow|rgb24".
5744 @subsection Examples
5748 Force libavfilter to use a format different from @var{yuv420p} for the
5749 input to the vflip filter:
5751 noformat=pix_fmts=yuv420p,vflip
5755 Convert the input video to any of the formats not contained in the list:
5757 noformat=yuv420p|yuv444p|yuv410p
5763 Add noise on video input frame.
5765 The filter accepts the following options:
5773 Set noise seed for specific pixel component or all pixel components in case
5774 of @var{all_seed}. Default value is @code{123457}.
5776 @item all_strength, alls
5777 @item c0_strength, c0s
5778 @item c1_strength, c1s
5779 @item c2_strength, c2s
5780 @item c3_strength, c3s
5781 Set noise strength for specific pixel component or all pixel components in case
5782 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
5784 @item all_flags, allf
5789 Set pixel component flags or set flags for all components if @var{all_flags}.
5790 Available values for component flags are:
5793 averaged temporal noise (smoother)
5795 mix random noise with a (semi)regular pattern
5797 temporal noise (noise pattern changes between frames)
5799 uniform noise (gaussian otherwise)
5803 @subsection Examples
5805 Add temporal and uniform noise to input video:
5807 noise=alls=20:allf=t+u
5812 Pass the video source unchanged to the output.
5816 Apply video transform using libopencv.
5818 To enable this filter install libopencv library and headers and
5819 configure FFmpeg with @code{--enable-libopencv}.
5821 This filter accepts the following parameters:
5826 The name of the libopencv filter to apply.
5829 The parameters to pass to the libopencv filter. If not specified the default
5834 Refer to the official libopencv documentation for more precise
5836 @url{http://opencv.willowgarage.com/documentation/c/image_filtering.html}
5838 Follows the list of supported libopencv filters.
5843 Dilate an image by using a specific structuring element.
5844 This filter corresponds to the libopencv function @code{cvDilate}.
5846 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
5848 @var{struct_el} represents a structuring element, and has the syntax:
5849 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
5851 @var{cols} and @var{rows} represent the number of columns and rows of
5852 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
5853 point, and @var{shape} the shape for the structuring element, and
5854 can be one of the values "rect", "cross", "ellipse", "custom".
5856 If the value for @var{shape} is "custom", it must be followed by a
5857 string of the form "=@var{filename}". The file with name
5858 @var{filename} is assumed to represent a binary image, with each
5859 printable character corresponding to a bright pixel. When a custom
5860 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
5861 or columns and rows of the read file are assumed instead.
5863 The default value for @var{struct_el} is "3x3+0x0/rect".
5865 @var{nb_iterations} specifies the number of times the transform is
5866 applied to the image, and defaults to 1.
5868 Follow some example:
5870 # use the default values
5873 # dilate using a structuring element with a 5x5 cross, iterate two times
5874 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
5876 # read the shape from the file diamond.shape, iterate two times
5877 # the file diamond.shape may contain a pattern of characters like this:
5883 # the specified cols and rows are ignored (but not the anchor point coordinates)
5884 ocv=dilate:0x0+2x2/custom=diamond.shape|2
5889 Erode an image by using a specific structuring element.
5890 This filter corresponds to the libopencv function @code{cvErode}.
5892 The filter accepts the parameters: @var{struct_el}:@var{nb_iterations},
5893 with the same syntax and semantics as the @ref{dilate} filter.
5897 Smooth the input video.
5899 The filter takes the following parameters:
5900 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
5902 @var{type} is the type of smooth filter to apply, and can be one of
5903 the following values: "blur", "blur_no_scale", "median", "gaussian",
5904 "bilateral". The default value is "gaussian".
5906 @var{param1}, @var{param2}, @var{param3}, and @var{param4} are
5907 parameters whose meanings depend on smooth type. @var{param1} and
5908 @var{param2} accept integer positive values or 0, @var{param3} and
5909 @var{param4} accept float values.
5911 The default value for @var{param1} is 3, the default value for the
5912 other parameters is 0.
5914 These parameters correspond to the parameters assigned to the
5915 libopencv function @code{cvSmooth}.
5920 Overlay one video on top of another.
5922 It takes two inputs and one output, the first input is the "main"
5923 video on which the second input is overlayed.
5925 This filter accepts the following parameters:
5927 A description of the accepted options follows.
5932 Set the expression for the x and y coordinates of the overlayed video
5933 on the main video. Default value is "0" for both expressions. In case
5934 the expression is invalid, it is set to a huge value (meaning that the
5935 overlay will not be displayed within the output visible area).
5938 The action to take when EOF is encountered on the secondary input, accepts one
5939 of the following values:
5943 repeat the last frame (the default)
5947 pass through the main input
5951 Set when the expressions for @option{x}, and @option{y} are evaluated.
5953 It accepts the following values:
5956 only evaluate expressions once during the filter initialization or
5957 when a command is processed
5960 evaluate expressions for each incoming frame
5963 Default value is @samp{frame}.
5966 If set to 1, force the output to terminate when the shortest input
5967 terminates. Default value is 0.
5970 Set the format for the output video.
5972 It accepts the following values:
5987 Default value is @samp{yuv420}.
5989 @item rgb @emph{(deprecated)}
5990 If set to 1, force the filter to accept inputs in the RGB
5991 color space. Default value is 0. This option is deprecated, use
5992 @option{format} instead.
5995 If set to 1, force the filter to draw the last overlay frame over the
5996 main input until the end of the stream. A value of 0 disables this
5997 behavior. Default value is 1.
6000 The @option{x}, and @option{y} expressions can contain the following
6006 main input width and height
6010 overlay input width and height
6014 the computed values for @var{x} and @var{y}. They are evaluated for
6019 horizontal and vertical chroma subsample values of the output
6020 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
6024 the number of input frame, starting from 0
6027 the position in the file of the input frame, NAN if unknown
6030 timestamp expressed in seconds, NAN if the input timestamp is unknown
6034 Note that the @var{n}, @var{pos}, @var{t} variables are available only
6035 when evaluation is done @emph{per frame}, and will evaluate to NAN
6036 when @option{eval} is set to @samp{init}.
6038 Be aware that frames are taken from each input video in timestamp
6039 order, hence, if their initial timestamps differ, it is a good idea
6040 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
6041 have them begin in the same zero timestamp, as it does the example for
6042 the @var{movie} filter.
6044 You can chain together more overlays but you should test the
6045 efficiency of such approach.
6047 @subsection Commands
6049 This filter supports the following commands:
6053 Modify the x and y of the overlay input.
6054 The command accepts the same syntax of the corresponding option.
6056 If the specified expression is not valid, it is kept at its current
6060 @subsection Examples
6064 Draw the overlay at 10 pixels from the bottom right corner of the main
6067 overlay=main_w-overlay_w-10:main_h-overlay_h-10
6070 Using named options the example above becomes:
6072 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
6076 Insert a transparent PNG logo in the bottom left corner of the input,
6077 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
6079 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
6083 Insert 2 different transparent PNG logos (second logo on bottom
6084 right corner) using the @command{ffmpeg} tool:
6086 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
6090 Add a transparent color layer on top of the main video, @code{WxH}
6091 must specify the size of the main input to the overlay filter:
6093 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
6097 Play an original video and a filtered version (here with the deshake
6098 filter) side by side using the @command{ffplay} tool:
6100 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
6103 The above command is the same as:
6105 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
6109 Make a sliding overlay appearing from the left to the right top part of the
6110 screen starting since time 2:
6112 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
6116 Compose output by putting two input videos side to side:
6118 ffmpeg -i left.avi -i right.avi -filter_complex "
6119 nullsrc=size=200x100 [background];
6120 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
6121 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
6122 [background][left] overlay=shortest=1 [background+left];
6123 [background+left][right] overlay=shortest=1:x=100 [left+right]
6128 mask 10-20 seconds of a video by applying the delogo filter to a section
6130 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
6131 -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]'
6136 Chain several overlays in cascade:
6138 nullsrc=s=200x200 [bg];
6139 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
6140 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
6141 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
6142 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
6143 [in3] null, [mid2] overlay=100:100 [out0]
6150 Apply Overcomplete Wavelet denoiser.
6152 The filter accepts the following options:
6158 Larger depth values will denoise lower frequency components more, but
6159 slow down filtering.
6161 Must be an int in the range 8-16, default is @code{8}.
6163 @item luma_strength, ls
6166 Must be a double value in the range 0-1000, default is @code{1.0}.
6168 @item chroma_strength, cs
6169 Set chroma strength.
6171 Must be a double value in the range 0-1000, default is @code{1.0}.
6176 Add paddings to the input image, and place the original input at the
6177 given coordinates @var{x}, @var{y}.
6179 This filter accepts the following parameters:
6184 Specify an expression for the size of the output image with the
6185 paddings added. If the value for @var{width} or @var{height} is 0, the
6186 corresponding input size is used for the output.
6188 The @var{width} expression can reference the value set by the
6189 @var{height} expression, and vice versa.
6191 The default value of @var{width} and @var{height} is 0.
6195 Specify an expression for the offsets where to place the input image
6196 in the padded area with respect to the top/left border of the output
6199 The @var{x} expression can reference the value set by the @var{y}
6200 expression, and vice versa.
6202 The default value of @var{x} and @var{y} is 0.
6205 Specify the color of the padded area. For the syntax of this option,
6206 check the "Color" section in the ffmpeg-utils manual.
6208 The default value of @var{color} is "black".
6211 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
6212 options are expressions containing the following constants:
6217 the input video width and height
6221 same as @var{in_w} and @var{in_h}
6225 the output width and height, that is the size of the padded area as
6226 specified by the @var{width} and @var{height} expressions
6230 same as @var{out_w} and @var{out_h}
6234 x and y offsets as specified by the @var{x} and @var{y}
6235 expressions, or NAN if not yet specified
6238 same as @var{iw} / @var{ih}
6241 input sample aspect ratio
6244 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
6248 horizontal and vertical chroma subsample values. For example for the
6249 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6252 @subsection Examples
6256 Add paddings with color "violet" to the input video. Output video
6257 size is 640x480, the top-left corner of the input video is placed at
6260 pad=640:480:0:40:violet
6263 The example above is equivalent to the following command:
6265 pad=width=640:height=480:x=0:y=40:color=violet
6269 Pad the input to get an output with dimensions increased by 3/2,
6270 and put the input video at the center of the padded area:
6272 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
6276 Pad the input to get a squared output with size equal to the maximum
6277 value between the input width and height, and put the input video at
6278 the center of the padded area:
6280 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
6284 Pad the input to get a final w/h ratio of 16:9:
6286 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
6290 In case of anamorphic video, in order to set the output display aspect
6291 correctly, it is necessary to use @var{sar} in the expression,
6292 according to the relation:
6294 (ih * X / ih) * sar = output_dar
6295 X = output_dar / sar
6298 Thus the previous example needs to be modified to:
6300 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
6304 Double output size and put the input video in the bottom-right
6305 corner of the output padded area:
6307 pad="2*iw:2*ih:ow-iw:oh-ih"
6311 @section perspective
6313 Correct perspective of video not recorded perpendicular to the screen.
6315 A description of the accepted parameters follows.
6326 Set coordinates expression for top left, top right, bottom left and bottom right corners.
6327 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
6329 The expressions can use the following variables:
6334 the width and height of video frame.
6338 Set interpolation for perspective correction.
6340 It accepts the following values:
6346 Default value is @samp{linear}.
6351 Delay interlaced video by one field time so that the field order changes.
6353 The intended use is to fix PAL movies that have been captured with the
6354 opposite field order to the film-to-video transfer.
6356 A description of the accepted parameters follows.
6362 It accepts the following values:
6365 Capture field order top-first, transfer bottom-first.
6366 Filter will delay the bottom field.
6369 Capture field order bottom-first, transfer top-first.
6370 Filter will delay the top field.
6373 Capture and transfer with the same field order. This mode only exists
6374 for the documentation of the other options to refer to, but if you
6375 actually select it, the filter will faithfully do nothing.
6378 Capture field order determined automatically by field flags, transfer
6380 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
6381 basis using field flags. If no field information is available,
6382 then this works just like @samp{u}.
6385 Capture unknown or varying, transfer opposite.
6386 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
6387 analyzing the images and selecting the alternative that produces best
6388 match between the fields.
6391 Capture top-first, transfer unknown or varying.
6392 Filter selects among @samp{t} and @samp{p} using image analysis.
6395 Capture bottom-first, transfer unknown or varying.
6396 Filter selects among @samp{b} and @samp{p} using image analysis.
6399 Capture determined by field flags, transfer unknown or varying.
6400 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
6401 image analysis. If no field information is available, then this works just
6402 like @samp{U}. This is the default mode.
6405 Both capture and transfer unknown or varying.
6406 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
6410 @section pixdesctest
6412 Pixel format descriptor test filter, mainly useful for internal
6413 testing. The output video should be equal to the input video.
6417 format=monow, pixdesctest
6420 can be used to test the monowhite pixel format descriptor definition.
6424 Enable the specified chain of postprocessing subfilters using libpostproc. This
6425 library should be automatically selected with a GPL build (@code{--enable-gpl}).
6426 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
6427 Each subfilter and some options have a short and a long name that can be used
6428 interchangeably, i.e. dr/dering are the same.
6430 The filters accept the following options:
6434 Set postprocessing subfilters string.
6437 All subfilters share common options to determine their scope:
6441 Honor the quality commands for this subfilter.
6444 Do chrominance filtering, too (default).
6447 Do luminance filtering only (no chrominance).
6450 Do chrominance filtering only (no luminance).
6453 These options can be appended after the subfilter name, separated by a '|'.
6455 Available subfilters are:
6458 @item hb/hdeblock[|difference[|flatness]]
6459 Horizontal deblocking filter
6462 Difference factor where higher values mean more deblocking (default: @code{32}).
6464 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6467 @item vb/vdeblock[|difference[|flatness]]
6468 Vertical deblocking filter
6471 Difference factor where higher values mean more deblocking (default: @code{32}).
6473 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6476 @item ha/hadeblock[|difference[|flatness]]
6477 Accurate horizontal deblocking filter
6480 Difference factor where higher values mean more deblocking (default: @code{32}).
6482 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6485 @item va/vadeblock[|difference[|flatness]]
6486 Accurate vertical deblocking filter
6489 Difference factor where higher values mean more deblocking (default: @code{32}).
6491 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6495 The horizontal and vertical deblocking filters share the difference and
6496 flatness values so you cannot set different horizontal and vertical
6501 Experimental horizontal deblocking filter
6504 Experimental vertical deblocking filter
6509 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
6512 larger -> stronger filtering
6514 larger -> stronger filtering
6516 larger -> stronger filtering
6519 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
6522 Stretch luminance to @code{0-255}.
6525 @item lb/linblenddeint
6526 Linear blend deinterlacing filter that deinterlaces the given block by
6527 filtering all lines with a @code{(1 2 1)} filter.
6529 @item li/linipoldeint
6530 Linear interpolating deinterlacing filter that deinterlaces the given block by
6531 linearly interpolating every second line.
6533 @item ci/cubicipoldeint
6534 Cubic interpolating deinterlacing filter deinterlaces the given block by
6535 cubically interpolating every second line.
6537 @item md/mediandeint
6538 Median deinterlacing filter that deinterlaces the given block by applying a
6539 median filter to every second line.
6541 @item fd/ffmpegdeint
6542 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
6543 second line with a @code{(-1 4 2 4 -1)} filter.
6546 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
6547 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
6549 @item fq/forceQuant[|quantizer]
6550 Overrides the quantizer table from the input with the constant quantizer you
6558 Default pp filter combination (@code{hb|a,vb|a,dr|a})
6561 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
6564 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
6567 @subsection Examples
6571 Apply horizontal and vertical deblocking, deringing and automatic
6572 brightness/contrast:
6578 Apply default filters without brightness/contrast correction:
6584 Apply default filters and temporal denoiser:
6586 pp=default/tmpnoise|1|2|3
6590 Apply deblocking on luminance only, and switch vertical deblocking on or off
6591 automatically depending on available CPU time:
6599 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
6600 Ratio) between two input videos.
6602 This filter takes in input two input videos, the first input is
6603 considered the "main" source and is passed unchanged to the
6604 output. The second input is used as a "reference" video for computing
6607 Both video inputs must have the same resolution and pixel format for
6608 this filter to work correctly. Also it assumes that both inputs
6609 have the same number of frames, which are compared one by one.
6611 The obtained average PSNR is printed through the logging system.
6613 The filter stores the accumulated MSE (mean squared error) of each
6614 frame, and at the end of the processing it is averaged across all frames
6615 equally, and the following formula is applied to obtain the PSNR:
6618 PSNR = 10*log10(MAX^2/MSE)
6621 Where MAX is the average of the maximum values of each component of the
6624 The description of the accepted parameters follows.
6628 If specified the filter will use the named file to save the PSNR of
6629 each individual frame.
6632 The file printed if @var{stats_file} is selected, contains a sequence of
6633 key/value pairs of the form @var{key}:@var{value} for each compared
6636 A description of each shown parameter follows:
6640 sequential number of the input frame, starting from 1
6643 Mean Square Error pixel-by-pixel average difference of the compared
6644 frames, averaged over all the image components.
6646 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
6647 Mean Square Error pixel-by-pixel average difference of the compared
6648 frames for the component specified by the suffix.
6650 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
6651 Peak Signal to Noise ratio of the compared frames for the component
6652 specified by the suffix.
6657 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
6658 [main][ref] psnr="stats_file=stats.log" [out]
6661 On this example the input file being processed is compared with the
6662 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
6663 is stored in @file{stats.log}.
6668 Pulldown reversal (inverse telecine) filter, capable of handling mixed
6669 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
6672 The pullup filter is designed to take advantage of future context in making
6673 its decisions. This filter is stateless in the sense that it does not lock
6674 onto a pattern to follow, but it instead looks forward to the following
6675 fields in order to identify matches and rebuild progressive frames.
6677 To produce content with an even framerate, insert the fps filter after
6678 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
6679 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
6681 The filter accepts the following options:
6688 These options set the amount of "junk" to ignore at the left, right, top, and
6689 bottom of the image, respectively. Left and right are in units of 8 pixels,
6690 while top and bottom are in units of 2 lines.
6691 The default is 8 pixels on each side.
6694 Set the strict breaks. Setting this option to 1 will reduce the chances of
6695 filter generating an occasional mismatched frame, but it may also cause an
6696 excessive number of frames to be dropped during high motion sequences.
6697 Conversely, setting it to -1 will make filter match fields more easily.
6698 This may help processing of video where there is slight blurring between
6699 the fields, but may also cause there to be interlaced frames in the output.
6700 Default value is @code{0}.
6703 Set the metric plane to use. It accepts the following values:
6709 Use chroma blue plane.
6712 Use chroma red plane.
6715 This option may be set to use chroma plane instead of the default luma plane
6716 for doing filter's computations. This may improve accuracy on very clean
6717 source material, but more likely will decrease accuracy, especially if there
6718 is chroma noise (rainbow effect) or any grayscale video.
6719 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
6720 load and make pullup usable in realtime on slow machines.
6723 For best results (without duplicated frames in the output file) it is
6724 necessary to change the output frame rate. For example, to inverse
6725 telecine NTSC input:
6727 ffmpeg -i input -vf pullup -r 24000/1001 ...
6732 Suppress a TV station logo, using an image file to determine which
6733 pixels comprise the logo. It works by filling in the pixels that
6734 comprise the logo with neighboring pixels.
6736 The filter accepts the following options:
6740 Set the filter bitmap file, which can be any image format supported by
6741 libavformat. The width and height of the image file must match those of the
6742 video stream being processed.
6745 Pixels in the provided bitmap image with a value of zero are not
6746 considered part of the logo, non-zero pixels are considered part of
6747 the logo. If you use white (255) for the logo and black (0) for the
6748 rest, you will be safe. For making the filter bitmap, it is
6749 recommended to take a screen capture of a black frame with the logo
6750 visible, and then using a threshold filter followed by the erode
6751 filter once or twice.
6753 If needed, little splotches can be fixed manually. Remember that if
6754 logo pixels are not covered, the filter quality will be much
6755 reduced. Marking too many pixels as part of the logo does not hurt as
6756 much, but it will increase the amount of blurring needed to cover over
6757 the image and will destroy more information than necessary, and extra
6758 pixels will slow things down on a large logo.
6762 Rotate video by an arbitrary angle expressed in radians.
6764 The filter accepts the following options:
6766 A description of the optional parameters follows.
6769 Set an expression for the angle by which to rotate the input video
6770 clockwise, expressed as a number of radians. A negative value will
6771 result in a counter-clockwise rotation. By default it is set to "0".
6773 This expression is evaluated for each frame.
6776 Set the output width expression, default value is "iw".
6777 This expression is evaluated just once during configuration.
6780 Set the output height expression, default value is "ih".
6781 This expression is evaluated just once during configuration.
6784 Enable bilinear interpolation if set to 1, a value of 0 disables
6785 it. Default value is 1.
6788 Set the color used to fill the output area not covered by the rotated
6789 image. For the generalsyntax of this option, check the "Color" section in the
6790 ffmpeg-utils manual. If the special value "none" is selected then no
6791 background is printed (useful for example if the background is never shown).
6793 Default value is "black".
6796 The expressions for the angle and the output size can contain the
6797 following constants and functions:
6801 sequential number of the input frame, starting from 0. It is always NAN
6802 before the first frame is filtered.
6805 time in seconds of the input frame, it is set to 0 when the filter is
6806 configured. It is always NAN before the first frame is filtered.
6810 horizontal and vertical chroma subsample values. For example for the
6811 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6815 the input video width and height
6819 the output width and height, that is the size of the padded area as
6820 specified by the @var{width} and @var{height} expressions
6824 the minimal width/height required for completely containing the input
6825 video rotated by @var{a} radians.
6827 These are only available when computing the @option{out_w} and
6828 @option{out_h} expressions.
6831 @subsection Examples
6835 Rotate the input by PI/6 radians clockwise:
6841 Rotate the input by PI/6 radians counter-clockwise:
6847 Rotate the input by 45 degrees clockwise:
6853 Apply a constant rotation with period T, starting from an angle of PI/3:
6855 rotate=PI/3+2*PI*t/T
6859 Make the input video rotation oscillating with a period of T
6860 seconds and an amplitude of A radians:
6862 rotate=A*sin(2*PI/T*t)
6866 Rotate the video, output size is choosen so that the whole rotating
6867 input video is always completely contained in the output:
6869 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
6873 Rotate the video, reduce the output size so that no background is ever
6876 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
6880 @subsection Commands
6882 The filter supports the following commands:
6886 Set the angle expression.
6887 The command accepts the same syntax of the corresponding option.
6889 If the specified expression is not valid, it is kept at its current
6895 Apply Shape Adaptive Blur.
6897 The filter accepts the following options:
6900 @item luma_radius, lr
6901 Set luma blur filter strength, must be a value in range 0.1-4.0, default
6902 value is 1.0. A greater value will result in a more blurred image, and
6903 in slower processing.
6905 @item luma_pre_filter_radius, lpfr
6906 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
6909 @item luma_strength, ls
6910 Set luma maximum difference between pixels to still be considered, must
6911 be a value in the 0.1-100.0 range, default value is 1.0.
6913 @item chroma_radius, cr
6914 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
6915 greater value will result in a more blurred image, and in slower
6918 @item chroma_pre_filter_radius, cpfr
6919 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
6921 @item chroma_strength, cs
6922 Set chroma maximum difference between pixels to still be considered,
6923 must be a value in the 0.1-100.0 range.
6926 Each chroma option value, if not explicitly specified, is set to the
6927 corresponding luma option value.
6932 Scale (resize) the input video, using the libswscale library.
6934 The scale filter forces the output display aspect ratio to be the same
6935 of the input, by changing the output sample aspect ratio.
6937 If the input image format is different from the format requested by
6938 the next filter, the scale filter will convert the input to the
6942 The filter accepts the following options, or any of the options
6943 supported by the libswscale scaler.
6945 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
6946 the complete list of scaler options.
6951 Set the output video dimension expression. Default value is the input
6954 If the value is 0, the input width is used for the output.
6956 If one of the values is -1, the scale filter will use a value that
6957 maintains the aspect ratio of the input image, calculated from the
6958 other specified dimension. If both of them are -1, the input size is
6961 If one of the values is -n with n > 1, the scale filter will also use a value
6962 that maintains the aspect ratio of the input image, calculated from the other
6963 specified dimension. After that it will, however, make sure that the calculated
6964 dimension is divisible by n and adjust the value if necessary.
6966 See below for the list of accepted constants for use in the dimension
6970 Set the interlacing mode. It accepts the following values:
6974 Force interlaced aware scaling.
6977 Do not apply interlaced scaling.
6980 Select interlaced aware scaling depending on whether the source frames
6981 are flagged as interlaced or not.
6984 Default value is @samp{0}.
6987 Set libswscale scaling flags. See
6988 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
6989 complete list of values. If not explictly specified the filter applies
6993 Set the video size. For the syntax of this option, check the "Video size"
6994 section in the ffmpeg-utils manual.
6996 @item in_color_matrix
6997 @item out_color_matrix
6998 Set in/output YCbCr color space type.
7000 This allows the autodetected value to be overridden as well as allows forcing
7001 a specific value used for the output and encoder.
7003 If not specified, the color space type depends on the pixel format.
7009 Choose automatically.
7012 Format conforming to International Telecommunication Union (ITU)
7013 Recommendation BT.709.
7016 Set color space conforming to the United States Federal Communications
7017 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
7020 Set color space conforming to:
7024 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
7027 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
7030 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
7035 Set color space conforming to SMPTE ST 240:1999.
7040 Set in/output YCbCr sample range.
7042 This allows the autodetected value to be overridden as well as allows forcing
7043 a specific value used for the output and encoder. If not specified, the
7044 range depends on the pixel format. Possible values:
7048 Choose automatically.
7051 Set full range (0-255 in case of 8-bit luma).
7054 Set "MPEG" range (16-235 in case of 8-bit luma).
7057 @item force_original_aspect_ratio
7058 Enable decreasing or increasing output video width or height if necessary to
7059 keep the original aspect ratio. Possible values:
7063 Scale the video as specified and disable this feature.
7066 The output video dimensions will automatically be decreased if needed.
7069 The output video dimensions will automatically be increased if needed.
7073 One useful instance of this option is that when you know a specific device's
7074 maximum allowed resolution, you can use this to limit the output video to
7075 that, while retaining the aspect ratio. For example, device A allows
7076 1280x720 playback, and your video is 1920x800. Using this option (set it to
7077 decrease) and specifying 1280x720 to the command line makes the output
7080 Please note that this is a different thing than specifying -1 for @option{w}
7081 or @option{h}, you still need to specify the output resolution for this option
7086 The values of the @option{w} and @option{h} options are expressions
7087 containing the following constants:
7092 the input width and height
7096 same as @var{in_w} and @var{in_h}
7100 the output (scaled) width and height
7104 same as @var{out_w} and @var{out_h}
7107 same as @var{iw} / @var{ih}
7110 input sample aspect ratio
7113 input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
7117 horizontal and vertical input chroma subsample values. For example for the
7118 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7122 horizontal and vertical output chroma subsample values. For example for the
7123 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7126 @subsection Examples
7130 Scale the input video to a size of 200x100:
7135 This is equivalent to:
7146 Specify a size abbreviation for the output size:
7151 which can also be written as:
7157 Scale the input to 2x:
7163 The above is the same as:
7169 Scale the input to 2x with forced interlaced scaling:
7171 scale=2*iw:2*ih:interl=1
7175 Scale the input to half size:
7181 Increase the width, and set the height to the same size:
7187 Seek for Greek harmony:
7194 Increase the height, and set the width to 3/2 of the height:
7196 scale=w=3/2*oh:h=3/5*ih
7200 Increase the size, but make the size a multiple of the chroma
7203 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
7207 Increase the width to a maximum of 500 pixels, keep the same input
7210 scale=w='min(500\, iw*3/2):h=-1'
7214 @section separatefields
7216 The @code{separatefields} takes a frame-based video input and splits
7217 each frame into its components fields, producing a new half height clip
7218 with twice the frame rate and twice the frame count.
7220 This filter use field-dominance information in frame to decide which
7221 of each pair of fields to place first in the output.
7222 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
7224 @section setdar, setsar
7226 The @code{setdar} filter sets the Display Aspect Ratio for the filter
7229 This is done by changing the specified Sample (aka Pixel) Aspect
7230 Ratio, according to the following equation:
7232 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
7235 Keep in mind that the @code{setdar} filter does not modify the pixel
7236 dimensions of the video frame. Also the display aspect ratio set by
7237 this filter may be changed by later filters in the filterchain,
7238 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
7241 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
7242 the filter output video.
7244 Note that as a consequence of the application of this filter, the
7245 output display aspect ratio will change according to the equation
7248 Keep in mind that the sample aspect ratio set by the @code{setsar}
7249 filter may be changed by later filters in the filterchain, e.g. if
7250 another "setsar" or a "setdar" filter is applied.
7252 The filters accept the following options:
7255 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
7256 Set the aspect ratio used by the filter.
7258 The parameter can be a floating point number string, an expression, or
7259 a string of the form @var{num}:@var{den}, where @var{num} and
7260 @var{den} are the numerator and denominator of the aspect ratio. If
7261 the parameter is not specified, it is assumed the value "0".
7262 In case the form "@var{num}:@var{den}" is used, the @code{:} character
7266 Set the maximum integer value to use for expressing numerator and
7267 denominator when reducing the expressed aspect ratio to a rational.
7268 Default value is @code{100}.
7272 The parameter @var{sar} is an expression containing
7273 the following constants:
7277 the corresponding mathematical approximated values for e
7278 (euler number), pi (greek PI), phi (golden ratio)
7281 the input width and height
7284 same as @var{w} / @var{h}
7287 input sample aspect ratio
7290 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
7293 horizontal and vertical chroma subsample values. For example for the
7294 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7297 @subsection Examples
7302 To change the display aspect ratio to 16:9, specify one of the following:
7310 To change the sample aspect ratio to 10:11, specify:
7316 To set a display aspect ratio of 16:9, and specify a maximum integer value of
7317 1000 in the aspect ratio reduction, use the command:
7319 setdar=ratio=16/9:max=1000
7327 Force field for the output video frame.
7329 The @code{setfield} filter marks the interlace type field for the
7330 output frames. It does not change the input frame, but only sets the
7331 corresponding property, which affects how the frame is treated by
7332 following filters (e.g. @code{fieldorder} or @code{yadif}).
7334 The filter accepts the following options:
7339 Available values are:
7343 Keep the same field property.
7346 Mark the frame as bottom-field-first.
7349 Mark the frame as top-field-first.
7352 Mark the frame as progressive.
7358 Show a line containing various information for each input video frame.
7359 The input video is not modified.
7361 The shown line contains a sequence of key/value pairs of the form
7362 @var{key}:@var{value}.
7364 A description of each shown parameter follows:
7368 sequential number of the input frame, starting from 0
7371 Presentation TimeStamp of the input frame, expressed as a number of
7372 time base units. The time base unit depends on the filter input pad.
7375 Presentation TimeStamp of the input frame, expressed as a number of
7379 position of the frame in the input stream, -1 if this information in
7380 unavailable and/or meaningless (for example in case of synthetic video)
7386 sample aspect ratio of the input frame, expressed in the form
7390 size of the input frame. For the syntax of this option, check the "Video size"
7391 section in the ffmpeg-utils manual.
7394 interlaced mode ("P" for "progressive", "T" for top field first, "B"
7395 for bottom field first)
7398 1 if the frame is a key frame, 0 otherwise
7401 picture type of the input frame ("I" for an I-frame, "P" for a
7402 P-frame, "B" for a B-frame, "?" for unknown type).
7403 Check also the documentation of the @code{AVPictureType} enum and of
7404 the @code{av_get_picture_type_char} function defined in
7405 @file{libavutil/avutil.h}.
7408 Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame
7410 @item plane_checksum
7411 Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
7412 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]"
7418 Blur the input video without impacting the outlines.
7420 The filter accepts the following options:
7423 @item luma_radius, lr
7424 Set the luma radius. The option value must be a float number in
7425 the range [0.1,5.0] that specifies the variance of the gaussian filter
7426 used to blur the image (slower if larger). Default value is 1.0.
7428 @item luma_strength, ls
7429 Set the luma strength. The option value must be a float number
7430 in the range [-1.0,1.0] that configures the blurring. A value included
7431 in [0.0,1.0] will blur the image whereas a value included in
7432 [-1.0,0.0] will sharpen the image. Default value is 1.0.
7434 @item luma_threshold, lt
7435 Set the luma threshold used as a coefficient to determine
7436 whether a pixel should be blurred or not. The option value must be an
7437 integer in the range [-30,30]. A value of 0 will filter all the image,
7438 a value included in [0,30] will filter flat areas and a value included
7439 in [-30,0] will filter edges. Default value is 0.
7441 @item chroma_radius, cr
7442 Set the chroma radius. The option value must be a float number in
7443 the range [0.1,5.0] that specifies the variance of the gaussian filter
7444 used to blur the image (slower if larger). Default value is 1.0.
7446 @item chroma_strength, cs
7447 Set the chroma strength. The option value must be a float number
7448 in the range [-1.0,1.0] that configures the blurring. A value included
7449 in [0.0,1.0] will blur the image whereas a value included in
7450 [-1.0,0.0] will sharpen the image. Default value is 1.0.
7452 @item chroma_threshold, ct
7453 Set the chroma threshold used as a coefficient to determine
7454 whether a pixel should be blurred or not. The option value must be an
7455 integer in the range [-30,30]. A value of 0 will filter all the image,
7456 a value included in [0,30] will filter flat areas and a value included
7457 in [-30,0] will filter edges. Default value is 0.
7460 If a chroma option is not explicitly set, the corresponding luma value
7465 Convert between different stereoscopic image formats.
7467 The filters accept the following options:
7471 Set stereoscopic image format of input.
7473 Available values for input image formats are:
7476 side by side parallel (left eye left, right eye right)
7479 side by side crosseye (right eye left, left eye right)
7482 side by side parallel with half width resolution
7483 (left eye left, right eye right)
7486 side by side crosseye with half width resolution
7487 (right eye left, left eye right)
7490 above-below (left eye above, right eye below)
7493 above-below (right eye above, left eye below)
7496 above-below with half height resolution
7497 (left eye above, right eye below)
7500 above-below with half height resolution
7501 (right eye above, left eye below)
7504 alternating frames (left eye first, right eye second)
7507 alternating frames (right eye first, left eye second)
7509 Default value is @samp{sbsl}.
7513 Set stereoscopic image format of output.
7515 Available values for output image formats are all the input formats as well as:
7518 anaglyph red/blue gray
7519 (red filter on left eye, blue filter on right eye)
7522 anaglyph red/green gray
7523 (red filter on left eye, green filter on right eye)
7526 anaglyph red/cyan gray
7527 (red filter on left eye, cyan filter on right eye)
7530 anaglyph red/cyan half colored
7531 (red filter on left eye, cyan filter on right eye)
7534 anaglyph red/cyan color
7535 (red filter on left eye, cyan filter on right eye)
7538 anaglyph red/cyan color optimized with the least squares projection of dubois
7539 (red filter on left eye, cyan filter on right eye)
7542 anaglyph green/magenta gray
7543 (green filter on left eye, magenta filter on right eye)
7546 anaglyph green/magenta half colored
7547 (green filter on left eye, magenta filter on right eye)
7550 anaglyph green/magenta colored
7551 (green filter on left eye, magenta filter on right eye)
7554 anaglyph green/magenta color optimized with the least squares projection of dubois
7555 (green filter on left eye, magenta filter on right eye)
7558 anaglyph yellow/blue gray
7559 (yellow filter on left eye, blue filter on right eye)
7562 anaglyph yellow/blue half colored
7563 (yellow filter on left eye, blue filter on right eye)
7566 anaglyph yellow/blue colored
7567 (yellow filter on left eye, blue filter on right eye)
7570 anaglyph yellow/blue color optimized with the least squares projection of dubois
7571 (yellow filter on left eye, blue filter on right eye)
7574 interleaved rows (left eye has top row, right eye starts on next row)
7577 interleaved rows (right eye has top row, left eye starts on next row)
7580 mono output (left eye only)
7583 mono output (right eye only)
7586 Default value is @samp{arcd}.
7589 @subsection Examples
7593 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
7599 Convert input video from above bellow (left eye above, right eye below) to side by side crosseye.
7607 Apply a simple postprocessing filter that compresses and decompresses the image
7608 at several (or - in the case of @option{quality} level @code{6} - all) shifts
7609 and average the results.
7611 The filter accepts the following options:
7615 Set quality. This option defines the number of levels for averaging. It accepts
7616 an integer in the range 0-6. If set to @code{0}, the filter will have no
7617 effect. A value of @code{6} means the higher quality. For each increment of
7618 that value the speed drops by a factor of approximately 2. Default value is
7622 Force a constant quantization parameter. If not set, the filter will use the QP
7623 from the video stream (if available).
7626 Set thresholding mode. Available modes are:
7630 Set hard thresholding (default).
7632 Set soft thresholding (better de-ringing effect, but likely blurrier).
7636 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
7637 option may cause flicker since the B-Frames have often larger QP. Default is
7638 @code{0} (not enabled).
7644 Draw subtitles on top of input video using the libass library.
7646 To enable compilation of this filter you need to configure FFmpeg with
7647 @code{--enable-libass}. This filter also requires a build with libavcodec and
7648 libavformat to convert the passed subtitles file to ASS (Advanced Substation
7649 Alpha) subtitles format.
7651 The filter accepts the following options:
7655 Set the filename of the subtitle file to read. It must be specified.
7658 Specify the size of the original video, the video for which the ASS file
7659 was composed. For the syntax of this option, check the "Video size" section in
7660 the ffmpeg-utils manual. Due to a misdesign in ASS aspect ratio arithmetic,
7661 this is necessary to correctly scale the fonts if the aspect ratio has been
7665 Set subtitles input character encoding. @code{subtitles} filter only. Only
7666 useful if not UTF-8.
7669 If the first key is not specified, it is assumed that the first value
7670 specifies the @option{filename}.
7672 For example, to render the file @file{sub.srt} on top of the input
7673 video, use the command:
7678 which is equivalent to:
7680 subtitles=filename=sub.srt
7685 Scale the input by 2x and smooth using the Super2xSaI (Scale and
7686 Interpolate) pixel art scaling algorithm.
7688 Useful for enlarging pixel art images without reducing sharpness.
7695 Apply telecine process to the video.
7697 This filter accepts the following options:
7706 The default value is @code{top}.
7710 A string of numbers representing the pulldown pattern you wish to apply.
7711 The default value is @code{23}.
7715 Some typical patterns:
7720 24p: 2332 (preferred)
7727 24p: 222222222223 ("Euro pulldown")
7733 Select the most representative frame in a given sequence of consecutive frames.
7735 The filter accepts the following options:
7739 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
7740 will pick one of them, and then handle the next batch of @var{n} frames until
7741 the end. Default is @code{100}.
7744 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
7745 value will result in a higher memory usage, so a high value is not recommended.
7747 @subsection Examples
7751 Extract one picture each 50 frames:
7757 Complete example of a thumbnail creation with @command{ffmpeg}:
7759 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
7765 Tile several successive frames together.
7767 The filter accepts the following options:
7772 Set the grid size (i.e. the number of lines and columns). For the syntax of
7773 this option, check the "Video size" section in the ffmpeg-utils manual.
7776 Set the maximum number of frames to render in the given area. It must be less
7777 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
7778 the area will be used.
7781 Set the outer border margin in pixels.
7784 Set the inner border thickness (i.e. the number of pixels between frames). For
7785 more advanced padding options (such as having different values for the edges),
7786 refer to the pad video filter.
7789 Specify the color of the unused areaFor the syntax of this option, check the
7790 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
7794 @subsection Examples
7798 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
7800 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
7802 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
7803 duplicating each output frame to accomodate the originally detected frame
7807 Display @code{5} pictures in an area of @code{3x2} frames,
7808 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
7809 mixed flat and named options:
7811 tile=3x2:nb_frames=5:padding=7:margin=2
7817 Perform various types of temporal field interlacing.
7819 Frames are counted starting from 1, so the first input frame is
7822 The filter accepts the following options:
7827 Specify the mode of the interlacing. This option can also be specified
7828 as a value alone. See below for a list of values for this option.
7830 Available values are:
7834 Move odd frames into the upper field, even into the lower field,
7835 generating a double height frame at half frame rate.
7838 Only output even frames, odd frames are dropped, generating a frame with
7839 unchanged height at half frame rate.
7842 Only output odd frames, even frames are dropped, generating a frame with
7843 unchanged height at half frame rate.
7846 Expand each frame to full height, but pad alternate lines with black,
7847 generating a frame with double height at the same input frame rate.
7849 @item interleave_top, 4
7850 Interleave the upper field from odd frames with the lower field from
7851 even frames, generating a frame with unchanged height at half frame rate.
7853 @item interleave_bottom, 5
7854 Interleave the lower field from odd frames with the upper field from
7855 even frames, generating a frame with unchanged height at half frame rate.
7857 @item interlacex2, 6
7858 Double frame rate with unchanged height. Frames are inserted each
7859 containing the second temporal field from the previous input frame and
7860 the first temporal field from the next input frame. This mode relies on
7861 the top_field_first flag. Useful for interlaced video displays with no
7862 field synchronisation.
7865 Numeric values are deprecated but are accepted for backward
7866 compatibility reasons.
7868 Default mode is @code{merge}.
7871 Specify flags influencing the filter process.
7873 Available value for @var{flags} is:
7876 @item low_pass_filter, vlfp
7877 Enable vertical low-pass filtering in the filter.
7878 Vertical low-pass filtering is required when creating an interlaced
7879 destination from a progressive source which contains high-frequency
7880 vertical detail. Filtering will reduce interlace 'twitter' and Moire
7883 Vertical low-pass filtering can only be enabled for @option{mode}
7884 @var{interleave_top} and @var{interleave_bottom}.
7891 Transpose rows with columns in the input video and optionally flip it.
7893 This filter accepts the following options:
7898 Specify the transposition direction.
7900 Can assume the following values:
7902 @item 0, 4, cclock_flip
7903 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
7911 Rotate by 90 degrees clockwise, that is:
7919 Rotate by 90 degrees counterclockwise, that is:
7926 @item 3, 7, clock_flip
7927 Rotate by 90 degrees clockwise and vertically flip, that is:
7935 For values between 4-7, the transposition is only done if the input
7936 video geometry is portrait and not landscape. These values are
7937 deprecated, the @code{passthrough} option should be used instead.
7939 Numerical values are deprecated, and should be dropped in favor of
7943 Do not apply the transposition if the input geometry matches the one
7944 specified by the specified value. It accepts the following values:
7947 Always apply transposition.
7949 Preserve portrait geometry (when @var{height} >= @var{width}).
7951 Preserve landscape geometry (when @var{width} >= @var{height}).
7954 Default value is @code{none}.
7957 For example to rotate by 90 degrees clockwise and preserve portrait
7960 transpose=dir=1:passthrough=portrait
7963 The command above can also be specified as:
7965 transpose=1:portrait
7969 Trim the input so that the output contains one continuous subpart of the input.
7971 This filter accepts the following options:
7974 Specify time of the start of the kept section, i.e. the frame with the
7975 timestamp @var{start} will be the first frame in the output.
7978 Specify time of the first frame that will be dropped, i.e. the frame
7979 immediately preceding the one with the timestamp @var{end} will be the last
7980 frame in the output.
7983 Same as @var{start}, except this option sets the start timestamp in timebase
7984 units instead of seconds.
7987 Same as @var{end}, except this option sets the end timestamp in timebase units
7991 Specify maximum duration of the output.
7994 Number of the first frame that should be passed to output.
7997 Number of the first frame that should be dropped.
8000 @option{start}, @option{end}, @option{duration} are expressed as time
8001 duration specifications, check the "Time duration" section in the
8002 ffmpeg-utils manual.
8004 Note that the first two sets of the start/end options and the @option{duration}
8005 option look at the frame timestamp, while the _frame variants simply count the
8006 frames that pass through the filter. Also note that this filter does not modify
8007 the timestamps. If you wish that the output timestamps start at zero, insert a
8008 setpts filter after the trim filter.
8010 If multiple start or end options are set, this filter tries to be greedy and
8011 keep all the frames that match at least one of the specified constraints. To keep
8012 only the part that matches all the constraints at once, chain multiple trim
8015 The defaults are such that all the input is kept. So it is possible to set e.g.
8016 just the end values to keep everything before the specified time.
8021 drop everything except the second minute of input
8023 ffmpeg -i INPUT -vf trim=60:120
8027 keep only the first second
8029 ffmpeg -i INPUT -vf trim=duration=1
8037 Sharpen or blur the input video.
8039 It accepts the following parameters:
8042 @item luma_msize_x, lx
8043 Set the luma matrix horizontal size. It must be an odd integer between
8044 3 and 63, default value is 5.
8046 @item luma_msize_y, ly
8047 Set the luma matrix vertical size. It must be an odd integer between 3
8048 and 63, default value is 5.
8050 @item luma_amount, la
8051 Set the luma effect strength. It can be a float number, reasonable
8052 values lay between -1.5 and 1.5.
8054 Negative values will blur the input video, while positive values will
8055 sharpen it, a value of zero will disable the effect.
8057 Default value is 1.0.
8059 @item chroma_msize_x, cx
8060 Set the chroma matrix horizontal size. It must be an odd integer
8061 between 3 and 63, default value is 5.
8063 @item chroma_msize_y, cy
8064 Set the chroma matrix vertical size. It must be an odd integer
8065 between 3 and 63, default value is 5.
8067 @item chroma_amount, ca
8068 Set the chroma effect strength. It can be a float number, reasonable
8069 values lay between -1.5 and 1.5.
8071 Negative values will blur the input video, while positive values will
8072 sharpen it, a value of zero will disable the effect.
8074 Default value is 0.0.
8077 If set to 1, specify using OpenCL capabilities, only available if
8078 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
8082 All parameters are optional and default to the equivalent of the
8083 string '5:5:1.0:5:5:0.0'.
8085 @subsection Examples
8089 Apply strong luma sharpen effect:
8091 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
8095 Apply strong blur of both luma and chroma parameters:
8097 unsharp=7:7:-2:7:7:-2
8101 @anchor{vidstabdetect}
8102 @section vidstabdetect
8104 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
8105 @ref{vidstabtransform} for pass 2.
8107 This filter generates a file with relative translation and rotation
8108 transform information about subsequent frames, which is then used by
8109 the @ref{vidstabtransform} filter.
8111 To enable compilation of this filter you need to configure FFmpeg with
8112 @code{--enable-libvidstab}.
8114 This filter accepts the following options:
8118 Set the path to the file used to write the transforms information.
8119 Default value is @file{transforms.trf}.
8122 Set how shaky the video is and how quick the camera is. It accepts an
8123 integer in the range 1-10, a value of 1 means little shakiness, a
8124 value of 10 means strong shakiness. Default value is 5.
8127 Set the accuracy of the detection process. It must be a value in the
8128 range 1-15. A value of 1 means low accuracy, a value of 15 means high
8129 accuracy. Default value is 15.
8132 Set stepsize of the search process. The region around minimum is
8133 scanned with 1 pixel resolution. Default value is 6.
8136 Set minimum contrast. Below this value a local measurement field is
8137 discarded. Must be a floating point value in the range 0-1. Default
8141 Set reference frame number for tripod mode.
8143 If enabled, the motion of the frames is compared to a reference frame
8144 in the filtered stream, identified by the specified number. The idea
8145 is to compensate all movements in a more-or-less static scene and keep
8146 the camera view absolutely still.
8148 If set to 0, it is disabled. The frames are counted starting from 1.
8151 Show fields and transforms in the resulting frames. It accepts an
8152 integer in the range 0-2. Default value is 0, which disables any
8156 @subsection Examples
8166 Analyze strongly shaky movie and put the results in file
8167 @file{mytransforms.trf}:
8169 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
8173 Visualize the result of internal transformations in the resulting
8176 vidstabdetect=show=1
8180 Analyze a video with medium shakiness using @command{ffmpeg}:
8182 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
8186 @anchor{vidstabtransform}
8187 @section vidstabtransform
8189 Video stabilization/deshaking: pass 2 of 2,
8190 see @ref{vidstabdetect} for pass 1.
8192 Read a file with transform information for each frame and
8193 apply/compensate them. Together with the @ref{vidstabdetect}
8194 filter this can be used to deshake videos. See also
8195 @url{http://public.hronopik.de/vid.stab}. It is important to also use
8196 the unsharp filter, see below.
8198 To enable compilation of this filter you need to configure FFmpeg with
8199 @code{--enable-libvidstab}.
8205 Set path to the file used to read the transforms. Default value is
8206 @file{transforms.trf}).
8209 Set the number of frames (value*2 + 1) used for lowpass filtering the
8210 camera movements. Default value is 10.
8212 For example a number of 10 means that 21 frames are used (10 in the
8213 past and 10 in the future) to smoothen the motion in the video. A
8214 larger values leads to a smoother video, but limits the acceleration
8215 of the camera (pan/tilt movements). 0 is a special case where a
8216 static camera is simulated.
8219 Set the camera path optimization algorithm.
8221 Accepted values are:
8224 gaussian kernel low-pass filter on camera motion (default)
8226 averaging on transformations
8230 Set maximal number of pixels to translate frames. Default value is -1,
8234 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
8235 value is -1, meaning no limit.
8238 Specify how to deal with borders that may be visible due to movement
8241 Available values are:
8244 keep image information from previous frame (default)
8246 fill the border black
8250 Invert transforms if set to 1. Default value is 0.
8253 Consider transforms as relative to previsou frame if set to 1,
8254 absolute if set to 0. Default value is 0.
8257 Set percentage to zoom. A positive value will result in a zoom-in
8258 effect, a negative value in a zoom-out effect. Default value is 0 (no
8262 Set optimal zooming to avoid borders.
8264 Accepted values are:
8269 optimal static zoom value is determined (only very strong movements
8270 will lead to visible borders) (default)
8272 optimal adaptive zoom value is determined (no borders will be
8273 visible), see @option{zoomspeed}
8276 Note that the value given at zoom is added to the one calculated here.
8279 Set percent to zoom maximally each frame (enabled when
8280 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
8284 Specify type of interpolation.
8286 Available values are:
8291 linear only horizontal
8293 linear in both directions (default)
8295 cubic in both directions (slow)
8299 Enable virtual tripod mode if set to 1, which is equivalent to
8300 @code{relative=0:smoothing=0}. Default value is 0.
8302 Use also @code{tripod} option of @ref{vidstabdetect}.
8305 Increase log verbosity if set to 1. Also the detected global motions
8306 are written to the temporary file @file{global_motions.trf}. Default
8310 @subsection Examples
8314 Use @command{ffmpeg} for a typical stabilization with default values:
8316 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
8319 Note the use of the unsharp filter which is always recommended.
8322 Zoom in a bit more and load transform data from a given file:
8324 vidstabtransform=zoom=5:input="mytransforms.trf"
8328 Smoothen the video even more:
8330 vidstabtransform=smoothing=30
8336 Flip the input video vertically.
8338 For example, to vertically flip a video with @command{ffmpeg}:
8340 ffmpeg -i in.avi -vf "vflip" out.avi
8345 Make or reverse a natural vignetting effect.
8347 The filter accepts the following options:
8351 Set lens angle expression as a number of radians.
8353 The value is clipped in the @code{[0,PI/2]} range.
8355 Default value: @code{"PI/5"}
8359 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
8363 Set forward/backward mode.
8365 Available modes are:
8368 The larger the distance from the central point, the darker the image becomes.
8371 The larger the distance from the central point, the brighter the image becomes.
8372 This can be used to reverse a vignette effect, though there is no automatic
8373 detection to extract the lens @option{angle} and other settings (yet). It can
8374 also be used to create a burning effect.
8377 Default value is @samp{forward}.
8380 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
8382 It accepts the following values:
8385 Evaluate expressions only once during the filter initialization.
8388 Evaluate expressions for each incoming frame. This is way slower than the
8389 @samp{init} mode since it requires all the scalers to be re-computed, but it
8390 allows advanced dynamic expressions.
8393 Default value is @samp{init}.
8396 Set dithering to reduce the circular banding effects. Default is @code{1}
8400 Set vignette aspect. This setting allows to adjust the shape of the vignette.
8401 Setting this value to the SAR of the input will make a rectangular vignetting
8402 following the dimensions of the video.
8404 Default is @code{1/1}.
8407 @subsection Expressions
8409 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
8410 following parameters.
8415 input width and height
8418 the number of input frame, starting from 0
8421 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
8422 @var{TB} units, NAN if undefined
8425 frame rate of the input video, NAN if the input frame rate is unknown
8428 the PTS (Presentation TimeStamp) of the filtered video frame,
8429 expressed in seconds, NAN if undefined
8432 time base of the input video
8436 @subsection Examples
8440 Apply simple strong vignetting effect:
8446 Make a flickering vignetting:
8448 vignette='PI/4+random(1)*PI/50':eval=frame
8455 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
8456 Deinterlacing Filter").
8458 Based on the process described by Martin Weston for BBC R&D, and
8459 implemented based on the de-interlace algorithm written by Jim
8460 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
8461 uses filter coefficients calculated by BBC R&D.
8463 There are two sets of filter coefficients, so called "simple":
8464 and "complex". Which set of filter coefficients is used can
8465 be set by passing an optional parameter:
8469 Set the interlacing filter coefficients. Accepts one of the following values:
8473 Simple filter coefficient set.
8475 More-complex filter coefficient set.
8477 Default value is @samp{complex}.
8480 Specify which frames to deinterlace. Accept one of the following values:
8484 Deinterlace all frames,
8486 Only deinterlace frames marked as interlaced.
8489 Default value is @samp{all}.
8495 Deinterlace the input video ("yadif" means "yet another deinterlacing
8498 This filter accepts the following options:
8504 The interlacing mode to adopt, accepts one of the following values:
8508 output 1 frame for each frame
8510 output 1 frame for each field
8511 @item 2, send_frame_nospatial
8512 like @code{send_frame} but skip spatial interlacing check
8513 @item 3, send_field_nospatial
8514 like @code{send_field} but skip spatial interlacing check
8517 Default value is @code{send_frame}.
8520 The picture field parity assumed for the input interlaced video, accepts one of
8521 the following values:
8525 assume top field first
8527 assume bottom field first
8529 enable automatic detection
8532 Default value is @code{auto}.
8533 If interlacing is unknown or decoder does not export this information,
8534 top field first will be assumed.
8537 Specify which frames to deinterlace. Accept one of the following
8542 deinterlace all frames
8544 only deinterlace frames marked as interlaced
8547 Default value is @code{all}.
8550 @c man end VIDEO FILTERS
8552 @chapter Video Sources
8553 @c man begin VIDEO SOURCES
8555 Below is a description of the currently available video sources.
8559 Buffer video frames, and make them available to the filter chain.
8561 This source is mainly intended for a programmatic use, in particular
8562 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
8564 This source accepts the following options:
8569 Specify the size (width and height) of the buffered video frames. For the
8570 syntax of this option, check the "Video size" section in the ffmpeg-utils
8580 A string representing the pixel format of the buffered video frames.
8581 It may be a number corresponding to a pixel format, or a pixel format
8585 Specify the timebase assumed by the timestamps of the buffered frames.
8588 Specify the frame rate expected for the video stream.
8590 @item pixel_aspect, sar
8591 Specify the sample aspect ratio assumed by the video frames.
8594 Specify the optional parameters to be used for the scale filter which
8595 is automatically inserted when an input change is detected in the
8596 input size or format.
8601 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
8604 will instruct the source to accept video frames with size 320x240 and
8605 with format "yuv410p", assuming 1/24 as the timestamps timebase and
8606 square pixels (1:1 sample aspect ratio).
8607 Since the pixel format with name "yuv410p" corresponds to the number 6
8608 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
8609 this example corresponds to:
8611 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
8614 Alternatively, the options can be specified as a flat string, but this
8615 syntax is deprecated:
8617 @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}]
8621 Create a pattern generated by an elementary cellular automaton.
8623 The initial state of the cellular automaton can be defined through the
8624 @option{filename}, and @option{pattern} options. If such options are
8625 not specified an initial state is created randomly.
8627 At each new frame a new row in the video is filled with the result of
8628 the cellular automaton next generation. The behavior when the whole
8629 frame is filled is defined by the @option{scroll} option.
8631 This source accepts the following options:
8635 Read the initial cellular automaton state, i.e. the starting row, from
8637 In the file, each non-whitespace character is considered an alive
8638 cell, a newline will terminate the row, and further characters in the
8639 file will be ignored.
8642 Read the initial cellular automaton state, i.e. the starting row, from
8643 the specified string.
8645 Each non-whitespace character in the string is considered an alive
8646 cell, a newline will terminate the row, and further characters in the
8647 string will be ignored.
8650 Set the video rate, that is the number of frames generated per second.
8653 @item random_fill_ratio, ratio
8654 Set the random fill ratio for the initial cellular automaton row. It
8655 is a floating point number value ranging from 0 to 1, defaults to
8658 This option is ignored when a file or a pattern is specified.
8660 @item random_seed, seed
8661 Set the seed for filling randomly the initial row, must be an integer
8662 included between 0 and UINT32_MAX. If not specified, or if explicitly
8663 set to -1, the filter will try to use a good random seed on a best
8667 Set the cellular automaton rule, it is a number ranging from 0 to 255.
8668 Default value is 110.
8671 Set the size of the output video. For the syntax of this option, check
8672 the "Video size" section in the ffmpeg-utils manual.
8674 If @option{filename} or @option{pattern} is specified, the size is set
8675 by default to the width of the specified initial state row, and the
8676 height is set to @var{width} * PHI.
8678 If @option{size} is set, it must contain the width of the specified
8679 pattern string, and the specified pattern will be centered in the
8682 If a filename or a pattern string is not specified, the size value
8683 defaults to "320x518" (used for a randomly generated initial state).
8686 If set to 1, scroll the output upward when all the rows in the output
8687 have been already filled. If set to 0, the new generated row will be
8688 written over the top row just after the bottom row is filled.
8691 @item start_full, full
8692 If set to 1, completely fill the output with generated rows before
8693 outputting the first frame.
8694 This is the default behavior, for disabling set the value to 0.
8697 If set to 1, stitch the left and right row edges together.
8698 This is the default behavior, for disabling set the value to 0.
8701 @subsection Examples
8705 Read the initial state from @file{pattern}, and specify an output of
8708 cellauto=f=pattern:s=200x400
8712 Generate a random initial row with a width of 200 cells, with a fill
8715 cellauto=ratio=2/3:s=200x200
8719 Create a pattern generated by rule 18 starting by a single alive cell
8720 centered on an initial row with width 100:
8722 cellauto=p=@@:s=100x400:full=0:rule=18
8726 Specify a more elaborated initial pattern:
8728 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
8735 Generate a Mandelbrot set fractal, and progressively zoom towards the
8736 point specified with @var{start_x} and @var{start_y}.
8738 This source accepts the following options:
8743 Set the terminal pts value. Default value is 400.
8746 Set the terminal scale value.
8747 Must be a floating point value. Default value is 0.3.
8750 Set the inner coloring mode, that is the algorithm used to draw the
8751 Mandelbrot fractal internal region.
8753 It shall assume one of the following values:
8758 Show time until convergence.
8760 Set color based on point closest to the origin of the iterations.
8765 Default value is @var{mincol}.
8768 Set the bailout value. Default value is 10.0.
8771 Set the maximum of iterations performed by the rendering
8772 algorithm. Default value is 7189.
8775 Set outer coloring mode.
8776 It shall assume one of following values:
8778 @item iteration_count
8779 Set iteration cound mode.
8780 @item normalized_iteration_count
8781 set normalized iteration count mode.
8783 Default value is @var{normalized_iteration_count}.
8786 Set frame rate, expressed as number of frames per second. Default
8790 Set frame size. For the syntax of this option, check the "Video
8791 size" section in the ffmpeg-utils manual. Default value is "640x480".
8794 Set the initial scale value. Default value is 3.0.
8797 Set the initial x position. Must be a floating point value between
8798 -100 and 100. Default value is -0.743643887037158704752191506114774.
8801 Set the initial y position. Must be a floating point value between
8802 -100 and 100. Default value is -0.131825904205311970493132056385139.
8807 Generate various test patterns, as generated by the MPlayer test filter.
8809 The size of the generated video is fixed, and is 256x256.
8810 This source is useful in particular for testing encoding features.
8812 This source accepts the following options:
8817 Specify the frame rate of the sourced video, as the number of frames
8818 generated per second. It has to be a string in the format
8819 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
8820 number or a valid video frame rate abbreviation. The default value is
8824 Set the video duration of the sourced video. The accepted syntax is:
8829 See also the function @code{av_parse_time()}.
8831 If not specified, or the expressed duration is negative, the video is
8832 supposed to be generated forever.
8836 Set the number or the name of the test to perform. Supported tests are:
8851 Default value is "all", which will cycle through the list of all tests.
8854 For example the following:
8859 will generate a "dc_luma" test pattern.
8863 Provide a frei0r source.
8865 To enable compilation of this filter you need to install the frei0r
8866 header and configure FFmpeg with @code{--enable-frei0r}.
8868 This source accepts the following options:
8873 The size of the video to generate. For the syntax of this option, check the
8874 "Video size" section in the ffmpeg-utils manual.
8877 Framerate of the generated video, may be a string of the form
8878 @var{num}/@var{den} or a frame rate abbreviation.
8881 The name to the frei0r source to load. For more information regarding frei0r and
8882 how to set the parameters read the section @ref{frei0r} in the description of
8886 A '|'-separated list of parameters to pass to the frei0r source.
8890 For example, to generate a frei0r partik0l source with size 200x200
8891 and frame rate 10 which is overlayed on the overlay filter main input:
8893 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
8898 Generate a life pattern.
8900 This source is based on a generalization of John Conway's life game.
8902 The sourced input represents a life grid, each pixel represents a cell
8903 which can be in one of two possible states, alive or dead. Every cell
8904 interacts with its eight neighbours, which are the cells that are
8905 horizontally, vertically, or diagonally adjacent.
8907 At each interaction the grid evolves according to the adopted rule,
8908 which specifies the number of neighbor alive cells which will make a
8909 cell stay alive or born. The @option{rule} option allows to specify
8912 This source accepts the following options:
8916 Set the file from which to read the initial grid state. In the file,
8917 each non-whitespace character is considered an alive cell, and newline
8918 is used to delimit the end of each row.
8920 If this option is not specified, the initial grid is generated
8924 Set the video rate, that is the number of frames generated per second.
8927 @item random_fill_ratio, ratio
8928 Set the random fill ratio for the initial random grid. It is a
8929 floating point number value ranging from 0 to 1, defaults to 1/PHI.
8930 It is ignored when a file is specified.
8932 @item random_seed, seed
8933 Set the seed for filling the initial random grid, must be an integer
8934 included between 0 and UINT32_MAX. If not specified, or if explicitly
8935 set to -1, the filter will try to use a good random seed on a best
8941 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
8942 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
8943 @var{NS} specifies the number of alive neighbor cells which make a
8944 live cell stay alive, and @var{NB} the number of alive neighbor cells
8945 which make a dead cell to become alive (i.e. to "born").
8946 "s" and "b" can be used in place of "S" and "B", respectively.
8948 Alternatively a rule can be specified by an 18-bits integer. The 9
8949 high order bits are used to encode the next cell state if it is alive
8950 for each number of neighbor alive cells, the low order bits specify
8951 the rule for "borning" new cells. Higher order bits encode for an
8952 higher number of neighbor cells.
8953 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
8954 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
8956 Default value is "S23/B3", which is the original Conway's game of life
8957 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
8958 cells, and will born a new cell if there are three alive cells around
8962 Set the size of the output video. For the syntax of this option, check the
8963 "Video size" section in the ffmpeg-utils manual.
8965 If @option{filename} is specified, the size is set by default to the
8966 same size of the input file. If @option{size} is set, it must contain
8967 the size specified in the input file, and the initial grid defined in
8968 that file is centered in the larger resulting area.
8970 If a filename is not specified, the size value defaults to "320x240"
8971 (used for a randomly generated initial grid).
8974 If set to 1, stitch the left and right grid edges together, and the
8975 top and bottom edges also. Defaults to 1.
8978 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
8979 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
8980 value from 0 to 255.
8983 Set the color of living (or new born) cells.
8986 Set the color of dead cells. If @option{mold} is set, this is the first color
8987 used to represent a dead cell.
8990 Set mold color, for definitely dead and moldy cells.
8992 For the syntax of these 3 color options, check the "Color" section in the
8993 ffmpeg-utils manual.
8996 @subsection Examples
9000 Read a grid from @file{pattern}, and center it on a grid of size
9003 life=f=pattern:s=300x300
9007 Generate a random grid of size 200x200, with a fill ratio of 2/3:
9009 life=ratio=2/3:s=200x200
9013 Specify a custom rule for evolving a randomly generated grid:
9019 Full example with slow death effect (mold) using @command{ffplay}:
9021 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
9026 @anchor{haldclutsrc}
9030 @anchor{smptehdbars}
9032 @section color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc
9034 The @code{color} source provides an uniformly colored input.
9036 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
9037 @ref{haldclut} filter.
9039 The @code{nullsrc} source returns unprocessed video frames. It is
9040 mainly useful to be employed in analysis / debugging tools, or as the
9041 source for filters which ignore the input data.
9043 The @code{rgbtestsrc} source generates an RGB test pattern useful for
9044 detecting RGB vs BGR issues. You should see a red, green and blue
9045 stripe from top to bottom.
9047 The @code{smptebars} source generates a color bars pattern, based on
9048 the SMPTE Engineering Guideline EG 1-1990.
9050 The @code{smptehdbars} source generates a color bars pattern, based on
9051 the SMPTE RP 219-2002.
9053 The @code{testsrc} source generates a test video pattern, showing a
9054 color pattern, a scrolling gradient and a timestamp. This is mainly
9055 intended for testing purposes.
9057 The sources accept the following options:
9062 Specify the color of the source, only available in the @code{color}
9063 source. For the syntax of this option, check the "Color" section in the
9064 ffmpeg-utils manual.
9067 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
9068 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
9069 pixels to be used as identity matrix for 3D lookup tables. Each component is
9070 coded on a @code{1/(N*N)} scale.
9073 Specify the size of the sourced video. For the syntax of this option, check the
9074 "Video size" section in the ffmpeg-utils manual. The default value is
9077 This option is not available with the @code{haldclutsrc} filter.
9080 Specify the frame rate of the sourced video, as the number of frames
9081 generated per second. It has to be a string in the format
9082 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
9083 number or a valid video frame rate abbreviation. The default value is
9087 Set the sample aspect ratio of the sourced video.
9090 Set the video duration of the sourced video. The accepted syntax is:
9092 [-]HH[:MM[:SS[.m...]]]
9095 See also the function @code{av_parse_time()}.
9097 If not specified, or the expressed duration is negative, the video is
9098 supposed to be generated forever.
9101 Set the number of decimals to show in the timestamp, only available in the
9102 @code{testsrc} source.
9104 The displayed timestamp value will correspond to the original
9105 timestamp value multiplied by the power of 10 of the specified
9106 value. Default value is 0.
9109 For example the following:
9111 testsrc=duration=5.3:size=qcif:rate=10
9114 will generate a video with a duration of 5.3 seconds, with size
9115 176x144 and a frame rate of 10 frames per second.
9117 The following graph description will generate a red source
9118 with an opacity of 0.2, with size "qcif" and a frame rate of 10
9121 color=c=red@@0.2:s=qcif:r=10
9124 If the input content is to be ignored, @code{nullsrc} can be used. The
9125 following command generates noise in the luminance plane by employing
9126 the @code{geq} filter:
9128 nullsrc=s=256x256, geq=random(1)*255:128:128
9131 @subsection Commands
9133 The @code{color} source supports the following commands:
9137 Set the color of the created image. Accepts the same syntax of the
9138 corresponding @option{color} option.
9141 @c man end VIDEO SOURCES
9143 @chapter Video Sinks
9144 @c man begin VIDEO SINKS
9146 Below is a description of the currently available video sinks.
9150 Buffer video frames, and make them available to the end of the filter
9153 This sink is mainly intended for a programmatic use, in particular
9154 through the interface defined in @file{libavfilter/buffersink.h}
9155 or the options system.
9157 It accepts a pointer to an AVBufferSinkContext structure, which
9158 defines the incoming buffers' formats, to be passed as the opaque
9159 parameter to @code{avfilter_init_filter} for initialization.
9163 Null video sink, do absolutely nothing with the input video. It is
9164 mainly useful as a template and to be employed in analysis / debugging
9167 @c man end VIDEO SINKS
9169 @chapter Multimedia Filters
9170 @c man begin MULTIMEDIA FILTERS
9172 Below is a description of the currently available multimedia filters.
9174 @section avectorscope
9176 Convert input audio to a video output, representing the audio vector
9179 The filter is used to measure the difference between channels of stereo
9180 audio stream. A monoaural signal, consisting of identical left and right
9181 signal, results in straight vertical line. Any stereo separation is visible
9182 as a deviation from this line, creating a Lissajous figure.
9183 If the straight (or deviation from it) but horizontal line appears this
9184 indicates that the left and right channels are out of phase.
9186 The filter accepts the following options:
9190 Set the vectorscope mode.
9192 Available values are:
9195 Lissajous rotated by 45 degrees.
9198 Same as above but not rotated.
9201 Default value is @samp{lissajous}.
9204 Set the video size for the output. For the syntax of this option, check the "Video size"
9205 section in the ffmpeg-utils manual. Default value is @code{400x400}.
9208 Set the output frame rate. Default value is @code{25}.
9213 Specify the red, green and blue contrast. Default values are @code{40}, @code{160} and @code{80}.
9214 Allowed range is @code{[0, 255]}.
9219 Specify the red, green and blue fade. Default values are @code{15}, @code{10} and @code{5}.
9220 Allowed range is @code{[0, 255]}.
9223 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
9226 @subsection Examples
9230 Complete example using @command{ffplay}:
9232 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
9233 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
9239 Concatenate audio and video streams, joining them together one after the
9242 The filter works on segments of synchronized video and audio streams. All
9243 segments must have the same number of streams of each type, and that will
9244 also be the number of streams at output.
9246 The filter accepts the following options:
9251 Set the number of segments. Default is 2.
9254 Set the number of output video streams, that is also the number of video
9255 streams in each segment. Default is 1.
9258 Set the number of output audio streams, that is also the number of video
9259 streams in each segment. Default is 0.
9262 Activate unsafe mode: do not fail if segments have a different format.
9266 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
9267 @var{a} audio outputs.
9269 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
9270 segment, in the same order as the outputs, then the inputs for the second
9273 Related streams do not always have exactly the same duration, for various
9274 reasons including codec frame size or sloppy authoring. For that reason,
9275 related synchronized streams (e.g. a video and its audio track) should be
9276 concatenated at once. The concat filter will use the duration of the longest
9277 stream in each segment (except the last one), and if necessary pad shorter
9278 audio streams with silence.
9280 For this filter to work correctly, all segments must start at timestamp 0.
9282 All corresponding streams must have the same parameters in all segments; the
9283 filtering system will automatically select a common pixel format for video
9284 streams, and a common sample format, sample rate and channel layout for
9285 audio streams, but other settings, such as resolution, must be converted
9286 explicitly by the user.
9288 Different frame rates are acceptable but will result in variable frame rate
9289 at output; be sure to configure the output file to handle it.
9291 @subsection Examples
9295 Concatenate an opening, an episode and an ending, all in bilingual version
9296 (video in stream 0, audio in streams 1 and 2):
9298 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
9299 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
9300 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
9301 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
9305 Concatenate two parts, handling audio and video separately, using the
9306 (a)movie sources, and adjusting the resolution:
9308 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
9309 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
9310 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
9312 Note that a desync will happen at the stitch if the audio and video streams
9313 do not have exactly the same duration in the first file.
9319 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
9320 it unchanged. By default, it logs a message at a frequency of 10Hz with the
9321 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
9322 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
9324 The filter also has a video output (see the @var{video} option) with a real
9325 time graph to observe the loudness evolution. The graphic contains the logged
9326 message mentioned above, so it is not printed anymore when this option is set,
9327 unless the verbose logging is set. The main graphing area contains the
9328 short-term loudness (3 seconds of analysis), and the gauge on the right is for
9329 the momentary loudness (400 milliseconds).
9331 More information about the Loudness Recommendation EBU R128 on
9332 @url{http://tech.ebu.ch/loudness}.
9334 The filter accepts the following options:
9339 Activate the video output. The audio stream is passed unchanged whether this
9340 option is set or no. The video stream will be the first output stream if
9341 activated. Default is @code{0}.
9344 Set the video size. This option is for video only. For the syntax of this
9345 option, check the "Video size" section in the ffmpeg-utils manual. Default
9346 and minimum resolution is @code{640x480}.
9349 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
9350 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
9351 other integer value between this range is allowed.
9354 Set metadata injection. If set to @code{1}, the audio input will be segmented
9355 into 100ms output frames, each of them containing various loudness information
9356 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
9358 Default is @code{0}.
9361 Force the frame logging level.
9363 Available values are:
9366 information logging level
9368 verbose logging level
9371 By default, the logging level is set to @var{info}. If the @option{video} or
9372 the @option{metadata} options are set, it switches to @var{verbose}.
9377 Available modes can be cumulated (the option is a @code{flag} type). Possible
9381 Disable any peak mode (default).
9383 Enable sample-peak mode.
9385 Simple peak mode looking for the higher sample value. It logs a message
9386 for sample-peak (identified by @code{SPK}).
9388 Enable true-peak mode.
9390 If enabled, the peak lookup is done on an over-sampled version of the input
9391 stream for better peak accuracy. It logs a message for true-peak.
9392 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
9393 This mode requires a build with @code{libswresample}.
9398 @subsection Examples
9402 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
9404 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
9408 Run an analysis with @command{ffmpeg}:
9410 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
9414 @section interleave, ainterleave
9416 Temporally interleave frames from several inputs.
9418 @code{interleave} works with video inputs, @code{ainterleave} with audio.
9420 These filters read frames from several inputs and send the oldest
9421 queued frame to the output.
9423 Input streams must have a well defined, monotonically increasing frame
9426 In order to submit one frame to output, these filters need to enqueue
9427 at least one frame for each input, so they cannot work in case one
9428 input is not yet terminated and will not receive incoming frames.
9430 For example consider the case when one input is a @code{select} filter
9431 which always drop input frames. The @code{interleave} filter will keep
9432 reading from that input, but it will never be able to send new frames
9433 to output until the input will send an end-of-stream signal.
9435 Also, depending on inputs synchronization, the filters will drop
9436 frames in case one input receives more frames than the other ones, and
9437 the queue is already filled.
9439 These filters accept the following options:
9443 Set the number of different inputs, it is 2 by default.
9446 @subsection Examples
9450 Interleave frames belonging to different streams using @command{ffmpeg}:
9452 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
9456 Add flickering blur effect:
9458 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
9462 @section perms, aperms
9464 Set read/write permissions for the output frames.
9466 These filters are mainly aimed at developers to test direct path in the
9467 following filter in the filtergraph.
9469 The filters accept the following options:
9473 Select the permissions mode.
9475 It accepts the following values:
9478 Do nothing. This is the default.
9480 Set all the output frames read-only.
9482 Set all the output frames directly writable.
9484 Make the frame read-only if writable, and writable if read-only.
9486 Set each output frame read-only or writable randomly.
9490 Set the seed for the @var{random} mode, must be an integer included between
9491 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
9492 @code{-1}, the filter will try to use a good random seed on a best effort
9496 Note: in case of auto-inserted filter between the permission filter and the
9497 following one, the permission might not be received as expected in that
9498 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
9499 perms/aperms filter can avoid this problem.
9501 @section select, aselect
9503 Select frames to pass in output.
9505 This filter accepts the following options:
9510 Set expression, which is evaluated for each input frame.
9512 If the expression is evaluated to zero, the frame is discarded.
9514 If the evaluation result is negative or NaN, the frame is sent to the
9515 first output; otherwise it is sent to the output with index
9516 @code{ceil(val)-1}, assuming that the input index starts from 0.
9518 For example a value of @code{1.2} corresponds to the output with index
9519 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
9522 Set the number of outputs. The output to which to send the selected
9523 frame is based on the result of the evaluation. Default value is 1.
9526 The expression can contain the following constants:
9530 the sequential number of the filtered frame, starting from 0
9533 the sequential number of the selected frame, starting from 0
9535 @item prev_selected_n
9536 the sequential number of the last selected frame, NAN if undefined
9539 timebase of the input timestamps
9542 the PTS (Presentation TimeStamp) of the filtered video frame,
9543 expressed in @var{TB} units, NAN if undefined
9546 the PTS (Presentation TimeStamp) of the filtered video frame,
9547 expressed in seconds, NAN if undefined
9550 the PTS of the previously filtered video frame, NAN if undefined
9552 @item prev_selected_pts
9553 the PTS of the last previously filtered video frame, NAN if undefined
9555 @item prev_selected_t
9556 the PTS of the last previously selected video frame, NAN if undefined
9559 the PTS of the first video frame in the video, NAN if undefined
9562 the time of the first video frame in the video, NAN if undefined
9564 @item pict_type @emph{(video only)}
9565 the type of the filtered frame, can assume one of the following
9577 @item interlace_type @emph{(video only)}
9578 the frame interlace type, can assume one of the following values:
9581 the frame is progressive (not interlaced)
9583 the frame is top-field-first
9585 the frame is bottom-field-first
9588 @item consumed_sample_n @emph{(audio only)}
9589 the number of selected samples before the current frame
9591 @item samples_n @emph{(audio only)}
9592 the number of samples in the current frame
9594 @item sample_rate @emph{(audio only)}
9595 the input sample rate
9598 1 if the filtered frame is a key-frame, 0 otherwise
9601 the position in the file of the filtered frame, -1 if the information
9602 is not available (e.g. for synthetic video)
9604 @item scene @emph{(video only)}
9605 value between 0 and 1 to indicate a new scene; a low value reflects a low
9606 probability for the current frame to introduce a new scene, while a higher
9607 value means the current frame is more likely to be one (see the example below)
9611 The default value of the select expression is "1".
9613 @subsection Examples
9617 Select all frames in input:
9622 The example above is the same as:
9634 Select only I-frames:
9636 select='eq(pict_type\,I)'
9640 Select one frame every 100:
9642 select='not(mod(n\,100))'
9646 Select only frames contained in the 10-20 time interval:
9648 select=between(t\,10\,20)
9652 Select only I frames contained in the 10-20 time interval:
9654 select=between(t\,10\,20)*eq(pict_type\,I)
9658 Select frames with a minimum distance of 10 seconds:
9660 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
9664 Use aselect to select only audio frames with samples number > 100:
9666 aselect='gt(samples_n\,100)'
9670 Create a mosaic of the first scenes:
9672 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
9675 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
9679 Send even and odd frames to separate outputs, and compose them:
9681 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
9685 @section sendcmd, asendcmd
9687 Send commands to filters in the filtergraph.
9689 These filters read commands to be sent to other filters in the
9692 @code{sendcmd} must be inserted between two video filters,
9693 @code{asendcmd} must be inserted between two audio filters, but apart
9694 from that they act the same way.
9696 The specification of commands can be provided in the filter arguments
9697 with the @var{commands} option, or in a file specified by the
9698 @var{filename} option.
9700 These filters accept the following options:
9703 Set the commands to be read and sent to the other filters.
9705 Set the filename of the commands to be read and sent to the other
9709 @subsection Commands syntax
9711 A commands description consists of a sequence of interval
9712 specifications, comprising a list of commands to be executed when a
9713 particular event related to that interval occurs. The occurring event
9714 is typically the current frame time entering or leaving a given time
9717 An interval is specified by the following syntax:
9719 @var{START}[-@var{END}] @var{COMMANDS};
9722 The time interval is specified by the @var{START} and @var{END} times.
9723 @var{END} is optional and defaults to the maximum time.
9725 The current frame time is considered within the specified interval if
9726 it is included in the interval [@var{START}, @var{END}), that is when
9727 the time is greater or equal to @var{START} and is lesser than
9730 @var{COMMANDS} consists of a sequence of one or more command
9731 specifications, separated by ",", relating to that interval. The
9732 syntax of a command specification is given by:
9734 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
9737 @var{FLAGS} is optional and specifies the type of events relating to
9738 the time interval which enable sending the specified command, and must
9739 be a non-null sequence of identifier flags separated by "+" or "|" and
9740 enclosed between "[" and "]".
9742 The following flags are recognized:
9745 The command is sent when the current frame timestamp enters the
9746 specified interval. In other words, the command is sent when the
9747 previous frame timestamp was not in the given interval, and the
9751 The command is sent when the current frame timestamp leaves the
9752 specified interval. In other words, the command is sent when the
9753 previous frame timestamp was in the given interval, and the
9757 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
9760 @var{TARGET} specifies the target of the command, usually the name of
9761 the filter class or a specific filter instance name.
9763 @var{COMMAND} specifies the name of the command for the target filter.
9765 @var{ARG} is optional and specifies the optional list of argument for
9766 the given @var{COMMAND}.
9768 Between one interval specification and another, whitespaces, or
9769 sequences of characters starting with @code{#} until the end of line,
9770 are ignored and can be used to annotate comments.
9772 A simplified BNF description of the commands specification syntax
9775 @var{COMMAND_FLAG} ::= "enter" | "leave"
9776 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
9777 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
9778 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
9779 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
9780 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
9783 @subsection Examples
9787 Specify audio tempo change at second 4:
9789 asendcmd=c='4.0 atempo tempo 1.5',atempo
9793 Specify a list of drawtext and hue commands in a file.
9795 # show text in the interval 5-10
9796 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
9797 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
9799 # desaturate the image in the interval 15-20
9800 15.0-20.0 [enter] hue s 0,
9801 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
9803 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
9805 # apply an exponential saturation fade-out effect, starting from time 25
9806 25 [enter] hue s exp(25-t)
9809 A filtergraph allowing to read and process the above command list
9810 stored in a file @file{test.cmd}, can be specified with:
9812 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
9817 @section setpts, asetpts
9819 Change the PTS (presentation timestamp) of the input frames.
9821 @code{setpts} works on video frames, @code{asetpts} on audio frames.
9823 This filter accepts the following options:
9828 The expression which is evaluated for each frame to construct its timestamp.
9832 The expression is evaluated through the eval API and can contain the following
9837 frame rate, only defined for constant frame-rate video
9840 the presentation timestamp in input
9843 the count of the input frame for video or the number of consumed samples,
9844 not including the current frame for audio, starting from 0.
9846 @item NB_CONSUMED_SAMPLES
9847 the number of consumed samples, not including the current frame (only
9851 the number of samples in the current frame (only audio)
9853 @item SAMPLE_RATE, SR
9857 the PTS of the first frame
9860 the time in seconds of the first frame
9863 tell if the current frame is interlaced
9866 the time in seconds of the current frame
9869 original position in the file of the frame, or undefined if undefined
9870 for the current frame
9876 previous input time in seconds
9882 previous output time in seconds
9885 wallclock (RTC) time in microseconds. This is deprecated, use time(0)
9889 wallclock (RTC) time at the start of the movie in microseconds
9892 timebase of the input timestamps
9896 @subsection Examples
9900 Start counting PTS from zero
9906 Apply fast motion effect:
9912 Apply slow motion effect:
9918 Set fixed rate of 25 frames per second:
9924 Set fixed rate 25 fps with some jitter:
9926 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
9930 Apply an offset of 10 seconds to the input PTS:
9936 Generate timestamps from a "live source" and rebase onto the current timebase:
9938 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
9942 Generate timestamps by counting samples:
9949 @section settb, asettb
9951 Set the timebase to use for the output frames timestamps.
9952 It is mainly useful for testing timebase configuration.
9954 This filter accepts the following options:
9959 The expression which is evaluated into the output timebase.
9963 The value for @option{tb} is an arithmetic expression representing a
9964 rational. The expression can contain the constants "AVTB" (the default
9965 timebase), "intb" (the input timebase) and "sr" (the sample rate,
9966 audio only). Default value is "intb".
9968 @subsection Examples
9972 Set the timebase to 1/25:
9978 Set the timebase to 1/10:
9984 Set the timebase to 1001/1000:
9990 Set the timebase to 2*intb:
9996 Set the default timebase value:
10002 @section showspectrum
10004 Convert input audio to a video output, representing the audio frequency
10007 The filter accepts the following options:
10011 Specify the video size for the output. For the syntax of this option, check
10012 the "Video size" section in the ffmpeg-utils manual. Default value is
10016 Specify if the spectrum should slide along the window. Default value is
10020 Specify display mode.
10022 It accepts the following values:
10025 all channels are displayed in the same row
10027 all channels are displayed in separate rows
10030 Default value is @samp{combined}.
10033 Specify display color mode.
10035 It accepts the following values:
10038 each channel is displayed in a separate color
10040 each channel is is displayed using the same color scheme
10043 Default value is @samp{channel}.
10046 Specify scale used for calculating intensity color values.
10048 It accepts the following values:
10053 square root, default
10060 Default value is @samp{sqrt}.
10063 Set saturation modifier for displayed colors. Negative values provide
10064 alternative color scheme. @code{0} is no saturation at all.
10065 Saturation must be in [-10.0, 10.0] range.
10066 Default value is @code{1}.
10069 Set window function.
10071 It accepts the following values:
10074 No samples pre-processing (do not expect this to be faster)
10083 Default value is @code{hann}.
10086 The usage is very similar to the showwaves filter; see the examples in that
10089 @subsection Examples
10093 Large window with logarithmic color scaling:
10095 showspectrum=s=1280x480:scale=log
10099 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
10101 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
10102 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
10108 Convert input audio to a video output, representing the samples waves.
10110 The filter accepts the following options:
10114 Specify the video size for the output. For the syntax of this option, check
10115 the "Video size" section in the ffmpeg-utils manual. Default value
10121 Available values are:
10124 Draw a point for each sample.
10127 Draw a vertical line for each sample.
10130 Default value is @code{point}.
10133 Set the number of samples which are printed on the same column. A
10134 larger value will decrease the frame rate. Must be a positive
10135 integer. This option can be set only if the value for @var{rate}
10136 is not explicitly specified.
10139 Set the (approximate) output frame rate. This is done by setting the
10140 option @var{n}. Default value is "25".
10144 @subsection Examples
10148 Output the input file audio and the corresponding video representation
10151 amovie=a.mp3,asplit[out0],showwaves[out1]
10155 Create a synthetic signal and show it with showwaves, forcing a
10156 frame rate of 30 frames per second:
10158 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
10162 @section split, asplit
10164 Split input into several identical outputs.
10166 @code{asplit} works with audio input, @code{split} with video.
10168 The filter accepts a single parameter which specifies the number of outputs. If
10169 unspecified, it defaults to 2.
10171 @subsection Examples
10175 Create two separate outputs from the same input:
10177 [in] split [out0][out1]
10181 To create 3 or more outputs, you need to specify the number of
10184 [in] asplit=3 [out0][out1][out2]
10188 Create two separate outputs from the same input, one cropped and
10191 [in] split [splitout1][splitout2];
10192 [splitout1] crop=100:100:0:0 [cropout];
10193 [splitout2] pad=200:200:100:100 [padout];
10197 Create 5 copies of the input audio with @command{ffmpeg}:
10199 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
10205 Receive commands sent through a libzmq client, and forward them to
10206 filters in the filtergraph.
10208 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
10209 must be inserted between two video filters, @code{azmq} between two
10212 To enable these filters you need to install the libzmq library and
10213 headers and configure FFmpeg with @code{--enable-libzmq}.
10215 For more information about libzmq see:
10216 @url{http://www.zeromq.org/}
10218 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
10219 receives messages sent through a network interface defined by the
10220 @option{bind_address} option.
10222 The received message must be in the form:
10224 @var{TARGET} @var{COMMAND} [@var{ARG}]
10227 @var{TARGET} specifies the target of the command, usually the name of
10228 the filter class or a specific filter instance name.
10230 @var{COMMAND} specifies the name of the command for the target filter.
10232 @var{ARG} is optional and specifies the optional argument list for the
10233 given @var{COMMAND}.
10235 Upon reception, the message is processed and the corresponding command
10236 is injected into the filtergraph. Depending on the result, the filter
10237 will send a reply to the client, adopting the format:
10239 @var{ERROR_CODE} @var{ERROR_REASON}
10243 @var{MESSAGE} is optional.
10245 @subsection Examples
10247 Look at @file{tools/zmqsend} for an example of a zmq client which can
10248 be used to send commands processed by these filters.
10250 Consider the following filtergraph generated by @command{ffplay}
10252 ffplay -dumpgraph 1 -f lavfi "
10253 color=s=100x100:c=red [l];
10254 color=s=100x100:c=blue [r];
10255 nullsrc=s=200x100, zmq [bg];
10256 [bg][l] overlay [bg+l];
10257 [bg+l][r] overlay=x=100 "
10260 To change the color of the left side of the video, the following
10261 command can be used:
10263 echo Parsed_color_0 c yellow | tools/zmqsend
10266 To change the right side:
10268 echo Parsed_color_1 c pink | tools/zmqsend
10271 @c man end MULTIMEDIA FILTERS
10273 @chapter Multimedia Sources
10274 @c man begin MULTIMEDIA SOURCES
10276 Below is a description of the currently available multimedia sources.
10280 This is the same as @ref{movie} source, except it selects an audio
10286 Read audio and/or video stream(s) from a movie container.
10288 This filter accepts the following options:
10292 The name of the resource to read (not necessarily a file but also a device or a
10293 stream accessed through some protocol).
10295 @item format_name, f
10296 Specifies the format assumed for the movie to read, and can be either
10297 the name of a container or an input device. If not specified the
10298 format is guessed from @var{movie_name} or by probing.
10300 @item seek_point, sp
10301 Specifies the seek point in seconds, the frames will be output
10302 starting from this seek point, the parameter is evaluated with
10303 @code{av_strtod} so the numerical value may be suffixed by an IS
10304 postfix. Default value is "0".
10307 Specifies the streams to read. Several streams can be specified,
10308 separated by "+". The source will then have as many outputs, in the
10309 same order. The syntax is explained in the ``Stream specifiers''
10310 section in the ffmpeg manual. Two special names, "dv" and "da" specify
10311 respectively the default (best suited) video and audio stream. Default
10312 is "dv", or "da" if the filter is called as "amovie".
10314 @item stream_index, si
10315 Specifies the index of the video stream to read. If the value is -1,
10316 the best suited video stream will be automatically selected. Default
10317 value is "-1". Deprecated. If the filter is called "amovie", it will select
10318 audio instead of video.
10321 Specifies how many times to read the stream in sequence.
10322 If the value is less than 1, the stream will be read again and again.
10323 Default value is "1".
10325 Note that when the movie is looped the source timestamps are not
10326 changed, so it will generate non monotonically increasing timestamps.
10329 This filter allows to overlay a second video on top of main input of
10330 a filtergraph as shown in this graph:
10332 input -----------> deltapts0 --> overlay --> output
10335 movie --> scale--> deltapts1 -------+
10338 @subsection Examples
10342 Skip 3.2 seconds from the start of the avi file in.avi, and overlay it
10343 on top of the input labelled as "in":
10345 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
10346 [in] setpts=PTS-STARTPTS [main];
10347 [main][over] overlay=16:16 [out]
10351 Read from a video4linux2 device, and overlay it on top of the input
10354 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
10355 [in] setpts=PTS-STARTPTS [main];
10356 [main][over] overlay=16:16 [out]
10360 Read the first video stream and the audio stream with id 0x81 from
10361 dvd.vob; the video is connected to the pad named "video" and the audio is
10362 connected to the pad named "audio":
10364 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
10368 @c man end MULTIMEDIA SOURCES