1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-02 03:06:28 +02:00
Commit Graph

7587 Commits

Author SHA1 Message Date
Ruiling Song
83f9da7768 avfilter/vf_gblur: add x86 SIMD optimizations
The horizontal pass get ~2x performance with the patch
under single thread.

Tested overall performance using the command(avx2 enabled):
./ffmpeg -i 1080p.mp4 -vf gblur -f null /dev/null
./ffmpeg -i 1080p.mp4 -vf gblur=threads=1 -f null /dev/null
For single thread, the fps improves from 43 to 60, about 40%.
For multi-thread, the fps improves from 110 to 130, about 20%.

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
2019-06-12 08:53:11 +08:00
Limin Wang
5fc8d87ba6 libavfilter/vf_cover_rect.c: free the allocated frame
Signed-off-by: Limin Wang <lance.lmwang@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
2019-06-11 23:53:26 +02:00
Paul B Mahol
80e68ce116 avfilter/vf_normalize: fix filtering of RGB0 formats 2019-06-10 23:11:57 +02:00
Zhong Li
e62f625163 lavf/qsvvpp: add P010 output format support
Signed-off-by: Zhong Li <zhong.li@intel.com>
2019-06-10 13:48:41 +08:00
Zhong Li
ef56c28b25 lavf/qsvvpp: add extra_hw_frames support
extra_hw_frames is needed for some cases.
(eg: qsv decoding + qsv vpp + qsv look_ahead encoding, transcoding
 failed if no extra_hw_frames supported:
 ffmpeg -hwaccel qsv -c:v h264_qsv -i bbb_sunflower_1080p_30fps_normal_2000frames.mp4 \
-vf vpp_qsv=w=1280:h=720:extra_hw_frames=100 -v verbose  -c:v h264_qsv \
-look_ahead 1 -look_ahead_depth 100 out1.mp4
)

Signed-off-by: Zhong Li <zhong.li@intel.com>
2019-06-10 13:48:06 +08:00
Paul B Mahol
a9bf656577 avfilter/vf_lut2: add slice threading 2019-06-09 14:07:00 +02:00
Mark Thompson
a53f9fde45 lavfi/vaapi: Fix build with libva 2.1 and 2.2
While the VAProcColorProperties structure was added in libva 2.1, the
colour primaries/transfer/space fields were not added until libva 2.3.
2019-06-07 21:15:29 +01:00
Xuewei Meng
78e1d7f421 libavfilter: Add derain filter
Remove the rain in the input image/video by applying the derain
methods based on convolutional neural networks. Training scripts
as well as scripts for model generation are provided in the
repository at https://github.com/XueweiMeng/derain_filter.git.

Signed-off-by: Xuewei Meng <xwmeng96@gmail.com>
2019-06-06 13:59:43 +08:00
Gyan Doshi
3be4490014 avfilter/framesync: fix shortest with eof_action=pass
Shifted check of shortest to after repeatlast,
to ensure shortest=1 is always honoured.
2019-06-06 10:19:14 +05:30
Paul B Mahol
d3f236b300 avilter/avf_showwaves: switch to activate in showwaves filter 2019-06-04 09:28:07 +02:00
Paul B Mahol
97ddc5d1f7 avfilter/af_anlmdn: add support for commands 2019-06-03 14:20:19 +02:00
Jun Zhao
0b7bfa8ad7 lavf/sr: Refine the coding style for init
We perfer the coding style like:

/* some stuff */
if (error) {
    /* error handling */
    return -(errorcode);
}
/* normal actions */
do_something()

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-06-03 13:04:23 +08:00
Jun Zhao
5c1fbc4239 lavf/sr: Don't need to check NULL before sws_freeContext
sws_freeContext have check the NULL pointer, so don't need to check
NULL before sws_freeContext.

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-06-03 13:04:23 +08:00
Jun Zhao
51b0e81216 lavf/sr: Dump input pixel format in error message
Dump input pixel format in error message, it's will help to debugging

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-06-03 13:04:23 +08:00
Zhong Li
165eabf19b lavf/qsv: use av_cold for init/uninit
Signed-off-by: Zhong Li <zhong.li@intel.com>
2019-06-03 05:39:06 +08:00
Ruiling Song
94ceeba9f9 avfilter/vf_unsharp: enable slice threading
benchmarking with a simple command:
ffmpeg -i 1080p.mp4 -vf unsharp=la=3:ca=3 -an -f null /dev/null
with the patch, the fps increase from 50 to 120 on my local machine (i7-6770HQ).

Signed-off-by: Ruiling Song <ruiling.song@intel.com>
2019-06-03 10:49:57 +08:00
Jun Zhao
bbad0bc5ff lavfi/lut: Add slice threading support
Used the command for 1080p h264 clip as follow:

a). ffmpeg -i input -vf lutyuv="u=128:v=128" -f null /dev/null
b). ffmpeg -i input -vf lutrgb="g=0:b=0" -f null /dev/null

after enabled the slice threading, the fps change from:

a). 144fps to 258fps (lutyuv)
b). 94fps  to 153fps (lutrgb)

in Intel(R) Core(TM) i5-8265U CPU @ 1.60GHz

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-06-03 09:38:28 +08:00
Jun Zhao
360bee8ca4 lavfi/colorlevels: Add slice threading support
Add slice threading support, use the command like:

./ffmpeg -i input -vf colorlevels -f null /dev/null

with 1080p h264 clip, the fps from 39 fps to 79 fps
in the local(Intel(R) Core(TM) i5-8265U CPU @ 1.60GHz)

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-06-03 09:36:14 +08:00
Mark Thompson
ef2f89bbcc vf_scale_vaapi: Add options to configure output colour properties
The "out_color_matrix" and "out_range" properties match the same options
in vf_scale; the others attempt to follow the same pattern.
2019-06-02 17:30:41 +01:00
Mark Thompson
5051b7f898 lavfi/vaapi: Improve support for colour properties
Attempts to pick the set of supported colour properties best matching the
input.  Output is then set with the same values, except for the colour
matrix which may change when converting between RGB and YUV.
2019-06-02 17:30:41 +01:00
Mark Thompson
5fb9eb9ed2 vf_misc_vaapi: Add missing return value checks
Parameter buffer creation can fail.
2019-06-02 17:30:41 +01:00
Mark Thompson
6ed34a4379 lavfi/vaapi: Factorise out common code for parameter buffer setup
Also enables cropping on all VAAPI filters, inherited from the existing
support in scale_vaapi.
2019-06-02 17:30:41 +01:00
Mark Thompson
f1b359aaf5 vf_crop: Add support for cropping hardware frames
Set the cropping fields in the AVFrame.
2019-06-02 17:30:41 +01:00
Paul B Mahol
fcfe85220d avfilter/af_anlmdn: avoid creating frames with zero samples 2019-06-02 13:00:35 +02:00
Paul B Mahol
1a266a1ef9 avfilter/af_anlmdn: try to recover when cache becomes negative 2019-06-02 12:59:57 +02:00
Paul B Mahol
b535577465 avfilter/af_anlmdn: add smooth factor option 2019-06-02 11:58:49 +02:00
Paul B Mahol
cbaa60329a avfilter/vf_xmedian: remove limitation of only odd number of inputs 2019-06-02 11:07:46 +02:00
Carl Eugen Hoyos
d1b4ea2f61 lavfi/showspatial: Remove two write-only variables.
Fixes two warnings:
libavfilter/avf_showspatial.c:157:26: warning: variable ‘w’ set but not used
libavfilter/avf_showspatial.c:157:23: warning: variable ‘h’ set but not used
2019-05-31 22:30:46 +02:00
Paul B Mahol
b19550367f avfilter: add showspatial multimedia filter 2019-05-31 12:53:20 +02:00
Linjie Fu
6895b350c3 lavf/qsvvpp: avoid the double-free when working in sys memory mode
Currently, picref will be freed by calling av_frame_free(&picref) in
submit_frame() in qsvvpp.c when working in system memory mode,and normally it
is freed in filter_frame() in vf_vpp_qsv.c when working in other modes.

Double free happens when working in system memory mode, remove to
fix the memory issue.

Reproduce:
ffmpeg -init_hw_device qsv=foo -filter_hw_device foo -f rawvideo -pix_fmt nv12 -s:v 852x480 \
-i 852x480.nv12 -vf 'vpp_qsv=w=500:h=400' -f rawvideo -pix_fmt nv12 qsv.nv12

Signed-off-by: Linjie Fu <linjie.fu@intel.com>
Signed-off-by: Zhong Li <zhong.li@intel.com>
2019-05-30 09:33:21 +08:00
Paul B Mahol
819ed1df94 avfilter/avf_aphasemeter: make use of av_rescale 2019-05-28 11:10:51 +02:00
Paul B Mahol
62c8707e2a avfilter/avf_showwaves: make use of av_rescale_q 2019-05-28 11:07:47 +02:00
Paul B Mahol
cd75396897 avfilter/avf_ahistogram: make use of av_rescale 2019-05-28 10:51:57 +02:00
Paul B Mahol
d2e76a1f96 avfilter/avf_showvolume: make use of av_rescale 2019-05-28 10:49:09 +02:00
Paul B Mahol
618f8bb0a4 avfilter/avf_abitscope: make use of av_rescale 2019-05-28 10:46:38 +02:00
Paul B Mahol
b16806b09d avfilter/avf_avectorscope: make use of av_rescale 2019-05-28 10:43:25 +02:00
Jun Zhao
9b069eb14e lavfi/sr: Change the backend type from flags to int
native and tensorflow is exclusive, so change the type from
flags to int.

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-05-28 09:13:36 +08:00
Jun Zhao
b983e71bde lavfi/sr: Remove slice thread flag
sr didn't enable the slice threading, so remove the flag

Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-05-28 09:13:36 +08:00
Paul B Mahol
5ddaedcaec avfilter/avf_showvolume: switch to activate 2019-05-27 18:44:19 +02:00
Paul B Mahol
293489d835 avfilter/af_superequalizer: switch to activate 2019-05-27 17:36:30 +02:00
Paul B Mahol
fc0de1c04b avfilter/avf_showfreqs: switch to activate 2019-05-26 17:11:17 +02:00
Marton Balint
1eb54323c8 avfilter/f_loop: warn about unset loop sizes
Signed-off-by: Marton Balint <cus@passwd.hu>
2019-05-24 21:39:07 +02:00
Marton Balint
87e0247a6c avfilter/f_loop: fix video loop issues with 0 size or when size is bigger than input
Fixes infinte loop with -vf loop=loop=1 and also fixes looping when the input
is less frames than the specified loop size.

Possible regressions since ef1aadffc7.

Signed-off-by: Marton Balint <cus@passwd.hu>
2019-05-24 21:39:07 +02:00
Josh Allmann
4cfc34d9a8 avfilter/vf_scale_cuda: reset frame size after acquiring from hwframe
Signed-off-by: Timo Rothenpieler <timo@rothenpieler.org>
2019-05-24 19:03:34 +02:00
Paul B Mahol
5073adcc42 avfilter/vf_xmedian: set sar to outlink 2019-05-24 18:37:41 +02:00
Paul B Mahol
a1992637ac avfilter/vf_mix: set sar to outlink 2019-05-24 18:37:41 +02:00
Paul B Mahol
a9b5af0d19 avfilter/vf_mix: use time_base from framesync 2019-05-24 18:33:16 +02:00
Ruiling Song
1d74150a7d lavfi/opencl: add nlmeans_opencl filter
Reviewed-by: Mark Thompson <sw@jkqxz.net>
Signed-off-by: Ruiling Song <ruiling.song@intel.com>
2019-05-24 15:09:22 +08:00
Xuewei Meng
023ea5e360 libavfilter/dnn_native: Add support of dilated convolution in dnn_native.
Add dilation parameter in dnn native to support dilated convolution.

Signed-off-by: Xuewei Meng <xwmeng96@gmail.com>
Signed-off-by: Steven Liu <lq@onvideo.cn>
2019-05-24 11:49:50 +08:00
Jun Zhao
98663a0bb8 lavfi/lensfun: Use ff_filter_get_nb_threads() get number of threads
ff_filter_get_nb_threads() respect AVFilterContext.nb_threads and
graph->nb_threads both, in most case, we perfer this API than using
ctx->graph->nb_threads directly.

Reviewed-by: Steven Liu <lq@onvideo.cn>
Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-05-22 17:34:18 +08:00
Xuewei Meng
ecc096513c libavfilter/dnn_native: Add multiple activation functions in dnn native
Add "Leaky_relu" and "None" option in activation function.

Reviewed-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Xuewei Meng <xwmeng96@gmail.com>
Signed-off-by: Steven Liu <lq@onvideo.cn>
2019-05-22 10:28:07 +08:00
Xuewei Meng
e45e6005ce libavfilter/dnn_native: Add multiple padding methods in dnn native
Add another two padding methods "VALID" and "SAME" as tensorflow,
and keep the existing "SAME_CLAMP_TO_EDGE" method suggested by sr filter.
As "SAME_CLAMP_TO_EDGE"can keep the output with the same size as original input,
and gives a slight better result as mentioned by sr filter.

Reviewed-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Xuewei Meng <xwmeng96@gmail.com>
Signed-off-by: Steven Liu <lq@onvideo.cn>
2019-05-21 12:53:58 +08:00
Paul B Mahol
96c79d4e1f avfilter/vf_ocr: also export confidence of result 2019-05-17 19:10:12 +02:00
Paul B Mahol
6276b4db97 avfilter/vf_separatefields: switch to activate
Fixes timestamp of last output frame.
2019-05-16 12:31:57 +02:00
Gyan Doshi
8cf947ca4c avfilter/drawtext: stop resource leak
Fixes Coverity CID 1445099
2019-05-16 10:30:28 +05:30
Philip Langdale
13168ecbdc avfilter/vf_scale_cuda: Simplify output plane addressing
I'm not sure why this was written the way it was originally. We
initialise the plane addresses correctly in hwcontext_cuda so
why try and play games to calculate the plane offsets directly
in this code?
2019-05-15 11:40:50 -07:00
Philip Langdale
bfbde996cf avfilter/vf_scale_cuda: Add support for YUV444P16
This format is interesting because it's what you get for decoded
10/12bit HEVC 4:4:4.
2019-05-15 11:40:50 -07:00
Philip Langdale
89bd7554b2 avfilter/vf_scale_cuda: Fix incorrect scaling of > 8bit content
When i converted the filter to use texture objects instead of
texture references, I incorrect dropped the `pixel_size` scaling
factor when setting `pitchInBytes`. `src_pitch` is in pixels and
so must be scaled up.
2019-05-15 11:40:50 -07:00
Paul B Mahol
f49cec2ba8 avfilter: add asr filter 2019-05-14 15:17:14 +02:00
Andreas Rheinhardt
670251de56 avfilter/vf_stack: Don't modify const strings
b3b7ba62 introduced undefined behaviour: A (non-modifiable) string
literal has been assigned to a modifiable string; said string was indeed
modified later via av_strtok.
This of course caused compiler warnings because of the discarded
qualifier; these are in particular fixed by this commit.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
2019-05-14 14:47:51 +02:00
Paul B Mahol
8ef163ba9e avfilter: add xmedian filter 2019-05-14 14:43:56 +02:00
Paul B Mahol
f1c9d6fe70 avfilter/vf_chromahold: implement blend option 2019-05-13 12:15:32 +02:00
Gyan Doshi
87db1ca632 avfilter/drawtext: make command processing error-resilient
Prevents crash or interruption in text rendering
if new option string contains invalid values.
2019-05-13 10:34:32 +05:30
Paul B Mahol
bd6942b4c0 avfilter/af_astats: fix calculations involving last sample 2019-05-12 17:12:06 +02:00
Paul B Mahol
2a4a364f6a avfilter/af_astats: fix initial values of variables 2019-05-12 17:12:06 +02:00
Paul B Mahol
59fb8cae5e avfilter: add colorhold filter
Fixes #7671.
2019-05-11 20:16:45 +02:00
Jun Zhao
c967128952 lavf/cover_rect: Fix logic check issue
Fix logic check issue #6741

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Jun Zhao <barryjzhao@tencent.com>
2019-05-10 09:07:29 +08:00
Pavel Koshevoy
181031906e avfilter/af_atempo: Make ffplay display correct timestamps when seeking
NOTE: this is a refinement of the patch from Paul B Mahol
offset all output timestamps by same amount of first input timestamp
2019-05-08 23:38:10 -06:00
Guo, Yejun
c636dc9819 libavfilter/dnn: add more data type support for dnn model input
currently, only float is supported as model input, actually, there
are other data types, this patch adds uint8.

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>
2019-05-08 12:33:00 -03:00
Guo, Yejun
25c1cd909f libavfilter/dnn: support multiple outputs for tensorflow model
some models such as ssd, yolo have more than one output.

the clean up code in this patch is a little complex, it is because
that set_input_output_tf could be called for many times together
with ff_dnn_execute_model_tf, we have to clean resources for the
case that the two interfaces are called interleaved.

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>
2019-05-08 12:33:00 -03:00
Guo, Yejun
7adfb6132e libavfilter/dnn: avoid memcpy for tensorflow dnn output
use TF_Tensor's cpu address to avoid extra memcpy.

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>
2019-05-08 12:33:00 -03:00
Guo, Yejun
e2b92896c4 libavfilter/dnn: determine dnn output during execute_model instead of set_input_output
Currently, within interface set_input_output, the dims/memory of the tensorflow
dnn model output is determined by executing the model with zero input,
actually, the output dims might vary with different input data for networks
such as object detection models faster-rcnn, ssd and yolo.

This patch moves the logic from set_input_output to execute_model which
is suitable for all the cases. Since interface changed, and so dnn_backend_native
also changes.

In vf_sr.c, it knows it's srcnn or espcn by executing the model with zero input,
so execute_model has to be called in function config_props

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>
2019-05-08 12:33:00 -03:00
Guo, Yejun
05f86f05bb libavfilter/dnn: remove limit for the name of DNN model input/output
remove the requirment that the name of DNN model input/output
should be "x"/"y",

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>
2019-05-08 12:33:00 -03:00
Guo, Yejun
05aec8bb13 libavfilter/vf_sr: refine code to remove keyword 'else'
remove 'else' since there is always 'return' in 'if' scope,
so the code will be clean for later maintenance

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>
2019-05-08 12:33:00 -03:00
Guo, Yejun
014b6a56f8 libavfilter/dnn_backend_tf.c: set layer_add_res for input layer
otherwise, the following check will return error if layer_add_res
is randomly initialized.

Signed-off-by: Guo, Yejun <yejun.guo@intel.com>
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>
2019-05-08 12:33:00 -03:00
Paul B Mahol
c539dd992c avfilter/af_afftfilt: switch to activate 2019-05-08 15:03:22 +02:00
Ruiling Song
06ba4783a0 lavfi/gblur: doing several columns at the same time
Instead of doing each column one by one, doing several columns
together gives about 30% better performance.

Reviewed-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Ruiling Song <ruiling.song@intel.com>
2019-05-08 10:39:43 +08:00
Gyan Doshi
b3b7ba623f avfilter/vf_xstack: set better error msg for
missing layout
2019-05-07 16:45:11 +05:30
Paul B Mahol
e668260ba9 avfilter/af_rubberband: also do not ignore failures 2019-05-06 21:09:43 +02:00
Paul B Mahol
066864aca8 avfilter/af_rubberband: handle case when no frame is given 2019-05-06 21:04:52 +02:00
Marton Balint
328a96839d avfilter/vf_freezedetect: fix missing freeze_start when the freeze length is around the detection duration
Fixes ticket #7875.

Signed-off-by: Marton Balint <cus@passwd.hu>
2019-05-05 20:07:35 +02:00
Paul B Mahol
73f234fdf8 avfilter/f_interleave: switch to activate 2019-05-05 19:51:44 +02:00
Paul B Mahol
38baaa1617 avfilter/vf_vibrance: add alternate option 2019-05-05 13:29:34 +02:00
Paul B Mahol
526bc2205b avfilter/vf_vibrance: factor some calculations out of loop 2019-05-05 13:09:37 +02:00
Moritz Barsnick
98541f7032 avfilter/f_realtime: add option to scale speed 2019-05-04 19:39:38 +02:00
Paul B Mahol
e94447cd49 avfilter/af_amultiply: use pts from frame from first input 2019-05-04 11:28:14 +02:00
Paul B Mahol
3d8e99eba3 avfilter/vf_vmafmotion: remove unused header 2019-05-04 11:09:47 +02:00
Paul B Mahol
16154c019f avfilter/vf_blend: remove unused header 2019-05-03 20:48:18 +02:00
Paul B Mahol
d3d1d5936f avfilter/vf_tonemap: add slice threading 2019-05-01 21:19:58 +02:00
Paul B Mahol
a7fa89fa0b avfilter/avf_abitscope: switch to activate 2019-05-01 21:17:55 +02:00
Paul B Mahol
255feeccc5 avfilter/af_biquads: add timeline support 2019-05-01 12:59:54 +02:00
Paul B Mahol
01204c04b0 avfilter/af_stereowiden: add timeline support 2019-05-01 12:40:25 +02:00
Paul B Mahol
0ee65ea95c avfilter/af_extrastereo: add timeline support 2019-05-01 12:33:56 +02:00
Paul B Mahol
d010bfad31 avfilter/af_crossfeed: add timeline support 2019-05-01 12:31:52 +02:00
Paul B Mahol
5a39b797ac avfilter/avf_showspectrum: convert some doubles to floats 2019-04-30 20:36:33 +02:00
Paul B Mahol
38c170d54c avfilter/af_asoftclip: add timeline support 2019-04-30 11:02:36 +02:00
Paul B Mahol
9cc10418b3 avfilter/af_dynaudnorm: add timeline support 2019-04-30 10:56:39 +02:00
Paul B Mahol
c82d81ba42 avfilter/af_dynaudnorm: always update delay when flushing 2019-04-30 10:05:23 +02:00
Paul B Mahol
ac551c54b1 avfilter/vf_normalize: add timeline support 2019-04-29 23:28:40 +02:00
Paul B Mahol
e6c7d838ea avfilter/af_adeclick: switch to activate 2019-04-29 22:19:15 +02:00