1 @chapter Filtergraph description
2 @c man begin FILTERGRAPH DESCRIPTION
4 A filtergraph is a directed graph of connected filters. It can contain
5 cycles, and there can be multiple links between a pair of
6 filters. Each link has one input pad on one side connecting it to one
7 filter from which it takes its input, and one output pad on the other
8 side connecting it to the one filter accepting its output.
10 Each filter in a filtergraph is an instance of a filter class
11 registered in the application, which defines the features and the
12 number of input and output pads of the filter.
14 A filter with no input pads is called a "source", a filter with no
15 output pads is called a "sink".
17 @section Filtergraph syntax
19 A filtergraph can be represented using a textual representation, which
20 is recognized by the @code{-vf} option of the ff*
21 tools, and by the @code{avfilter_graph_parse()} function defined in
22 @file{libavfilter/avfiltergraph.h}.
24 A filterchain consists of a sequence of connected filters, each one
25 connected to the previous one in the sequence. A filterchain is
26 represented by a list of ","-separated filter descriptions.
28 A filtergraph consists of a sequence of filterchains. A sequence of
29 filterchains is represented by a list of ";"-separated filterchain
32 A filter is represented by a string of the form:
33 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
35 @var{filter_name} is the name of the filter class of which the
36 described filter is an instance of, and has to be the name of one of
37 the filter classes registered in the program.
38 The name of the filter class is optionally followed by a string
41 @var{arguments} is a string which contains the parameters used to
42 initialize the filter instance, and are described in the filter
45 The list of arguments can be quoted using the character "'" as initial
46 and ending mark, and the character '\' for escaping the characters
47 within the quoted text; otherwise the argument string is considered
48 terminated when the next special character (belonging to the set
49 "[]=;,") is encountered.
51 The name and arguments of the filter are optionally preceded and
52 followed by a list of link labels.
53 A link label allows to name a link and associate it to a filter output
54 or input pad. The preceding labels @var{in_link_1}
55 ... @var{in_link_N}, are associated to the filter input pads,
56 the following labels @var{out_link_1} ... @var{out_link_M}, are
57 associated to the output pads.
59 When two link labels with the same name are found in the
60 filtergraph, a link between the corresponding input and output pad is
63 If an output pad is not labelled, it is linked by default to the first
64 unlabelled input pad of the next filter in the filterchain.
65 For example in the filterchain:
67 nullsrc, split[L1], [L2]overlay, nullsink
69 the split filter instance has two output pads, and the overlay filter
70 instance two input pads. The first output pad of split is labelled
71 "L1", the first input pad of overlay is labelled "L2", and the second
72 output pad of split is linked to the second input pad of overlay,
73 which are both unlabelled.
75 In a complete filterchain all the unlabelled filter input and output
76 pads must be connected. A filtergraph is considered valid if all the
77 filter input and output pads of all the filterchains are connected.
79 Follows a BNF description for the filtergraph syntax:
81 @var{NAME} ::= sequence of alphanumeric characters and '_'
82 @var{LINKLABEL} ::= "[" @var{NAME} "]"
83 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
84 @var{FILTER_ARGUMENTS} ::= sequence of chars (eventually quoted)
85 @var{FILTER} ::= [@var{LINKNAMES}] @var{NAME} ["=" @var{ARGUMENTS}] [@var{LINKNAMES}]
86 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
87 @var{FILTERGRAPH} ::= @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
90 @c man end FILTERGRAPH DESCRIPTION
92 @chapter Audio Filters
93 @c man begin AUDIO FILTERS
95 When you configure your FFmpeg build, you can disable any of the
96 existing filters using @code{--disable-filters}.
97 The configure output will show the audio filters included in your
100 Below is a description of the currently available audio filters.
104 Convert the input audio format to the specified formats.
106 The filter accepts a string of the form:
107 "@var{sample_format}:@var{channel_layout}".
109 @var{sample_format} specifies the sample format, and can be a string or the
110 corresponding numeric value defined in @file{libavutil/samplefmt.h}. Use 'p'
111 suffix for a planar sample format.
113 @var{channel_layout} specifies the channel layout, and can be a string
114 or the corresponding number value defined in @file{libavutil/audioconvert.h}.
116 The special parameter "auto", signifies that the filter will
117 automatically select the output format depending on the output filter.
119 Some examples follow.
123 Convert input to float, planar, stereo:
129 Convert input to unsigned 8-bit, automatically select out channel layout:
137 Convert the input audio to one of the specified formats. The framework will
138 negotiate the most appropriate format to minimize conversions.
140 The filter accepts three lists of formats, separated by ":", in the form:
141 "@var{sample_formats}:@var{channel_layouts}:@var{packing_formats}".
143 Elements in each list are separated by "," which has to be escaped in the
144 filtergraph specification.
146 The special parameter "all", in place of a list of elements, signifies all
149 Some examples follow:
151 aformat=u8\\,s16:mono:packed
153 aformat=s16:mono\\,stereo:all
158 Merge two audio streams into a single multi-channel stream.
160 This filter does not need any argument.
162 If the channel layouts of the inputs are disjoint, and therefore compatible,
163 the channel layout of the output will be set accordingly and the channels
164 will be reordered as necessary. If the channel layouts of the inputs are not
165 disjoint, the output will have all the channels of the first input then all
166 the channels of the second input, in that order, and the channel layout of
167 the output will be the default value corresponding to the total number of
170 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
171 is FC+BL+BR, then the output will be in 5.1, with the channels in the
172 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
173 first input, b1 is the first channel of the second input).
175 On the other hand, if both input are in stereo, the output channels will be
176 in the default order: a1, a2, b1, b2, and the channel layout will be
177 arbitrarily set to 4.0, which may or may not be the expected value.
179 Both inputs must have the same sample rate, format and packing.
181 If inputs do not have the same duration, the output will stop with the
184 Example: merge two mono files into a stereo stream:
186 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
189 If you need to do multiple merges (for instance multiple mono audio streams in
190 a single video media), you can do:
193 amovie=input.mkv:si=0 [a0];
194 amovie=input.mkv:si=1 [a1];
195 amovie=input.mkv:si=2 [a2];
196 amovie=input.mkv:si=3 [a3];
197 amovie=input.mkv:si=4 [a4];
198 amovie=input.mkv:si=5 [a5];
199 [a0][a1] amerge [x0];
200 [x0][a2] amerge [x1];
201 [x1][a3] amerge [x2];
202 [x2][a4] amerge [x3];
203 [x3][a5] amerge" -c:a pcm_s16le output.mkv
208 Pass the audio source unchanged to the output.
212 Resample the input audio to the specified sample rate.
214 The filter accepts exactly one parameter, the output sample rate. If not
215 specified then the filter will automatically convert between its input
216 and output sample rates.
218 For example, to resample the input audio to 44100Hz:
225 Show a line containing various information for each input audio frame.
226 The input audio is not modified.
228 The shown line contains a sequence of key/value pairs of the form
229 @var{key}:@var{value}.
231 A description of each shown parameter follows:
235 sequential number of the input frame, starting from 0
238 presentation TimeStamp of the input frame, expressed as a number of
239 time base units. The time base unit depends on the filter input pad, and
240 is usually 1/@var{sample_rate}.
243 presentation TimeStamp of the input frame, expressed as a number of
247 position of the frame in the input stream, -1 if this information in
248 unavailable and/or meaningless (for example in case of synthetic audio)
254 channel layout description
257 number of samples (per each channel) contained in the filtered frame
260 sample rate for the audio frame
263 if the packing format is planar, 0 if packed
266 Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame
269 Adler-32 checksum (printed in hexadecimal) for each input frame plane,
270 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3} @var{c4} @var{c5}
276 Pass on the input audio to two outputs. Both outputs are identical to
281 [in] asplit[out0], showaudio[out1]
284 will create two separate outputs from the same input, one cropped and
289 Forward two audio streams and control the order the buffers are forwarded.
291 The argument to the filter is an expression deciding which stream should be
292 forwarded next: if the result is negative, the first stream is forwarded; if
293 the result is positive or zero, the second stream is forwarded. It can use
294 the following variables:
298 number of buffers forwarded so far on each stream
300 number of samples forwarded so far on each stream
302 current timestamp of each stream
305 The default value is @code{t1-t2}, which means to always forward the stream
306 that has a smaller timestamp.
308 Example: stress-test @code{amerge} by randomly sending buffers on the wrong
309 input, while avoiding too much of a desynchronization:
311 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
312 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
318 Make audio easier to listen to on headphones.
320 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
321 so that when listened to on headphones the stereo image is moved from
322 inside your head (standard for headphones) to outside and in front of
323 the listener (standard for speakers).
329 Mix channels with specific gain levels. The filter accepts the output
330 channel layout followed by a set of channels definitions.
332 This filter is also designed to remap efficiently the channels of an audio
335 The filter accepts parameters of the form:
336 "@var{l}:@var{outdef}:@var{outdef}:..."
340 output channel layout or number of channels
343 output channel specification, of the form:
344 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
347 output channel to define, either a channel name (FL, FR, etc.) or a channel
348 number (c0, c1, etc.)
351 multiplicative coefficient for the channel, 1 leaving the volume unchanged
354 input channel to use, see out_name for details; it is not possible to mix
355 named and numbered input channels
358 If the `=' in a channel specification is replaced by `<', then the gains for
359 that specification will be renormalized so that the total is 1, thus
360 avoiding clipping noise.
362 @subsection Mixing examples
364 For example, if you want to down-mix from stereo to mono, but with a bigger
365 factor for the left channel:
367 pan=1:c0=0.9*c0+0.1*c1
370 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
373 pan=stereo: FL < FL + 0.5*FC + 0.6*BL + 0.6*SL : FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
376 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
377 that should be preferred (see "-ac" option) unless you have very specific
380 @subsection Remapping examples
382 The channel remapping will be effective if, and only if:
385 @item gain coefficients are zeroes or ones,
386 @item only one input per channel output,
389 If all these conditions are satisfied, the filter will notify the user ("Pure
390 channel mapping detected"), and use an optimized and lossless method to do the
393 For example, if you have a 5.1 source and want a stereo audio stream by
394 dropping the extra channels:
396 pan="stereo: c0=FL : c1=FR"
399 Given the same source, you can also switch front left and front right channels
400 and keep the input channel layout:
402 pan="5.1: c0=c1 : c1=c0 : c2=c2 : c3=c3 : c4=c4 : c5=c5"
405 If the input is a stereo audio stream, you can mute the front left channel (and
406 still keep the stereo channel layout) with:
411 Still with a stereo audio stream input, you can copy the right channel in both
412 front left and right:
414 pan="stereo: c0=FR : c1=FR"
417 @section silencedetect
419 Detect silence in an audio stream.
421 This filter logs a message when it detects that the input audio volume is less
422 or equal to a noise tolerance value for a duration greater or equal to the
423 minimum detected noise duration.
425 The printed times and duration are expressed in seconds.
429 Set silence duration until notification (default is 2 seconds).
432 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
433 specified value) or amplitude ratio. Default is -60dB, or 0.001.
436 Detect 5 seconds of silence with -50dB noise tolerance:
438 silencedetect=n=-50dB:d=5
441 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
442 tolerance in @file{silence.mp3}:
444 ffmpeg -f lavfi -i amovie=silence.mp3,silencedetect=noise=0.0001 -f null -
449 Adjust the input audio volume.
451 The filter accepts exactly one parameter @var{vol}, which expresses
452 how the audio volume will be increased or decreased.
454 Output values are clipped to the maximum value.
456 If @var{vol} is expressed as a decimal number, the output audio
457 volume is given by the relation:
459 @var{output_volume} = @var{vol} * @var{input_volume}
462 If @var{vol} is expressed as a decimal number followed by the string
463 "dB", the value represents the requested change in decibels of the
464 input audio power, and the output audio volume is given by the
467 @var{output_volume} = 10^(@var{vol}/20) * @var{input_volume}
470 Otherwise @var{vol} is considered an expression and its evaluated
471 value is used for computing the output audio volume according to the
474 Default value for @var{vol} is 1.0.
480 Half the input audio volume:
485 The above example is equivalent to:
491 Decrease input audio power by 12 decibels:
497 @c man end AUDIO FILTERS
499 @chapter Audio Sources
500 @c man begin AUDIO SOURCES
502 Below is a description of the currently available audio sources.
506 Buffer audio frames, and make them available to the filter chain.
508 This source is mainly intended for a programmatic use, in particular
509 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
511 It accepts the following mandatory parameters:
512 @var{sample_rate}:@var{sample_fmt}:@var{channel_layout}:@var{packing}
517 The sample rate of the incoming audio buffers.
520 The sample format of the incoming audio buffers.
521 Either a sample format name or its corresponging integer representation from
522 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
525 The channel layout of the incoming audio buffers.
526 Either a channel layout name from channel_layout_map in
527 @file{libavutil/audioconvert.c} or its corresponding integer representation
528 from the AV_CH_LAYOUT_* macros in @file{libavutil/audioconvert.h}
531 Either "packed" or "planar", or their integer representation: 0 or 1
538 abuffer=44100:s16:stereo:planar
541 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
542 Since the sample format with name "s16" corresponds to the number
543 1 and the "stereo" channel layout corresponds to the value 0x3, this is
546 abuffer=44100:1:0x3:1
551 Generate an audio signal specified by an expression.
553 This source accepts in input one or more expressions (one for each
554 channel), which are evaluated and used to generate a corresponding
557 It accepts the syntax: @var{exprs}[::@var{options}].
558 @var{exprs} is a list of expressions separated by ":", one for each
559 separate channel. The output channel layout depends on the number of
560 provided expressions, up to 8 channels are supported.
562 @var{options} is an optional sequence of @var{key}=@var{value} pairs,
565 The description of the accepted options follows.
570 Set the minimum duration of the sourced audio. See the function
571 @code{av_parse_time()} for the accepted format.
572 Note that the resulting duration may be greater than the specified
573 duration, as the generated audio is always cut at the end of a
576 If not specified, or the expressed duration is negative, the audio is
577 supposed to be generated forever.
580 Set the number of samples per channel per each output frame,
584 Specify the sample rate, default to 44100.
587 Each expression in @var{exprs} can contain the following constants:
591 number of the evaluated sample, starting from 0
594 time of the evaluated sample expressed in seconds, starting from 0
613 Generate a sin signal with frequency of 440 Hz, set sample rate to
616 aevalsrc="sin(440*2*PI*t)::s=8000"
620 Generate white noise:
622 aevalsrc="-2+random(0)"
626 Generate an amplitude modulated signal:
628 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
632 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
634 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) : 0.1*sin(2*PI*(360+2.5/2)*t)"
641 Read an audio stream from a movie container.
643 It accepts the syntax: @var{movie_name}[:@var{options}] where
644 @var{movie_name} is the name of the resource to read (not necessarily
645 a file but also a device or a stream accessed through some protocol),
646 and @var{options} is an optional sequence of @var{key}=@var{value}
647 pairs, separated by ":".
649 The description of the accepted options follows.
654 Specify the format assumed for the movie to read, and can be either
655 the name of a container or an input device. If not specified the
656 format is guessed from @var{movie_name} or by probing.
659 Specify the seek point in seconds, the frames will be output
660 starting from this seek point, the parameter is evaluated with
661 @code{av_strtod} so the numerical value may be suffixed by an IS
662 postfix. Default value is "0".
664 @item stream_index, si
665 Specify the index of the audio stream to read. If the value is -1,
666 the best suited audio stream will be automatically selected. Default
673 Null audio source, return unprocessed audio frames. It is mainly useful
674 as a template and to be employed in analysis / debugging tools, or as
675 the source for filters which ignore the input data (for example the sox
678 It accepts an optional sequence of @var{key}=@var{value} pairs,
681 The description of the accepted options follows.
686 Specify the sample rate, and defaults to 44100.
688 @item channel_layout, cl
690 Specify the channel layout, and can be either an integer or a string
691 representing a channel layout. The default value of @var{channel_layout}
694 Check the channel_layout_map definition in
695 @file{libavcodec/audioconvert.c} for the mapping between strings and
696 channel layout values.
699 Set the number of samples per requested frames.
703 Follow some examples:
705 # set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
706 anullsrc=r=48000:cl=4
709 anullsrc=r=48000:cl=mono
712 @c man end AUDIO SOURCES
715 @c man begin AUDIO SINKS
717 Below is a description of the currently available audio sinks.
721 Buffer audio frames, and make them available to the end of filter chain.
723 This sink is mainly intended for programmatic use, in particular
724 through the interface defined in @file{libavfilter/buffersink.h}.
726 It requires a pointer to an AVABufferSinkContext structure, which
727 defines the incoming buffers' formats, to be passed as the opaque
728 parameter to @code{avfilter_init_filter} for initialization.
732 Null audio sink, do absolutely nothing with the input audio. It is
733 mainly useful as a template and to be employed in analysis / debugging
736 @c man end AUDIO SINKS
738 @chapter Video Filters
739 @c man begin VIDEO FILTERS
741 When you configure your FFmpeg build, you can disable any of the
742 existing filters using @code{--disable-filters}.
743 The configure output will show the video filters included in your
746 Below is a description of the currently available video filters.
750 Draw ASS (Advanced Substation Alpha) subtitles on top of input video
751 using the libass library.
753 To enable compilation of this filter you need to configure FFmpeg with
754 @code{--enable-libass}.
756 This filter accepts the syntax: @var{ass_filename}[:@var{options}],
757 where @var{ass_filename} is the filename of the ASS file to read, and
758 @var{options} is an optional sequence of @var{key}=@var{value} pairs,
761 A description of the accepted options follows.
765 Specifies the size of the original video, the video for which the ASS file
766 was composed. Due to a misdesign in ASS aspect ratio arithmetic, this is
767 necessary to correctly scale the fonts if the aspect ratio has been changed.
770 For example, to render the file @file{sub.ass} on top of the input
771 video, use the command:
778 Compute the bounding box for the non-black pixels in the input frame
781 This filter computes the bounding box containing all the pixels with a
782 luminance value greater than the minimum allowed value.
783 The parameters describing the bounding box are printed on the filter
788 Detect video intervals that are (almost) completely black. Can be
789 useful to detect chapter transitions, commercials, or invalid
790 recordings. Output lines contains the time for the start, end and
791 duration of the detected black interval expressed in seconds.
793 In order to display the output lines, you need to set the loglevel at
794 least to the AV_LOG_INFO value.
796 This filter accepts a list of options in the form of
797 @var{key}=@var{value} pairs separated by ":". A description of the
798 accepted options follows.
801 @item black_min_duration, d
802 Set the minimum detected black duration expressed in seconds. It must
803 be a non-negative floating point number.
805 Default value is 2.0.
807 @item picture_black_ratio_th, pic_th
808 Set the threshold for considering a picture "black".
809 Express the minimum value for the ratio:
811 @var{nb_black_pixels} / @var{nb_pixels}
814 for which a picture is considered black.
815 Default value is 0.98.
817 @item pixel_black_th, pix_th
818 Set the threshold for considering a pixel "black".
820 The threshold expresses the maximum pixel luminance value for which a
821 pixel is considered "black". The provided value is scaled according to
822 the following equation:
824 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
827 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
828 the input video format, the range is [0-255] for YUV full-range
829 formats and [16-235] for YUV non full-range formats.
831 Default value is 0.10.
834 The following example sets the maximum pixel threshold to the minimum
835 value, and detects only black intervals of 2 or more seconds:
837 blackdetect=d=2:pix_th=0.00
842 Detect frames that are (almost) completely black. Can be useful to
843 detect chapter transitions or commercials. Output lines consist of
844 the frame number of the detected frame, the percentage of blackness,
845 the position in the file if known or -1 and the timestamp in seconds.
847 In order to display the output lines, you need to set the loglevel at
848 least to the AV_LOG_INFO value.
850 The filter accepts the syntax:
852 blackframe[=@var{amount}:[@var{threshold}]]
855 @var{amount} is the percentage of the pixels that have to be below the
856 threshold, and defaults to 98.
858 @var{threshold} is the threshold below which a pixel value is
859 considered black, and defaults to 32.
863 Apply boxblur algorithm to the input video.
865 This filter accepts the parameters:
866 @var{luma_radius}:@var{luma_power}:@var{chroma_radius}:@var{chroma_power}:@var{alpha_radius}:@var{alpha_power}
868 Chroma and alpha parameters are optional, if not specified they default
869 to the corresponding values set for @var{luma_radius} and
872 @var{luma_radius}, @var{chroma_radius}, and @var{alpha_radius} represent
873 the radius in pixels of the box used for blurring the corresponding
874 input plane. They are expressions, and can contain the following
878 the input width and height in pixels
881 the input chroma image width and height in pixels
884 horizontal and vertical chroma subsample values. For example for the
885 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
888 The radius must be a non-negative number, and must not be greater than
889 the value of the expression @code{min(w,h)/2} for the luma and alpha planes,
890 and of @code{min(cw,ch)/2} for the chroma planes.
892 @var{luma_power}, @var{chroma_power}, and @var{alpha_power} represent
893 how many times the boxblur filter is applied to the corresponding
896 Some examples follow:
901 Apply a boxblur filter with luma, chroma, and alpha radius
908 Set luma radius to 2, alpha and chroma radius to 0
914 Set luma and chroma radius to a fraction of the video dimension
916 boxblur=min(h\,w)/10:1:min(cw\,ch)/10:1
923 The colormatrix filter allows conversion between any of the following color
924 space: BT.709 (@var{bt709}), BT.601 (@var{bt601}), SMPTE-240M (@var{smpte240m})
927 The syntax of the parameters is @var{source}:@var{destination}:
930 colormatrix=bt601:smpte240m
935 Copy the input source unchanged to the output. Mainly useful for
940 Crop the input video to @var{out_w}:@var{out_h}:@var{x}:@var{y}:@var{keep_aspect}
942 The @var{keep_aspect} parameter is optional, if specified and set to a
943 non-zero value will force the output display aspect ratio to be the
944 same of the input, by changing the output sample aspect ratio.
946 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
947 expressions containing the following constants:
951 the computed values for @var{x} and @var{y}. They are evaluated for
955 the input width and height
958 same as @var{in_w} and @var{in_h}
961 the output (cropped) width and height
964 same as @var{out_w} and @var{out_h}
967 same as @var{iw} / @var{ih}
970 input sample aspect ratio
973 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
976 horizontal and vertical chroma subsample values. For example for the
977 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
980 the number of input frame, starting from 0
983 the position in the file of the input frame, NAN if unknown
986 timestamp expressed in seconds, NAN if the input timestamp is unknown
990 The @var{out_w} and @var{out_h} parameters specify the expressions for
991 the width and height of the output (cropped) video. They are
992 evaluated just at the configuration of the filter.
994 The default value of @var{out_w} is "in_w", and the default value of
995 @var{out_h} is "in_h".
997 The expression for @var{out_w} may depend on the value of @var{out_h},
998 and the expression for @var{out_h} may depend on @var{out_w}, but they
999 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
1000 evaluated after @var{out_w} and @var{out_h}.
1002 The @var{x} and @var{y} parameters specify the expressions for the
1003 position of the top-left corner of the output (non-cropped) area. They
1004 are evaluated for each frame. If the evaluated value is not valid, it
1005 is approximated to the nearest valid value.
1007 The default value of @var{x} is "(in_w-out_w)/2", and the default
1008 value for @var{y} is "(in_h-out_h)/2", which set the cropped area at
1009 the center of the input image.
1011 The expression for @var{x} may depend on @var{y}, and the expression
1012 for @var{y} may depend on @var{x}.
1014 Follow some examples:
1016 # crop the central input area with size 100x100
1019 # crop the central input area with size 2/3 of the input video
1020 "crop=2/3*in_w:2/3*in_h"
1022 # crop the input video central square
1025 # delimit the rectangle with the top-left corner placed at position
1026 # 100:100 and the right-bottom corner corresponding to the right-bottom
1027 # corner of the input image.
1028 crop=in_w-100:in_h-100:100:100
1030 # crop 10 pixels from the left and right borders, and 20 pixels from
1031 # the top and bottom borders
1032 "crop=in_w-2*10:in_h-2*20"
1034 # keep only the bottom right quarter of the input image
1035 "crop=in_w/2:in_h/2:in_w/2:in_h/2"
1037 # crop height for getting Greek harmony
1038 "crop=in_w:1/PHI*in_w"
1041 "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)"
1043 # erratic camera effect depending on timestamp
1044 "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)"
1046 # set x depending on the value of y
1047 "crop=in_w/2:in_h/2:y:10+10*sin(n/10)"
1052 Auto-detect crop size.
1054 Calculate necessary cropping parameters and prints the recommended
1055 parameters through the logging system. The detected dimensions
1056 correspond to the non-black area of the input video.
1058 It accepts the syntax:
1060 cropdetect[=@var{limit}[:@var{round}[:@var{reset}]]]
1066 Threshold, which can be optionally specified from nothing (0) to
1067 everything (255), defaults to 24.
1070 Value which the width/height should be divisible by, defaults to
1071 16. The offset is automatically adjusted to center the video. Use 2 to
1072 get only even dimensions (needed for 4:2:2 video). 16 is best when
1073 encoding to most video codecs.
1076 Counter that determines after how many frames cropdetect will reset
1077 the previously detected largest video area and start over to detect
1078 the current optimal crop area. Defaults to 0.
1080 This can be useful when channel logos distort the video area. 0
1081 indicates never reset and return the largest area encountered during
1087 Suppress a TV station logo by a simple interpolation of the surrounding
1088 pixels. Just set a rectangle covering the logo and watch it disappear
1089 (and sometimes something even uglier appear - your mileage may vary).
1091 The filter accepts parameters as a string of the form
1092 "@var{x}:@var{y}:@var{w}:@var{h}:@var{band}", or as a list of
1093 @var{key}=@var{value} pairs, separated by ":".
1095 The description of the accepted parameters follows.
1100 Specify the top left corner coordinates of the logo. They must be
1104 Specify the width and height of the logo to clear. They must be
1108 Specify the thickness of the fuzzy edge of the rectangle (added to
1109 @var{w} and @var{h}). The default value is 4.
1112 When set to 1, a green rectangle is drawn on the screen to simplify
1113 finding the right @var{x}, @var{y}, @var{w}, @var{h} parameters, and
1114 @var{band} is set to 4. The default value is 0.
1118 Some examples follow.
1123 Set a rectangle covering the area with top left corner coordinates 0,0
1124 and size 100x77, setting a band of size 10:
1126 delogo=0:0:100:77:10
1130 As the previous example, but use named options:
1132 delogo=x=0:y=0:w=100:h=77:band=10
1139 Attempt to fix small changes in horizontal and/or vertical shift. This
1140 filter helps remove camera shake from hand-holding a camera, bumping a
1141 tripod, moving on a vehicle, etc.
1143 The filter accepts parameters as a string of the form
1144 "@var{x}:@var{y}:@var{w}:@var{h}:@var{rx}:@var{ry}:@var{edge}:@var{blocksize}:@var{contrast}:@var{search}:@var{filename}"
1146 A description of the accepted parameters follows.
1151 Specify a rectangular area where to limit the search for motion
1153 If desired the search for motion vectors can be limited to a
1154 rectangular area of the frame defined by its top left corner, width
1155 and height. These parameters have the same meaning as the drawbox
1156 filter which can be used to visualise the position of the bounding
1159 This is useful when simultaneous movement of subjects within the frame
1160 might be confused for camera motion by the motion vector search.
1162 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
1163 then the full frame is used. This allows later options to be set
1164 without specifying the bounding box for the motion vector search.
1166 Default - search the whole frame.
1169 Specify the maximum extent of movement in x and y directions in the
1170 range 0-64 pixels. Default 16.
1173 Specify how to generate pixels to fill blanks at the edge of the
1174 frame. An integer from 0 to 3 as follows:
1177 Fill zeroes at blank locations
1179 Original image at blank locations
1181 Extruded edge value at blank locations
1183 Mirrored edge at blank locations
1186 The default setting is mirror edge at blank locations.
1189 Specify the blocksize to use for motion search. Range 4-128 pixels,
1193 Specify the contrast threshold for blocks. Only blocks with more than
1194 the specified contrast (difference between darkest and lightest
1195 pixels) will be considered. Range 1-255, default 125.
1198 Specify the search strategy 0 = exhaustive search, 1 = less exhaustive
1199 search. Default - exhaustive search.
1202 If set then a detailed log of the motion search is written to the
1209 Draw a colored box on the input image.
1211 It accepts the syntax:
1213 drawbox=@var{x}:@var{y}:@var{width}:@var{height}:@var{color}
1219 Specify the top left corner coordinates of the box. Default to 0.
1222 Specify the width and height of the box, if 0 they are interpreted as
1223 the input width and height. Default to 0.
1226 Specify the color of the box to write, it can be the name of a color
1227 (case insensitive match) or a 0xRRGGBB[AA] sequence.
1230 Follow some examples:
1232 # draw a black box around the edge of the input image
1235 # draw a box with color red and an opacity of 50%
1236 drawbox=10:20:200:60:red@@0.5"
1241 Draw text string or text from specified file on top of video using the
1242 libfreetype library.
1244 To enable compilation of this filter you need to configure FFmpeg with
1245 @code{--enable-libfreetype}.
1247 The filter also recognizes strftime() sequences in the provided text
1248 and expands them accordingly. Check the documentation of strftime().
1250 The filter accepts parameters as a list of @var{key}=@var{value} pairs,
1253 The description of the accepted parameters follows.
1258 The font file to be used for drawing text. Path must be included.
1259 This parameter is mandatory.
1262 The text string to be drawn. The text must be a sequence of UTF-8
1264 This parameter is mandatory if no file is specified with the parameter
1268 A text file containing text to be drawn. The text must be a sequence
1269 of UTF-8 encoded characters.
1271 This parameter is mandatory if no text string is specified with the
1272 parameter @var{text}.
1274 If both text and textfile are specified, an error is thrown.
1277 The expressions which specify the offsets where text will be drawn
1278 within the video frame. They are relative to the top/left border of the
1281 The default value of @var{x} and @var{y} is "0".
1283 See below for the list of accepted constants.
1286 The font size to be used for drawing text.
1287 The default value of @var{fontsize} is 16.
1290 The color to be used for drawing fonts.
1291 Either a string (e.g. "red") or in 0xRRGGBB[AA] format
1292 (e.g. "0xff000033"), possibly followed by an alpha specifier.
1293 The default value of @var{fontcolor} is "black".
1296 The color to be used for drawing box around text.
1297 Either a string (e.g. "yellow") or in 0xRRGGBB[AA] format
1298 (e.g. "0xff00ff"), possibly followed by an alpha specifier.
1299 The default value of @var{boxcolor} is "white".
1302 Used to draw a box around text using background color.
1303 Value should be either 1 (enable) or 0 (disable).
1304 The default value of @var{box} is 0.
1306 @item shadowx, shadowy
1307 The x and y offsets for the text shadow position with respect to the
1308 position of the text. They can be either positive or negative
1309 values. Default value for both is "0".
1312 The color to be used for drawing a shadow behind the drawn text. It
1313 can be a color name (e.g. "yellow") or a string in the 0xRRGGBB[AA]
1314 form (e.g. "0xff00ff"), possibly followed by an alpha specifier.
1315 The default value of @var{shadowcolor} is "black".
1318 Flags to be used for loading the fonts.
1320 The flags map the corresponding flags supported by libfreetype, and are
1321 a combination of the following values:
1328 @item vertical_layout
1329 @item force_autohint
1332 @item ignore_global_advance_width
1334 @item ignore_transform
1341 Default value is "render".
1343 For more information consult the documentation for the FT_LOAD_*
1347 The size in number of spaces to use for rendering the tab.
1351 If true, check and fix text coords to avoid clipping.
1354 The parameters for @var{x} and @var{y} are expressions containing the
1355 following constants:
1359 the input width and height
1362 the width of the rendered text
1365 the height of the rendered text
1368 the height of each text line
1371 input sample aspect ratio
1374 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
1377 horizontal and vertical chroma subsample values. For example for the
1378 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
1381 maximum glyph width, that is the maximum width for all the glyphs
1382 contained in the rendered text
1385 maximum glyph height, that is the maximum height for all the glyphs
1386 contained in the rendered text, it is equivalent to @var{ascent} -
1389 @item max_glyph_a, ascent
1391 the maximum distance from the baseline to the highest/upper grid
1392 coordinate used to place a glyph outline point, for all the rendered
1394 It is a positive value, due to the grid's orientation with the Y axis
1397 @item max_glyph_d, descent
1398 the maximum distance from the baseline to the lowest grid coordinate
1399 used to place a glyph outline point, for all the rendered glyphs.
1400 This is a negative value, due to the grid's orientation, with the Y axis
1404 the number of input frame, starting from 0
1407 timestamp expressed in seconds, NAN if the input timestamp is unknown
1410 initial timecode representation in "hh:mm:ss[:;.]ff" format. It can be used
1411 with or without text parameter. @var{rate} option must be specified.
1414 frame rate (timecode only)
1417 Some examples follow.
1422 Draw "Test Text" with font FreeSerif, using the default values for the
1423 optional parameters.
1426 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
1430 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
1431 and y=50 (counting from the top-left corner of the screen), text is
1432 yellow with a red box around it. Both the text and the box have an
1436 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
1437 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
1440 Note that the double quotes are not necessary if spaces are not used
1441 within the parameter list.
1444 Show the text at the center of the video frame:
1446 drawtext=fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
1450 Show a text line sliding from right to left in the last row of the video
1451 frame. The file @file{LONG_LINE} is assumed to contain a single line
1454 drawtext=fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t
1458 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
1460 drawtext=fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
1464 Draw a single green letter "g", at the center of the input video.
1465 The glyph baseline is placed at half screen height.
1467 drawtext=fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent
1472 For more information about libfreetype, check:
1473 @url{http://www.freetype.org/}.
1477 Apply fade-in/out effect to input video.
1479 It accepts the parameters:
1480 @var{type}:@var{start_frame}:@var{nb_frames}[:@var{options}]
1482 @var{type} specifies if the effect type, can be either "in" for
1483 fade-in, or "out" for a fade-out effect.
1485 @var{start_frame} specifies the number of the start frame for starting
1486 to apply the fade effect.
1488 @var{nb_frames} specifies the number of frames for which the fade
1489 effect has to last. At the end of the fade-in effect the output video
1490 will have the same intensity as the input video, at the end of the
1491 fade-out transition the output video will be completely black.
1493 @var{options} is an optional sequence of @var{key}=@var{value} pairs,
1494 separated by ":". The description of the accepted options follows.
1501 @item start_frame, s
1502 See @var{start_frame}.
1505 See @var{nb_frames}.
1508 If set to 1, fade only alpha channel, if one exists on the input.
1512 A few usage examples follow, usable too as test scenarios.
1514 # fade in first 30 frames of video
1517 # fade out last 45 frames of a 200-frame video
1520 # fade in first 25 frames and fade out last 25 frames of a 1000-frame video
1521 fade=in:0:25, fade=out:975:25
1523 # make first 5 frames black, then fade in from frame 5-24
1526 # fade in alpha over first 25 frames of video
1527 fade=in:0:25:alpha=1
1532 Transform the field order of the input video.
1534 It accepts one parameter which specifies the required field order that
1535 the input interlaced video will be transformed to. The parameter can
1536 assume one of the following values:
1540 output bottom field first
1542 output top field first
1545 Default value is "tff".
1547 Transformation is achieved by shifting the picture content up or down
1548 by one line, and filling the remaining line with appropriate picture content.
1549 This method is consistent with most broadcast field order converters.
1551 If the input video is not flagged as being interlaced, or it is already
1552 flagged as being of the required output field order then this filter does
1553 not alter the incoming video.
1555 This filter is very useful when converting to or from PAL DV material,
1556 which is bottom field first.
1560 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
1565 Buffer input images and send them when they are requested.
1567 This filter is mainly useful when auto-inserted by the libavfilter
1570 The filter does not take parameters.
1574 Convert the input video to one of the specified pixel formats.
1575 Libavfilter will try to pick one that is supported for the input to
1578 The filter accepts a list of pixel format names, separated by ":",
1579 for example "yuv420p:monow:rgb24".
1581 Some examples follow:
1583 # convert the input video to the format "yuv420p"
1586 # convert the input video to any of the formats in the list
1587 format=yuv420p:yuv444p:yuv410p
1593 Apply a frei0r effect to the input video.
1595 To enable compilation of this filter you need to install the frei0r
1596 header and configure FFmpeg with @code{--enable-frei0r}.
1598 The filter supports the syntax:
1600 @var{filter_name}[@{:|=@}@var{param1}:@var{param2}:...:@var{paramN}]
1603 @var{filter_name} is the name to the frei0r effect to load. If the
1604 environment variable @env{FREI0R_PATH} is defined, the frei0r effect
1605 is searched in each one of the directories specified by the colon
1606 separated list in @env{FREIOR_PATH}, otherwise in the standard frei0r
1607 paths, which are in this order: @file{HOME/.frei0r-1/lib/},
1608 @file{/usr/local/lib/frei0r-1/}, @file{/usr/lib/frei0r-1/}.
1610 @var{param1}, @var{param2}, ... , @var{paramN} specify the parameters
1611 for the frei0r effect.
1613 A frei0r effect parameter can be a boolean (whose values are specified
1614 with "y" and "n"), a double, a color (specified by the syntax
1615 @var{R}/@var{G}/@var{B}, @var{R}, @var{G}, and @var{B} being float
1616 numbers from 0.0 to 1.0) or by an @code{av_parse_color()} color
1617 description), a position (specified by the syntax @var{X}/@var{Y},
1618 @var{X} and @var{Y} being float numbers) and a string.
1620 The number and kind of parameters depend on the loaded effect. If an
1621 effect parameter is not specified the default value is set.
1623 Some examples follow:
1625 # apply the distort0r effect, set the first two double parameters
1626 frei0r=distort0r:0.5:0.01
1628 # apply the colordistance effect, takes a color as first parameter
1629 frei0r=colordistance:0.2/0.3/0.4
1630 frei0r=colordistance:violet
1631 frei0r=colordistance:0x112233
1633 # apply the perspective effect, specify the top left and top right
1635 frei0r=perspective:0.2/0.2:0.8/0.2
1638 For more information see:
1639 @url{http://piksel.org/frei0r}
1643 Fix the banding artifacts that are sometimes introduced into nearly flat
1644 regions by truncation to 8bit color depth.
1645 Interpolate the gradients that should go where the bands are, and
1648 This filter is designed for playback only. Do not use it prior to
1649 lossy compression, because compression tends to lose the dither and
1650 bring back the bands.
1652 The filter takes two optional parameters, separated by ':':
1653 @var{strength}:@var{radius}
1655 @var{strength} is the maximum amount by which the filter will change
1656 any one pixel. Also the threshold for detecting nearly flat
1657 regions. Acceptable values range from .51 to 255, default value is
1658 1.2, out-of-range values will be clipped to the valid range.
1660 @var{radius} is the neighborhood to fit the gradient to. A larger
1661 radius makes for smoother gradients, but also prevents the filter from
1662 modifying the pixels near detailed regions. Acceptable values are
1663 8-32, default value is 16, out-of-range values will be clipped to the
1667 # default parameters
1676 Flip the input video horizontally.
1678 For example to horizontally flip the input video with @command{ffmpeg}:
1680 ffmpeg -i in.avi -vf "hflip" out.avi
1685 High precision/quality 3d denoise filter. This filter aims to reduce
1686 image noise producing smooth images and making still images really
1687 still. It should enhance compressibility.
1689 It accepts the following optional parameters:
1690 @var{luma_spatial}:@var{chroma_spatial}:@var{luma_tmp}:@var{chroma_tmp}
1694 a non-negative float number which specifies spatial luma strength,
1697 @item chroma_spatial
1698 a non-negative float number which specifies spatial chroma strength,
1699 defaults to 3.0*@var{luma_spatial}/4.0
1702 a float number which specifies luma temporal strength, defaults to
1703 6.0*@var{luma_spatial}/4.0
1706 a float number which specifies chroma temporal strength, defaults to
1707 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}
1712 Interlaceing detect filter. This filter tries to detect if the input is
1713 interlaced or progressive. Top or bottom field first.
1715 @section lut, lutrgb, lutyuv
1717 Compute a look-up table for binding each pixel component input value
1718 to an output value, and apply it to input video.
1720 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
1721 to an RGB input video.
1723 These filters accept in input a ":"-separated list of options, which
1724 specify the expressions used for computing the lookup table for the
1725 corresponding pixel component values.
1727 The @var{lut} filter requires either YUV or RGB pixel formats in
1728 input, and accepts the options:
1731 first pixel component
1733 second pixel component
1735 third pixel component
1737 fourth pixel component, corresponds to the alpha component
1740 The exact component associated to each option depends on the format in
1743 The @var{lutrgb} filter requires RGB pixel formats in input, and
1744 accepts the options:
1756 The @var{lutyuv} filter requires YUV pixel formats in input, and
1757 accepts the options:
1760 Y/luminance component
1769 The expressions can contain the following constants and functions:
1773 the input width and height
1776 input value for the pixel component
1779 the input value clipped in the @var{minval}-@var{maxval} range
1782 maximum value for the pixel component
1785 minimum value for the pixel component
1788 the negated value for the pixel component value clipped in the
1789 @var{minval}-@var{maxval} range , it corresponds to the expression
1790 "maxval-clipval+minval"
1793 the computed value in @var{val} clipped in the
1794 @var{minval}-@var{maxval} range
1796 @item gammaval(gamma)
1797 the computed gamma correction value of the pixel component value
1798 clipped in the @var{minval}-@var{maxval} range, corresponds to the
1800 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
1804 All expressions default to "val".
1806 Some examples follow:
1808 # negate input video
1809 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
1810 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
1812 # the above is the same as
1813 lutrgb="r=negval:g=negval:b=negval"
1814 lutyuv="y=negval:u=negval:v=negval"
1819 # remove chroma components, turns the video into a graytone image
1820 lutyuv="u=128:v=128"
1822 # apply a luma burning effect
1825 # remove green and blue components
1828 # set a constant alpha channel value on input
1829 format=rgba,lutrgb=a="maxval-minval/2"
1831 # correct luminance gamma by a 0.5 factor
1832 lutyuv=y=gammaval(0.5)
1837 Apply an MPlayer filter to the input video.
1839 This filter provides a wrapper around most of the filters of
1842 This wrapper is considered experimental. Some of the wrapped filters
1843 may not work properly and we may drop support for them, as they will
1844 be implemented natively into FFmpeg. Thus you should avoid
1845 depending on them when writing portable scripts.
1847 The filters accepts the parameters:
1848 @var{filter_name}[:=]@var{filter_params}
1850 @var{filter_name} is the name of a supported MPlayer filter,
1851 @var{filter_params} is a string containing the parameters accepted by
1854 The list of the currently supported filters follows:
1907 The parameter syntax and behavior for the listed filters are the same
1908 of the corresponding MPlayer filters. For detailed instructions check
1909 the "VIDEO FILTERS" section in the MPlayer manual.
1911 Some examples follow:
1913 # remove a logo by interpolating the surrounding pixels
1914 mp=delogo=200:200:80:20:1
1916 # adjust gamma, brightness, contrast
1919 # tweak hue and saturation
1923 See also mplayer(1), @url{http://www.mplayerhq.hu/}.
1929 This filter accepts an integer in input, if non-zero it negates the
1930 alpha component (if available). The default value in input is 0.
1934 Force libavfilter not to use any of the specified pixel formats for the
1935 input to the next filter.
1937 The filter accepts a list of pixel format names, separated by ":",
1938 for example "yuv420p:monow:rgb24".
1940 Some examples follow:
1942 # force libavfilter to use a format different from "yuv420p" for the
1943 # input to the vflip filter
1944 noformat=yuv420p,vflip
1946 # convert the input video to any of the formats not contained in the list
1947 noformat=yuv420p:yuv444p:yuv410p
1952 Pass the video source unchanged to the output.
1956 Apply video transform using libopencv.
1958 To enable this filter install libopencv library and headers and
1959 configure FFmpeg with @code{--enable-libopencv}.
1961 The filter takes the parameters: @var{filter_name}@{:=@}@var{filter_params}.
1963 @var{filter_name} is the name of the libopencv filter to apply.
1965 @var{filter_params} specifies the parameters to pass to the libopencv
1966 filter. If not specified the default values are assumed.
1968 Refer to the official libopencv documentation for more precise
1970 @url{http://opencv.willowgarage.com/documentation/c/image_filtering.html}
1972 Follows the list of supported libopencv filters.
1977 Dilate an image by using a specific structuring element.
1978 This filter corresponds to the libopencv function @code{cvDilate}.
1980 It accepts the parameters: @var{struct_el}:@var{nb_iterations}.
1982 @var{struct_el} represents a structuring element, and has the syntax:
1983 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
1985 @var{cols} and @var{rows} represent the number of columns and rows of
1986 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
1987 point, and @var{shape} the shape for the structuring element, and
1988 can be one of the values "rect", "cross", "ellipse", "custom".
1990 If the value for @var{shape} is "custom", it must be followed by a
1991 string of the form "=@var{filename}". The file with name
1992 @var{filename} is assumed to represent a binary image, with each
1993 printable character corresponding to a bright pixel. When a custom
1994 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
1995 or columns and rows of the read file are assumed instead.
1997 The default value for @var{struct_el} is "3x3+0x0/rect".
1999 @var{nb_iterations} specifies the number of times the transform is
2000 applied to the image, and defaults to 1.
2002 Follow some example:
2004 # use the default values
2007 # dilate using a structuring element with a 5x5 cross, iterate two times
2008 ocv=dilate=5x5+2x2/cross:2
2010 # read the shape from the file diamond.shape, iterate two times
2011 # the file diamond.shape may contain a pattern of characters like this:
2017 # the specified cols and rows are ignored (but not the anchor point coordinates)
2018 ocv=0x0+2x2/custom=diamond.shape:2
2023 Erode an image by using a specific structuring element.
2024 This filter corresponds to the libopencv function @code{cvErode}.
2026 The filter accepts the parameters: @var{struct_el}:@var{nb_iterations},
2027 with the same syntax and semantics as the @ref{dilate} filter.
2031 Smooth the input video.
2033 The filter takes the following parameters:
2034 @var{type}:@var{param1}:@var{param2}:@var{param3}:@var{param4}.
2036 @var{type} is the type of smooth filter to apply, and can be one of
2037 the following values: "blur", "blur_no_scale", "median", "gaussian",
2038 "bilateral". The default value is "gaussian".
2040 @var{param1}, @var{param2}, @var{param3}, and @var{param4} are
2041 parameters whose meanings depend on smooth type. @var{param1} and
2042 @var{param2} accept integer positive values or 0, @var{param3} and
2043 @var{param4} accept float values.
2045 The default value for @var{param1} is 3, the default value for the
2046 other parameters is 0.
2048 These parameters correspond to the parameters assigned to the
2049 libopencv function @code{cvSmooth}.
2054 Overlay one video on top of another.
2056 It takes two inputs and one output, the first input is the "main"
2057 video on which the second input is overlayed.
2059 It accepts the parameters: @var{x}:@var{y}[:@var{options}].
2061 @var{x} is the x coordinate of the overlayed video on the main video,
2062 @var{y} is the y coordinate. @var{x} and @var{y} are expressions containing
2063 the following parameters:
2066 @item main_w, main_h
2067 main input width and height
2070 same as @var{main_w} and @var{main_h}
2072 @item overlay_w, overlay_h
2073 overlay input width and height
2076 same as @var{overlay_w} and @var{overlay_h}
2079 @var{options} is an optional list of @var{key}=@var{value} pairs,
2082 The description of the accepted options follows.
2086 If set to 1, force the filter to accept inputs in the RGB
2087 color space. Default value is 0.
2090 Be aware that frames are taken from each input video in timestamp
2091 order, hence, if their initial timestamps differ, it is a a good idea
2092 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
2093 have them begin in the same zero timestamp, as it does the example for
2094 the @var{movie} filter.
2096 Follow some examples:
2098 # draw the overlay at 10 pixels from the bottom right
2099 # corner of the main video.
2100 overlay=main_w-overlay_w-10:main_h-overlay_h-10
2102 # insert a transparent PNG logo in the bottom left corner of the input
2103 movie=logo.png [logo];
2104 [in][logo] overlay=10:main_h-overlay_h-10 [out]
2106 # insert 2 different transparent PNG logos (second logo on bottom
2108 movie=logo1.png [logo1];
2109 movie=logo2.png [logo2];
2110 [in][logo1] overlay=10:H-h-10 [in+logo1];
2111 [in+logo1][logo2] overlay=W-w-10:H-h-10 [out]
2113 # add a transparent color layer on top of the main video,
2114 # WxH specifies the size of the main input to the overlay filter
2115 color=red@.3:WxH [over]; [in][over] overlay [out]
2118 You can chain together more overlays but the efficiency of such
2119 approach is yet to be tested.
2123 Add paddings to the input image, and places the original input at the
2124 given coordinates @var{x}, @var{y}.
2126 It accepts the following parameters:
2127 @var{width}:@var{height}:@var{x}:@var{y}:@var{color}.
2129 The parameters @var{width}, @var{height}, @var{x}, and @var{y} are
2130 expressions containing the following constants:
2134 the input video width and height
2137 same as @var{in_w} and @var{in_h}
2140 the output width and height, that is the size of the padded area as
2141 specified by the @var{width} and @var{height} expressions
2144 same as @var{out_w} and @var{out_h}
2147 x and y offsets as specified by the @var{x} and @var{y}
2148 expressions, or NAN if not yet specified
2151 same as @var{iw} / @var{ih}
2154 input sample aspect ratio
2157 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
2160 horizontal and vertical chroma subsample values. For example for the
2161 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2164 Follows the description of the accepted parameters.
2169 Specify the size of the output image with the paddings added. If the
2170 value for @var{width} or @var{height} is 0, the corresponding input size
2171 is used for the output.
2173 The @var{width} expression can reference the value set by the
2174 @var{height} expression, and vice versa.
2176 The default value of @var{width} and @var{height} is 0.
2180 Specify the offsets where to place the input image in the padded area
2181 with respect to the top/left border of the output image.
2183 The @var{x} expression can reference the value set by the @var{y}
2184 expression, and vice versa.
2186 The default value of @var{x} and @var{y} is 0.
2190 Specify the color of the padded area, it can be the name of a color
2191 (case insensitive match) or a 0xRRGGBB[AA] sequence.
2193 The default value of @var{color} is "black".
2197 Some examples follow:
2200 # Add paddings with color "violet" to the input video. Output video
2201 # size is 640x480, the top-left corner of the input video is placed at
2203 pad=640:480:0:40:violet
2205 # pad the input to get an output with dimensions increased bt 3/2,
2206 # and put the input video at the center of the padded area
2207 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
2209 # pad the input to get a squared output with size equal to the maximum
2210 # value between the input width and height, and put the input video at
2211 # the center of the padded area
2212 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
2214 # pad the input to get a final w/h ratio of 16:9
2215 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
2217 # for anamorphic video, in order to set the output display aspect ratio,
2218 # it is necessary to use sar in the expression, according to the relation:
2219 # (ih * X / ih) * sar = output_dar
2220 # X = output_dar / sar
2221 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
2223 # double output size and put the input video in the bottom-right
2224 # corner of the output padded area
2225 pad="2*iw:2*ih:ow-iw:oh-ih"
2228 @section pixdesctest
2230 Pixel format descriptor test filter, mainly useful for internal
2231 testing. The output video should be equal to the input video.
2235 format=monow, pixdesctest
2238 can be used to test the monowhite pixel format descriptor definition.
2242 Scale the input video to @var{width}:@var{height}[:@var{interl}=@{1|-1@}] and/or convert the image format.
2244 The scale filter forces the output display aspect ratio to be the same
2245 of the input, by changing the output sample aspect ratio.
2247 The parameters @var{width} and @var{height} are expressions containing
2248 the following constants:
2252 the input width and height
2255 same as @var{in_w} and @var{in_h}
2258 the output (cropped) width and height
2261 same as @var{out_w} and @var{out_h}
2264 same as @var{iw} / @var{ih}
2267 input sample aspect ratio
2270 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
2273 horizontal and vertical chroma subsample values. For example for the
2274 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
2277 If the input image format is different from the format requested by
2278 the next filter, the scale filter will convert the input to the
2281 If the value for @var{width} or @var{height} is 0, the respective input
2282 size is used for the output.
2284 If the value for @var{width} or @var{height} is -1, the scale filter will
2285 use, for the respective output size, a value that maintains the aspect
2286 ratio of the input image.
2288 The default value of @var{width} and @var{height} is 0.
2290 Valid values for the optional parameter @var{interl} are:
2294 force interlaced aware scaling
2297 select interlaced aware scaling depending on whether the source frames
2298 are flagged as interlaced or not
2301 Unless @var{interl} is set to one of the above options, interlaced scaling will not be used.
2303 Some examples follow:
2305 # scale the input video to a size of 200x100.
2308 # scale the input to 2x
2310 # the above is the same as
2313 # scale the input to 2x with forced interlaced scaling
2314 scale=2*iw:2*ih:interl=1
2316 # scale the input to half size
2319 # increase the width, and set the height to the same size
2322 # seek for Greek harmony
2326 # increase the height, and set the width to 3/2 of the height
2329 # increase the size, but make the size a multiple of the chroma
2330 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
2332 # increase the width to a maximum of 500 pixels, keep the same input aspect ratio
2333 scale='min(500\, iw*3/2):-1'
2337 Select frames to pass in output.
2339 It accepts in input an expression, which is evaluated for each input
2340 frame. If the expression is evaluated to a non-zero value, the frame
2341 is selected and passed to the output, otherwise it is discarded.
2343 The expression can contain the following constants:
2347 the sequential number of the filtered frame, starting from 0
2350 the sequential number of the selected frame, starting from 0
2352 @item prev_selected_n
2353 the sequential number of the last selected frame, NAN if undefined
2356 timebase of the input timestamps
2359 the PTS (Presentation TimeStamp) of the filtered video frame,
2360 expressed in @var{TB} units, NAN if undefined
2363 the PTS (Presentation TimeStamp) of the filtered video frame,
2364 expressed in seconds, NAN if undefined
2367 the PTS of the previously filtered video frame, NAN if undefined
2369 @item prev_selected_pts
2370 the PTS of the last previously filtered video frame, NAN if undefined
2372 @item prev_selected_t
2373 the PTS of the last previously selected video frame, NAN if undefined
2376 the PTS of the first video frame in the video, NAN if undefined
2379 the time of the first video frame in the video, NAN if undefined
2382 the type of the filtered frame, can assume one of the following
2394 @item interlace_type
2395 the frame interlace type, can assume one of the following values:
2398 the frame is progressive (not interlaced)
2400 the frame is top-field-first
2402 the frame is bottom-field-first
2406 1 if the filtered frame is a key-frame, 0 otherwise
2409 the position in the file of the filtered frame, -1 if the information
2410 is not available (e.g. for synthetic video)
2413 The default value of the select expression is "1".
2415 Some examples follow:
2418 # select all frames in input
2421 # the above is the same as:
2427 # select only I-frames
2428 select='eq(pict_type\,I)'
2430 # select one frame every 100
2431 select='not(mod(n\,100))'
2433 # select only frames contained in the 10-20 time interval
2434 select='gte(t\,10)*lte(t\,20)'
2436 # select only I frames contained in the 10-20 time interval
2437 select='gte(t\,10)*lte(t\,20)*eq(pict_type\,I)'
2439 # select frames with a minimum distance of 10 seconds
2440 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
2443 @section setdar, setsar
2445 The @code{setdar} filter sets the Display Aspect Ratio for the filter
2448 This is done by changing the specified Sample (aka Pixel) Aspect
2449 Ratio, according to the following equation:
2451 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
2454 Keep in mind that the @code{setdar} filter does not modify the pixel
2455 dimensions of the video frame. Also the display aspect ratio set by
2456 this filter may be changed by later filters in the filterchain,
2457 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
2460 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
2461 the filter output video.
2463 Note that as a consequence of the application of this filter, the
2464 output display aspect ratio will change according to the equation
2467 Keep in mind that the sample aspect ratio set by the @code{setsar}
2468 filter may be changed by later filters in the filterchain, e.g. if
2469 another "setsar" or a "setdar" filter is applied.
2471 The @code{setdar} and @code{setsar} filters accept a parameter string
2472 which represents the wanted aspect ratio. The parameter can
2473 be a floating point number string, an expression, or a string of the form
2474 @var{num}:@var{den}, where @var{num} and @var{den} are the numerator
2475 and denominator of the aspect ratio. If the parameter is not
2476 specified, it is assumed the value "0:1".
2478 For example to change the display aspect ratio to 16:9, specify:
2483 The example above is equivalent to:
2488 To change the sample aspect ratio to 10:11, specify:
2495 Force field for the output video frame.
2497 The @code{setfield} filter marks the interlace type field for the
2498 output frames. It does not change the input frame, but only sets the
2499 corresponding property, which affects how the frame is treated by
2500 followig filters (e.g. @code{fieldorder} or @code{yadif}).
2502 It accepts a parameter representing an integer or a string, which can
2503 assume the following values:
2506 Keep the same field property.
2509 Mark the frame as bottom-field-first.
2512 Mark the frame as top-field-first.
2517 Change the PTS (presentation timestamp) of the input video frames.
2519 Accept in input an expression evaluated through the eval API, which
2520 can contain the following constants:
2524 the presentation timestamp in input
2527 the count of the input frame, starting from 0.
2530 the PTS of the first video frame
2533 tell if the current frame is interlaced
2536 original position in the file of the frame, or undefined if undefined
2537 for the current frame
2547 Some examples follow:
2550 # start counting PTS from zero
2562 # fixed rate 25 fps with some jitter
2563 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
2568 Set the timebase to use for the output frames timestamps.
2569 It is mainly useful for testing timebase configuration.
2571 It accepts in input an arithmetic expression representing a rational.
2572 The expression can contain the constants "AVTB" (the
2573 default timebase), and "intb" (the input timebase).
2575 The default value for the input is "intb".
2577 Follow some examples.
2580 # set the timebase to 1/25
2583 # set the timebase to 1/10
2586 #set the timebase to 1001/1000
2589 #set the timebase to 2*intb
2592 #set the default timebase value
2598 Show a line containing various information for each input video frame.
2599 The input video is not modified.
2601 The shown line contains a sequence of key/value pairs of the form
2602 @var{key}:@var{value}.
2604 A description of each shown parameter follows:
2608 sequential number of the input frame, starting from 0
2611 Presentation TimeStamp of the input frame, expressed as a number of
2612 time base units. The time base unit depends on the filter input pad.
2615 Presentation TimeStamp of the input frame, expressed as a number of
2619 position of the frame in the input stream, -1 if this information in
2620 unavailable and/or meaningless (for example in case of synthetic video)
2626 sample aspect ratio of the input frame, expressed in the form
2630 size of the input frame, expressed in the form
2631 @var{width}x@var{height}
2634 interlaced mode ("P" for "progressive", "T" for top field first, "B"
2635 for bottom field first)
2638 1 if the frame is a key frame, 0 otherwise
2641 picture type of the input frame ("I" for an I-frame, "P" for a
2642 P-frame, "B" for a B-frame, "?" for unknown type).
2643 Check also the documentation of the @code{AVPictureType} enum and of
2644 the @code{av_get_picture_type_char} function defined in
2645 @file{libavutil/avutil.h}.
2648 Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame
2650 @item plane_checksum
2651 Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
2652 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]"
2657 Pass the images of input video on to next video filter as multiple
2661 ffmpeg -i in.avi -vf "slicify=32" out.avi
2664 The filter accepts the slice height as parameter. If the parameter is
2665 not specified it will use the default value of 16.
2667 Adding this in the beginning of filter chains should make filtering
2668 faster due to better use of the memory cache.
2672 Pass on the input video to two outputs. Both outputs are identical to
2677 [in] split [splitout1][splitout2];
2678 [splitout1] crop=100:100:0:0 [cropout];
2679 [splitout2] pad=200:200:100:100 [padout];
2682 will create two separate outputs from the same input, one cropped and
2689 Select the most representative frame in a given sequence of consecutive frames.
2691 It accepts as argument the frames batch size to analyze (default @var{N}=100);
2692 in a set of @var{N} frames, the filter will pick one of them, and then handle
2693 the next batch of @var{N} frames until the end.
2695 Since the filter keeps track of the whole frames sequence, a bigger @var{N}
2696 value will result in a higher memory usage, so a high value is not recommended.
2698 The following example extract one picture each 50 frames:
2703 Complete example of a thumbnail creation with @command{ffmpeg}:
2705 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
2710 Tile several successive frames together.
2712 It accepts as argument the tile size (i.e. the number of lines and columns)
2713 in the form "@var{w}x@var{h}".
2715 For example, produce 8×8 PNG tiles of all keyframes (@option{-skip_frame
2718 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
2720 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
2721 duplicating each output frame to accomodate the originally detected frame
2726 Perform various types of temporal field interlacing.
2728 Frames are counted starting from 1, so the first input frame is
2731 This filter accepts a single parameter specifying the mode. Available
2736 Move odd frames into the upper field, even into the lower field,
2737 generating a double height frame at half framerate.
2740 Only output even frames, odd frames are dropped, generating a frame with
2741 unchanged height at half framerate.
2744 Only output odd frames, even frames are dropped, generating a frame with
2745 unchanged height at half framerate.
2748 Expand each frame to full height, but pad alternate lines with black,
2749 generating a frame with double height at the same input framerate.
2752 Interleave the upper field from odd frames with the lower field from
2753 even frames, generating a frame with unchanged height at half framerate.
2756 Interleave the lower field from odd frames with the upper field from
2757 even frames, generating a frame with unchanged height at half framerate.
2764 Transpose rows with columns in the input video and optionally flip it.
2766 It accepts a parameter representing an integer, which can assume the
2771 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
2779 Rotate by 90 degrees clockwise, that is:
2787 Rotate by 90 degrees counterclockwise, that is:
2795 Rotate by 90 degrees clockwise and vertically flip, that is:
2805 Sharpen or blur the input video.
2807 It accepts the following parameters:
2808 @var{luma_msize_x}:@var{luma_msize_y}:@var{luma_amount}:@var{chroma_msize_x}:@var{chroma_msize_y}:@var{chroma_amount}
2810 Negative values for the amount will blur the input video, while positive
2811 values will sharpen. All parameters are optional and default to the
2812 equivalent of the string '5:5:1.0:5:5:0.0'.
2817 Set the luma matrix horizontal size. It can be an integer between 3
2818 and 13, default value is 5.
2821 Set the luma matrix vertical size. It can be an integer between 3
2822 and 13, default value is 5.
2825 Set the luma effect strength. It can be a float number between -2.0
2826 and 5.0, default value is 1.0.
2828 @item chroma_msize_x
2829 Set the chroma matrix horizontal size. It can be an integer between 3
2830 and 13, default value is 5.
2832 @item chroma_msize_y
2833 Set the chroma matrix vertical size. It can be an integer between 3
2834 and 13, default value is 5.
2837 Set the chroma effect strength. It can be a float number between -2.0
2838 and 5.0, default value is 0.0.
2843 # Strong luma sharpen effect parameters
2846 # Strong blur of both luma and chroma parameters
2847 unsharp=7:7:-2:7:7:-2
2849 # Use the default values with @command{ffmpeg}
2850 ffmpeg -i in.avi -vf "unsharp" out.mp4
2855 Flip the input video vertically.
2858 ffmpeg -i in.avi -vf "vflip" out.avi
2863 Deinterlace the input video ("yadif" means "yet another deinterlacing
2866 It accepts the optional parameters: @var{mode}:@var{parity}:@var{auto}.
2868 @var{mode} specifies the interlacing mode to adopt, accepts one of the
2873 output 1 frame for each frame
2875 output 1 frame for each field
2877 like 0 but skips spatial interlacing check
2879 like 1 but skips spatial interlacing check
2884 @var{parity} specifies the picture field parity assumed for the input
2885 interlaced video, accepts one of the following values:
2889 assume top field first
2891 assume bottom field first
2893 enable automatic detection
2896 Default value is -1.
2897 If interlacing is unknown or decoder does not export this information,
2898 top field first will be assumed.
2900 @var{auto} specifies if deinterlacer should trust the interlaced flag
2901 and only deinterlace frames marked as interlaced
2905 deinterlace all frames
2907 only deinterlace frames marked as interlaced
2912 @c man end VIDEO FILTERS
2914 @chapter Video Sources
2915 @c man begin VIDEO SOURCES
2917 Below is a description of the currently available video sources.
2921 Buffer video frames, and make them available to the filter chain.
2923 This source is mainly intended for a programmatic use, in particular
2924 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
2926 It accepts the following parameters:
2927 @var{width}:@var{height}:@var{pix_fmt_string}:@var{timebase_num}:@var{timebase_den}:@var{sample_aspect_ratio_num}:@var{sample_aspect_ratio.den}:@var{scale_params}
2929 All the parameters but @var{scale_params} need to be explicitly
2932 Follows the list of the accepted parameters.
2937 Specify the width and height of the buffered video frames.
2939 @item pix_fmt_string
2940 A string representing the pixel format of the buffered video frames.
2941 It may be a number corresponding to a pixel format, or a pixel format
2944 @item timebase_num, timebase_den
2945 Specify numerator and denomitor of the timebase assumed by the
2946 timestamps of the buffered frames.
2948 @item sample_aspect_ratio.num, sample_aspect_ratio.den
2949 Specify numerator and denominator of the sample aspect ratio assumed
2950 by the video frames.
2953 Specify the optional parameters to be used for the scale filter which
2954 is automatically inserted when an input change is detected in the
2955 input size or format.
2960 buffer=320:240:yuv410p:1:24:1:1
2963 will instruct the source to accept video frames with size 320x240 and
2964 with format "yuv410p", assuming 1/24 as the timestamps timebase and
2965 square pixels (1:1 sample aspect ratio).
2966 Since the pixel format with name "yuv410p" corresponds to the number 6
2967 (check the enum PixelFormat definition in @file{libavutil/pixfmt.h}),
2968 this example corresponds to:
2970 buffer=320:240:6:1:24:1:1
2975 Create a pattern generated by an elementary cellular automaton.
2977 The initial state of the cellular automaton can be defined through the
2978 @option{filename}, and @option{pattern} options. If such options are
2979 not specified an initial state is created randomly.
2981 At each new frame a new row in the video is filled with the result of
2982 the cellular automaton next generation. The behavior when the whole
2983 frame is filled is defined by the @option{scroll} option.
2985 This source accepts a list of options in the form of
2986 @var{key}=@var{value} pairs separated by ":". A description of the
2987 accepted options follows.
2991 Read the initial cellular automaton state, i.e. the starting row, from
2993 In the file, each non-whitespace character is considered an alive
2994 cell, a newline will terminate the row, and further characters in the
2995 file will be ignored.
2998 Read the initial cellular automaton state, i.e. the starting row, from
2999 the specified string.
3001 Each non-whitespace character in the string is considered an alive
3002 cell, a newline will terminate the row, and further characters in the
3003 string will be ignored.
3006 Set the video rate, that is the number of frames generated per second.
3009 @item random_fill_ratio, ratio
3010 Set the random fill ratio for the initial cellular automaton row. It
3011 is a floating point number value ranging from 0 to 1, defaults to
3014 This option is ignored when a file or a pattern is specified.
3016 @item random_seed, seed
3017 Set the seed for filling randomly the initial row, must be an integer
3018 included between 0 and UINT32_MAX. If not specified, or if explicitly
3019 set to -1, the filter will try to use a good random seed on a best
3023 Set the cellular automaton rule, it is a number ranging from 0 to 255.
3024 Default value is 110.
3027 Set the size of the output video.
3029 If @option{filename} or @option{pattern} is specified, the size is set
3030 by default to the width of the specified initial state row, and the
3031 height is set to @var{width} * PHI.
3033 If @option{size} is set, it must contain the width of the specified
3034 pattern string, and the specified pattern will be centered in the
3037 If a filename or a pattern string is not specified, the size value
3038 defaults to "320x518" (used for a randomly generated initial state).
3041 If set to 1, scroll the output upward when all the rows in the output
3042 have been already filled. If set to 0, the new generated row will be
3043 written over the top row just after the bottom row is filled.
3046 @item start_full, full
3047 If set to 1, completely fill the output with generated rows before
3048 outputting the first frame.
3049 This is the default behavior, for disabling set the value to 0.
3052 If set to 1, stitch the left and right row edges together.
3053 This is the default behavior, for disabling set the value to 0.
3056 @subsection Examples
3060 Read the initial state from @file{pattern}, and specify an output of
3063 cellauto=f=pattern:s=200x400
3067 Generate a random initial row with a width of 200 cells, with a fill
3070 cellauto=ratio=2/3:s=200x200
3074 Create a pattern generated by rule 18 starting by a single alive cell
3075 centered on an initial row with width 100:
3077 cellauto=p=@@:s=100x400:full=0:rule=18
3081 Specify a more elaborated initial pattern:
3083 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
3090 Provide an uniformly colored input.
3092 It accepts the following parameters:
3093 @var{color}:@var{frame_size}:@var{frame_rate}
3095 Follows the description of the accepted parameters.
3100 Specify the color of the source. It can be the name of a color (case
3101 insensitive match) or a 0xRRGGBB[AA] sequence, possibly followed by an
3102 alpha specifier. The default value is "black".
3105 Specify the size of the sourced video, it may be a string of the form
3106 @var{width}x@var{height}, or the name of a size abbreviation. The
3107 default value is "320x240".
3110 Specify the frame rate of the sourced video, as the number of frames
3111 generated per second. It has to be a string in the format
3112 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
3113 number or a valid video frame rate abbreviation. The default value is
3118 For example the following graph description will generate a red source
3119 with an opacity of 0.2, with size "qcif" and a frame rate of 10
3120 frames per second, which will be overlayed over the source connected
3121 to the pad with identifier "in".
3124 "color=red@@0.2:qcif:10 [color]; [in][color] overlay [out]"
3129 Read a video stream from a movie container.
3131 It accepts the syntax: @var{movie_name}[:@var{options}] where
3132 @var{movie_name} is the name of the resource to read (not necessarily
3133 a file but also a device or a stream accessed through some protocol),
3134 and @var{options} is an optional sequence of @var{key}=@var{value}
3135 pairs, separated by ":".
3137 The description of the accepted options follows.
3141 @item format_name, f
3142 Specifies the format assumed for the movie to read, and can be either
3143 the name of a container or an input device. If not specified the
3144 format is guessed from @var{movie_name} or by probing.
3146 @item seek_point, sp
3147 Specifies the seek point in seconds, the frames will be output
3148 starting from this seek point, the parameter is evaluated with
3149 @code{av_strtod} so the numerical value may be suffixed by an IS
3150 postfix. Default value is "0".
3152 @item stream_index, si
3153 Specifies the index of the video stream to read. If the value is -1,
3154 the best suited video stream will be automatically selected. Default
3158 Specifies how many times to read the video stream in sequence.
3159 If the value is less than 1, the stream will be read again and again.
3160 Default value is "1".
3162 Note that when the movie is looped the source timestamps are not
3163 changed, so it will generate non monotonically increasing timestamps.
3166 This filter allows to overlay a second video on top of main input of
3167 a filtergraph as shown in this graph:
3169 input -----------> deltapts0 --> overlay --> output
3172 movie --> scale--> deltapts1 -------+
3175 Some examples follow:
3177 # skip 3.2 seconds from the start of the avi file in.avi, and overlay it
3178 # on top of the input labelled as "in".
3179 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [movie];
3180 [in] setpts=PTS-STARTPTS, [movie] overlay=16:16 [out]
3182 # read from a video4linux2 device, and overlay it on top of the input
3184 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [movie];
3185 [in] setpts=PTS-STARTPTS, [movie] overlay=16:16 [out]
3191 Generate various test patterns, as generated by the MPlayer test filter.
3193 The size of the generated video is fixed, and is 256x256.
3194 This source is useful in particular for testing encoding features.
3196 This source accepts an optional sequence of @var{key}=@var{value} pairs,
3197 separated by ":". The description of the accepted options follows.
3202 Specify the frame rate of the sourced video, as the number of frames
3203 generated per second. It has to be a string in the format
3204 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
3205 number or a valid video frame rate abbreviation. The default value is
3209 Set the video duration of the sourced video. The accepted syntax is:
3211 [-]HH[:MM[:SS[.m...]]]
3214 See also the function @code{av_parse_time()}.
3216 If not specified, or the expressed duration is negative, the video is
3217 supposed to be generated forever.
3221 Set the number or the name of the test to perform. Supported tests are:
3236 Default value is "all", which will cycle through the list of all tests.
3239 For example the following:
3244 will generate a "dc_luma" test pattern.
3248 Provide a frei0r source.
3250 To enable compilation of this filter you need to install the frei0r
3251 header and configure FFmpeg with @code{--enable-frei0r}.
3253 The source supports the syntax:
3255 @var{size}:@var{rate}:@var{src_name}[@{=|:@}@var{param1}:@var{param2}:...:@var{paramN}]
3258 @var{size} is the size of the video to generate, may be a string of the
3259 form @var{width}x@var{height} or a frame size abbreviation.
3260 @var{rate} is the rate of the video to generate, may be a string of
3261 the form @var{num}/@var{den} or a frame rate abbreviation.
3262 @var{src_name} is the name to the frei0r source to load. For more
3263 information regarding frei0r and how to set the parameters read the
3264 section @ref{frei0r} in the description of the video filters.
3266 Some examples follow:
3268 # generate a frei0r partik0l source with size 200x200 and frame rate 10
3269 # which is overlayed on the overlay filter main input
3270 frei0r_src=200x200:10:partik0l=1234 [overlay]; [in][overlay] overlay
3275 Generate a life pattern.
3277 This source is based on a generalization of John Conway's life game.
3279 The sourced input represents a life grid, each pixel represents a cell
3280 which can be in one of two possible states, alive or dead. Every cell
3281 interacts with its eight neighbours, which are the cells that are
3282 horizontally, vertically, or diagonally adjacent.
3284 At each interaction the grid evolves according to the adopted rule,
3285 which specifies the number of neighbor alive cells which will make a
3286 cell stay alive or born. The @option{rule} option allows to specify
3289 This source accepts a list of options in the form of
3290 @var{key}=@var{value} pairs separated by ":". A description of the
3291 accepted options follows.
3295 Set the file from which to read the initial grid state. In the file,
3296 each non-whitespace character is considered an alive cell, and newline
3297 is used to delimit the end of each row.
3299 If this option is not specified, the initial grid is generated
3303 Set the video rate, that is the number of frames generated per second.
3306 @item random_fill_ratio, ratio
3307 Set the random fill ratio for the initial random grid. It is a
3308 floating point number value ranging from 0 to 1, defaults to 1/PHI.
3309 It is ignored when a file is specified.
3311 @item random_seed, seed
3312 Set the seed for filling the initial random grid, must be an integer
3313 included between 0 and UINT32_MAX. If not specified, or if explicitly
3314 set to -1, the filter will try to use a good random seed on a best
3320 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
3321 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
3322 @var{NS} specifies the number of alive neighbor cells which make a
3323 live cell stay alive, and @var{NB} the number of alive neighbor cells
3324 which make a dead cell to become alive (i.e. to "born").
3325 "s" and "b" can be used in place of "S" and "B", respectively.
3327 Alternatively a rule can be specified by an 18-bits integer. The 9
3328 high order bits are used to encode the next cell state if it is alive
3329 for each number of neighbor alive cells, the low order bits specify
3330 the rule for "borning" new cells. Higher order bits encode for an
3331 higher number of neighbor cells.
3332 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
3333 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
3335 Default value is "S23/B3", which is the original Conway's game of life
3336 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
3337 cells, and will born a new cell if there are three alive cells around
3341 Set the size of the output video.
3343 If @option{filename} is specified, the size is set by default to the
3344 same size of the input file. If @option{size} is set, it must contain
3345 the size specified in the input file, and the initial grid defined in
3346 that file is centered in the larger resulting area.
3348 If a filename is not specified, the size value defaults to "320x240"
3349 (used for a randomly generated initial grid).
3352 If set to 1, stitch the left and right grid edges together, and the
3353 top and bottom edges also. Defaults to 1.
3356 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
3357 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
3358 value from 0 to 255.
3361 Set the color of living (or new born) cells.
3364 Set the color of dead cells. If @option{mold} is set, this is the first color
3365 used to represent a dead cell.
3368 Set mold color, for definitely dead and moldy cells.
3371 @subsection Examples
3375 Read a grid from @file{pattern}, and center it on a grid of size
3378 life=f=pattern:s=300x300
3382 Generate a random grid of size 200x200, with a fill ratio of 2/3:
3384 life=ratio=2/3:s=200x200
3388 Specify a custom rule for evolving a randomly generated grid:
3394 Full example with slow death effect (mold) using @command{ffplay}:
3396 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
3400 @section nullsrc, rgbtestsrc, testsrc
3402 The @code{nullsrc} source returns unprocessed video frames. It is
3403 mainly useful to be employed in analysis / debugging tools, or as the
3404 source for filters which ignore the input data.
3406 The @code{rgbtestsrc} source generates an RGB test pattern useful for
3407 detecting RGB vs BGR issues. You should see a red, green and blue
3408 stripe from top to bottom.
3410 The @code{testsrc} source generates a test video pattern, showing a
3411 color pattern, a scrolling gradient and a timestamp. This is mainly
3412 intended for testing purposes.
3414 These sources accept an optional sequence of @var{key}=@var{value} pairs,
3415 separated by ":". The description of the accepted options follows.
3420 Specify the size of the sourced video, it may be a string of the form
3421 @var{width}x@var{height}, or the name of a size abbreviation. The
3422 default value is "320x240".
3425 Specify the frame rate of the sourced video, as the number of frames
3426 generated per second. It has to be a string in the format
3427 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a float
3428 number or a valid video frame rate abbreviation. The default value is
3432 Set the sample aspect ratio of the sourced video.
3435 Set the video duration of the sourced video. The accepted syntax is:
3437 [-]HH[:MM[:SS[.m...]]]
3440 See also the function @code{av_parse_time()}.
3442 If not specified, or the expressed duration is negative, the video is
3443 supposed to be generated forever.
3446 Set the number of decimals to show in the timestamp, only used in the
3447 @code{testsrc} source.
3449 The displayed timestamp value will correspond to the original
3450 timestamp value multiplied by the power of 10 of the specified
3451 value. Default value is 0.
3454 For example the following:
3456 testsrc=duration=5.3:size=qcif:rate=10
3459 will generate a video with a duration of 5.3 seconds, with size
3460 176x144 and a frame rate of 10 frames per second.
3462 If the input content is to be ignored, @code{nullsrc} can be used. The
3463 following command generates noise in the luminance plane by employing
3464 the @code{mp=geq} filter:
3466 nullsrc=s=256x256, mp=geq=random(1)*255:128:128
3469 @c man end VIDEO SOURCES
3471 @chapter Video Sinks
3472 @c man begin VIDEO SINKS
3474 Below is a description of the currently available video sinks.
3478 Buffer video frames, and make them available to the end of the filter
3481 This sink is mainly intended for a programmatic use, in particular
3482 through the interface defined in @file{libavfilter/buffersink.h}.
3484 It does not require a string parameter in input, but you need to
3485 specify a pointer to a list of supported pixel formats terminated by
3486 -1 in the opaque parameter provided to @code{avfilter_init_filter}
3487 when initializing this sink.
3491 Null video sink, do absolutely nothing with the input video. It is
3492 mainly useful as a template and to be employed in analysis / debugging
3495 @c man end VIDEO SINKS