[FFmpeg-cvslog] libavfi/dnn: enable LibTorch xpu device option support

ffmpeg | branch: master | Wenbin Chen | Mon Jun 3 13:09:35 2024 +0800| [7560db937d1b4e412db1299d266d068bac56a7de] | committer: Guo Yejun libavfi/dnn: enable LibTorch xpu device option support Add xpu device support to libtorch backend. To enable xpu support you need to add "-Wl,--

[FFmpeg-cvslog] libavfilter/dnn_backend_openvino: Check bbox's height

ffmpeg | branch: master | Wenbin Chen | Tue Apr 2 16:13:05 2024 +0800| [8869f5ce864f197f3192c1660124e64da3f26989] | committer: Guo Yejun libavfilter/dnn_backend_openvino: Check bbox's height Check bbox's height with frame's height rather than frame's width. Signed

[FFmpeg-cvslog] libavfilter/dnn_io_proc: Take step into consideration when crop frame

ffmpeg | branch: master | Wenbin Chen | Tue Apr 2 16:13:06 2024 +0800| [478d97f3034fd7c0e1d8a1068b649e8fcc16f82f] | committer: Guo Yejun libavfilter/dnn_io_proc: Take step into consideration when crop frame Signed-off-by: Wenbin Chen Reviewed-by: Guo Yejun > http://git.videolan.

[FFmpeg-cvslog] doc: Add libtoch backend option to dnn_processing

ffmpeg | branch: master | Wenbin Chen | Mon Mar 25 10:15:08 2024 +0800| [ea2e0e92ed008fcea6c34c91ce523479e564131d] | committer: Guo Yejun doc: Add libtoch backend option to dnn_processing Signed-off-by: Wenbin Chen Reviewed-by: Guo Yejun > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] Changelog: add dnn libtorch backend entry

ffmpeg | branch: master | Wenbin Chen | Sat Mar 23 11:38:11 2024 +0100| [f34000541a3f7e81e25fe0a00401661f2c76a5ac] | committer: Stefano Sabatini Changelog: add dnn libtorch backend entry Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavfi/dnn: add LibTorch as one of DNN backend

ffmpeg | branch: master | Wenbin Chen | Fri Mar 15 12:42:49 2024 +0800| [f4e0664fd1bc94ee1a1afbd20c5a32ac33d93a2d] | committer: Guo Yejun libavfi/dnn: add LibTorch as one of DNN backend PyTorch is an open source machine learning framework that accelerates the path from research prototyping to

[FFmpeg-cvslog] libavfilter/dnn_interface: use dims to represent shapes

ffmpeg | branch: master | Wenbin Chen | Wed Jan 17 15:21:50 2024 +0800| [3de38b9da5c2ffddcf1c532bca78f989b0474494] | committer: Guo Yejun libavfilter/dnn_interface: use dims to represent shapes For detect and classify output, width and height make no sence, so change width, height to dims to

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Use class confidence to filt boxes

ffmpeg | branch: master | Wenbin Chen | Wed Jan 17 15:21:51 2024 +0800| [d52e8ed114301b44f2a1ff526c6164c43da35749] | committer: Guo Yejun libavfilter/vf_dnn_detect: Use class confidence to filt boxes Use class confidence instead of box_score to filt boxes, which is more accurate. Class

[FFmpeg-cvslog] libavfilter/dnn_bakcend_openvino: Add automatic input/output detection

ffmpeg | branch: master | Wenbin Chen | Wed Jan 17 15:21:49 2024 +0800| [c695de56b5ba8b2436e455c2284159759ca444d3] | committer: Guo Yejun libavfilter/dnn_bakcend_openvino: Add automatic input/output detection Now when using openvino backend, user doesn't need to set input/output nam

[FFmpeg-cvslog] libavfilter/dnn_backend_openvino: Add dynamic output support

ffmpeg | branch: master | Wenbin Chen | Wed Dec 27 12:16:57 2023 +0800| [86435582a62bd80b1af0e14ef703da18d6996994] | committer: Guo Yejun libavfilter/dnn_backend_openvino: Add dynamic output support Add dynamic outputs support. Some models don't have fixed output size. Its size ch

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Add two outputs ssd support

ffmpeg | branch: master | Wenbin Chen | Wed Dec 27 12:16:58 2023 +0800| [56c5930ec36be3bc63a27913d6f0a199912220c5] | committer: Guo Yejun libavfilter/vf_dnn_detect: Add two outputs ssd support For this kind of model, we can directly use its output as final result just like ssd model. The

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Add initialized value to function pointer

ffmpeg | branch: master | Wenbin Chen | Mon Dec 18 14:36:05 2023 +0800| [9415c2c11fc7141587c387e59d55a024fe38d0a2] | committer: Haihao Xiang libavfilter/vf_dnn_detect: Add initialized value to function pointer Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Fix a control flow issue

ffmpeg | branch: master | Wenbin Chen | Mon Dec 18 14:36:04 2023 +0800| [233c44f735faec193150f852d558aef7c4683b3b] | committer: Haihao Xiang libavfilter/vf_dnn_detect: Fix a control flow issue Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Add yolov4 support

ffmpeg | branch: master | Wenbin Chen | Tue Dec 12 10:33:34 2023 +0800| [1fa3346c70774d54b78c83d8ba69904215afbf78] | committer: Guo Yejun libavfilter/vf_dnn_detect: Add yolov4 support The difference of yolov4 is that sigmoid function needed to be applied on x, y coordinates. Also make it

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Add input pad

ffmpeg | branch: master | Wenbin Chen | Tue Dec 12 10:33:32 2023 +0800| [da02836b9d204ca002d973ef7b1e6f60a2316cb1] | committer: Guo Yejun libavfilter/vf_dnn_detect: Add input pad Add input pad to get model input resolution. Detection models always have fixed input size. And the output

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Add yolov3 support

ffmpeg | branch: master | Wenbin Chen | Tue Dec 12 10:33:33 2023 +0800| [a882fc029493ef2691360dcba360b0e998c628b4] | committer: Guo Yejun libavfilter/vf_dnn_detect: Add yolov3 support Add yolov3 support. The difference of yolov3 is that it has multiple outputs in different scale to perform

[FFmpeg-cvslog] libavfiter/dnn_backend_openvino: Add multiple output support

ffmpeg | branch: master | Wenbin Chen | Tue Dec 12 10:33:31 2023 +0800| [22652b576c2a0670d341648c68ca469ebe08f1a1] | committer: Guo Yejun libavfiter/dnn_backend_openvino: Add multiple output support Add multiple output support to openvino backend. You can use '&' to split di

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Set used pointer to NULL

ffmpeg | branch: master | Wenbin Chen | Thu Dec 14 10:49:28 2023 +0800| [22bebfa5e6cfcbd332a387b6f91ae38a4cbaf8b9] | committer: Guo Yejun libavfilter/vf_dnn_detect: Set used pointer to NULL Set used pointer to NULL in case it leaks the storage. Signed-off-by: Wenbin Chen > h

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Fix uninitialized variables error

ffmpeg | branch: master | Wenbin Chen | Thu Dec 14 10:49:27 2023 +0800| [a1e8892a8a2e5d32e31cc5b30a3727700de51017] | committer: Guo Yejun libavfilter/vf_dnn_detect: Fix uninitialized variables error Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Fix an incorrect expression

ffmpeg | branch: master | Wenbin Chen | Thu Dec 14 10:49:25 2023 +0800| [e01afa1c6d53eeae93210830968c1f0c111c1b36] | committer: Guo Yejun libavfilter/vf_dnn_detect: Fix an incorrect expression Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Add NULL pointer check

ffmpeg | branch: master | Wenbin Chen | Thu Dec 14 10:49:26 2023 +0800| [833722343aa52c77893f2330d1dc06b5423da6fe] | committer: Guo Yejun libavfilter/vf_dnn_detect: Add NULL pointer check Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Add model_type option.

ffmpeg | branch: master | Wenbin Chen | Tue Nov 21 10:20:17 2023 +0800| [caa5d123a710e11686458ff563a99a0fabc4f55c] | committer: Guo Yejun libavfilter/vf_dnn_detect: Add model_type option. There are many kinds of detection DNN model and they have different preprocess and postprocess methods

[FFmpeg-cvslog] libavfilter/vf_dnn_detect: Add yolo support

ffmpeg | branch: master | Wenbin Chen | Tue Nov 21 10:20:18 2023 +0800| [47b2328076f865a12dd0d50aed02fbef32f69490] | committer: Guo Yejun libavfilter/vf_dnn_detect: Add yolo support Add yolo support. Yolo model doesn't output final result. It outputs candidate boxes, so we need post-pr

[FFmpeg-cvslog] libavfilter/dnn/openvino: Reduce redundant memory allocation

ffmpeg | branch: master | Wenbin Chen | Thu Nov 9 16:13:10 2023 +0800| [fa81de4af045317aefeae941cf54f349c39e81e0] | committer: Guo Yejun libavfilter/dnn/openvino: Reduce redundant memory allocation We can directly get data ptr from tensor, so that extral memory allocation can be removed

[FFmpeg-cvslog] libavfilter/dnn: Add scale and mean preprocess to openvino backend

ffmpeg | branch: master | Wenbin Chen | Thu Sep 21 09:26:32 2023 +0800| [c8c925dc2920efccfb90deaf27a56e7b673cc54c] | committer: Guo Yejun libavfilter/dnn: Add scale and mean preprocess to openvino backend Dnn models has different data preprocess requirements. Scale and mean parameters are

[FFmpeg-cvslog] libavfilter/dnn: Initialze DNNData variables

ffmpeg | branch: master | Wenbin Chen | Thu Sep 21 09:26:33 2023 +0800| [58b6c0c327ab9b49e177694ac02fd8d782e20a4a] | committer: Guo Yejun libavfilter/dnn: Initialze DNNData variables Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavfilter/dnn: add layout option to openvino backend

ffmpeg | branch: master | Wenbin Chen | Thu Sep 21 09:26:31 2023 +0800| [74ce1d2d11a56b522a343d944bf5b184bb2d9212] | committer: Guo Yejun libavfilter/dnn: add layout option to openvino backend Dnn models have different input layout (NCHW or NHWC), so a "layout" option is added Use

[FFmpeg-cvslog] lavfi/dnn: Add OpenVINO API 2.0 support

ffmpeg | branch: master | Wenbin Chen | Tue Aug 15 16:26:31 2023 +0800| [e79bd1f1b158b84c4aa5083b5e2af2de8ede3b0e] | committer: Guo Yejun lavfi/dnn: Add OpenVINO API 2.0 support OpenVINO API 2.0 was released in March 2022, which introduced new features. This commit implements current OpenVINO

[FFmpeg-cvslog] libavfilter/vf_vpp_qsv: Fix a timestamp bug when framerate is not set

ffmpeg | branch: master | Wenbin Chen | Tue Aug 15 16:34:25 2023 +0800| [9ae4863cc57c38fa2f41f3b8c732fede4a2b78c6] | committer: Haihao Xiang libavfilter/vf_vpp_qsv: Fix a timestamp bug when framerate is not set If user doesn't set framerate when he creates a filter, the filter uses de

[FFmpeg-cvslog] libavcodec/qsvenc: Enable 444 encoding for RGB input

ffmpeg | branch: master | Wenbin Chen | Fri Dec 9 14:36:23 2022 +0800| [0825b461e64b5924fdee7055643db65870d2eaae] | committer: Haihao Xiang libavcodec/qsvenc: Enable 444 encoding for RGB input MSDK/VPL uses 420 chroma format as default to encode RGB, and this is not a proper usage. Now

[FFmpeg-cvslog] libavcodec/qsvenc: Flush cached frames before reset encoder

ffmpeg | branch: master | Wenbin Chen | Mon Feb 13 17:03:17 2023 +0800| [aeceefa6220ccb8eac625f78c6fa90d048ccd2de] | committer: Haihao Xiang libavcodec/qsvenc: Flush cached frames before reset encoder According to https://github.com/Intel-Media-SDK/MediaSDK/blob/master/doc/mediasdk-man.md

[FFmpeg-cvslog] libavcodec/qsvenc: Do not pass RGB solorspace to VPL/MSDK

ffmpeg | branch: master | Wenbin Chen | Mon Feb 13 16:49:26 2023 +0800| [e530d38bbf5f58e3e22f87dff635a50e267abb77] | committer: Haihao Xiang libavcodec/qsvenc: Do not pass RGB solorspace to VPL/MSDK When encode RGB frame, Intel driver convert RGB to YUV, so we cannot set RGB colorspace to VPL

[FFmpeg-cvslog] libavfilter/qsvvpp: check the return value

ffmpeg | branch: master | Wenbin Chen | Wed Feb 8 16:04:50 2023 +0800| [b7a335c5e9d2cd47dc8adeca2edc6714fa2d2996] | committer: Haihao Xiang libavfilter/qsvvpp: check the return value Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc_av1: Add max_frame_size support to av1_qsv encoder

ffmpeg | branch: master | Wenbin Chen | Mon Dec 26 14:02:01 2022 +0800| [13d04e30d7753811176e154a2828bf054a9a846a] | committer: Haihao Xiang libavcodec/qsvenc_av1: Add max_frame_size support to av1_qsv encoder Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] libavcodec/qsvenc_av1: Add low_delay_brc support to av1_qsv encoder

ffmpeg | branch: master | Wenbin Chen | Mon Dec 26 14:02:00 2022 +0800| [c8e73558fe0181a0b3e611f486bb8bc308af24b4] | committer: Haihao Xiang libavcodec/qsvenc_av1: Add low_delay_brc support to av1_qsv encoder Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git

[FFmpeg-cvslog] libavfilter/qsvvpp: Change the alignment to meet the requirement of YUV420P format

ffmpeg | branch: master | Wenbin Chen | Mon Dec 5 14:32:13 2022 +0800| [20f29411912efcfaa92f2eb02a9fddfcd5073271] | committer: Haihao Xiang libavfilter/qsvvpp: Change the alignment to meet the requirement of YUV420P format When process yuv420 frames, FFmpeg uses same alignment on Y/U/V

[FFmpeg-cvslog] libavcodec/qsvenc.c: Enable MFX_GOP_STRICT when adpative gop is disabled

ffmpeg | branch: master | Wenbin Chen | Thu Nov 24 10:47:29 2022 +0800| [47fff1eda155c30fc77cae74f40e71d4203c5043] | committer: Haihao Xiang libavcodec/qsvenc.c: Enable MFX_GOP_STRICT when adpative gop is disabled adaptive_i and adaptive_b cannot work with MFX_GOP_STRICT, so only enable

[FFmpeg-cvslog] doc/examples: Add qsv_transcode example

ffmpeg | branch: master | Wenbin Chen | Mon Oct 31 15:22:25 2022 +0800| [efe2b63c89462a46c0450da49832b3c9075b1222] | committer: Haihao Xiang doc/examples: Add qsv_transcode example Add qsv_transcode example which shows how to use qsv to do hardware accelerated transcoding, also show how to

[FFmpeg-cvslog] libavcodec/qsvenc: Add skip_frame support to qsvenc

ffmpeg | branch: master | Wenbin Chen | Wed Nov 2 16:11:51 2022 +0800| [dc4d3429e1b3d162b04372eace67156b52305164] | committer: Haihao Xiang libavcodec/qsvenc: Add skip_frame support to qsvenc Add skip_frame support to qsvenc. Use per-frame metadata "qsv_skip_frame" to control it.

[FFmpeg-cvslog] doc/encoders: Add doc for av1_qsv

ffmpeg | branch: master | Wenbin Chen | Thu Oct 13 16:25:24 2022 +0800| [d79c240196f43b93bd204363f1facc270029f113] | committer: Haihao Xiang doc/encoders: Add doc for av1_qsv Add doc for av1_qsv. Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc_av1: add av1_qsv encoder

ffmpeg | branch: master | Wenbin Chen | Thu Oct 13 16:25:23 2022 +0800| [dc9e4789a3b504c08c8cd24e990aa692dde50bc6] | committer: Haihao Xiang libavcodec/qsvenc_av1: add av1_qsv encoder It is available only when libvpl is enabled. MSDK doesn't support av1 encoding. sample command: ffmp

[FFmpeg-cvslog] libavcodec/qsvenc: Add pic_timing_sei reset support to qsv

ffmpeg | branch: master | Wenbin Chen | Fri Sep 23 10:44:12 2022 +0800| [3771d54989d761cab02bbd8a7c03dbc2add05a81] | committer: Haihao Xiang libavcodec/qsvenc: Add pic_timing_sei reset support to qsv Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc: Let runtime to decide targetUsage

ffmpeg | branch: master | Wenbin Chen | Mon Sep 26 17:36:30 2022 +0800| [5682046ce5ee16bca3159a766fb3f9caf537fe0c] | committer: Haihao Xiang libavcodec/qsvenc: Let runtime to decide targetUsage Set preset default value to MFX_TARGETUSAGE_UNKNOWN. Let runtime to decide the targetUsage, so that

[FFmpeg-cvslog] libavcodec/qsvenc: Let runtime to set default parameter.

ffmpeg | branch: master | Wenbin Chen | Mon Sep 26 17:36:29 2022 +0800| [a3c0a3ec604da635fba5d356d7ec53bbd7c38d78] | committer: Haihao Xiang libavcodec/qsvenc: Let runtime to set default parameter. Unset qsv_h264 and qsv_hevc's default settings. Let runtime to decide these parameters, so

[FFmpeg-cvslog] libavcodec/qsvenc: Add bitrate reset support to qsvenc

ffmpeg | branch: master | Wenbin Chen | Fri Sep 23 10:44:11 2022 +0800| [29a3ba869317054b3a795eddf632857a77485755] | committer: Haihao Xiang libavcodec/qsvenc: Add bitrate reset support to qsvenc Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc: Add framerate reset support to qsv

ffmpeg | branch: master | Wenbin Chen | Fri Sep 23 10:44:10 2022 +0800| [17df61083dc79e90fa0180811dc69fab077096aa] | committer: Haihao Xiang libavcodec/qsvenc: Add framerate reset support to qsv Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc: Add low_delay_brc reset support to qsvenc

ffmpeg | branch: master | Wenbin Chen | Tue Sep 6 17:22:58 2022 +0800| [a2fd553dd3079601249e56f6aa1a6fe711d0663c] | committer: Haihao Xiang libavcodec/qsvenc: Add low_delay_brc reset support to qsvenc Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc: Add max/min qp reset support in qsvenc

ffmpeg | branch: master | Wenbin Chen | Tue Sep 6 17:22:57 2022 +0800| [005c7a4f61ef047726270b59c08742f9ce6f7182] | committer: Haihao Xiang libavcodec/qsvenc: Add max/min qp reset support in qsvenc Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc: Add intra refresh reset support to qsvenc

ffmpeg | branch: master | Wenbin Chen | Tue Sep 6 17:22:56 2022 +0800| [9155ec096b0076bf5090c22c8c2f0f67bd8ee086] | committer: Haihao Xiang libavcodec/qsvenc: Add intra refresh reset support to qsvenc Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc: Add "slice" intra refresh type to qsvenc

ffmpeg | branch: master | Wenbin Chen | Tue Sep 6 17:22:55 2022 +0800| [f3ba1458b6787484978e9a8f1560dcabfefba92d] | committer: Haihao Xiang libavcodec/qsvenc: Add "slice" intra refresh type to qsvenc Add "slice" intra refresh type to h264_qsv and hevc_qsv. This type mean

[FFmpeg-cvslog] libavcodec/qsvenc: Add gop_size reset support to qsvenc

ffmpeg | branch: master | Wenbin Chen | Tue Sep 6 17:22:54 2022 +0800| [d60f657a50773ae8371585539bf484aa072bfc71] | committer: Haihao Xiang libavcodec/qsvenc: Add gop_size reset support to qsvenc Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc: Add max_frame_size reset support to qsv

ffmpeg | branch: master | Wenbin Chen | Tue Sep 6 17:22:53 2022 +0800| [c679de9be0d76e4eadc42f68421a569ff304c3e6] | committer: Haihao Xiang libavcodec/qsvenc: Add max_frame_size reset support to qsv Signed-off-by: Wenbin Chen > http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi

[FFmpeg-cvslog] libavcodec/qsvenc: skip parameter resetting on mjpeg_qsv

ffmpeg | branch: master | Wenbin Chen | Tue Sep 6 17:22:52 2022 +0800| [04e491cdf4a67e6533dc8ae29a8de1ba6837] | committer: Haihao Xiang libavcodec/qsvenc: skip parameter resetting on mjpeg_qsv mjpeg_qsv don't support dynamic resetting, so skip it. Signed-off-by: Wenbin Chen &

[FFmpeg-cvslog] libavcodec/qsvenc: Add adaptive_i/b to hevc_qsv

ffmpeg | branch: master | Wenbin Chen | Wed Aug 17 15:07:19 2022 +0800| [ae7d19680b7e7a401773c619e7c36f3faaf94d0d] | committer: Haihao Xiang libavcodec/qsvenc: Add adaptive_i/b to hevc_qsv Add adaptive_i/b feature to hevc_qsv. Adaptive_i allows changing of frame type from P and B to I

[FFmpeg-cvslog] libavfilter/vf_overlay_qsv: Use format of first input to set output format for overlay_qsv

ffmpeg | branch: master | Wenbin Chen | Thu Jul 7 11:12:52 2022 +0800| [636a629ff63ef60c9b31ba87e8e0a90e1f277c97] | committer: Haihao Xiang libavfilter/vf_overlay_qsv: Use format of first input to set output format for overlay_qsv overlay_qsv hard coded to use nv12 as output format. Now use

[FFmpeg-cvslog] libavcodec/qsvenc: Use parameter from AVCodecContext to reset qsv codec

ffmpeg | branch: master | Wenbin Chen | Thu Jul 14 16:43:25 2022 +0800| [52ad617af56e5bd009f488a26ceb4f73595b2477] | committer: Haihao Xiang libavcodec/qsvenc: Use parameter from AVCodecContext to reset qsv codec Using parameter from AVCodecContext to reset qsv codec is more suitable for

[FFmpeg-cvslog] libavcodec/qsvenc: Change the parameter log to be thread safe

ffmpeg | branch: master | Wenbin Chen | Fri Jul 8 15:14:51 2022 +0800| [8b82c9cdc74fd227726b81a7c1768b11536c4471] | committer: Haihao Xiang libavcodec/qsvenc: Change the parameter log to be thread safe Dividing one line log into several av_log() call is not thread safe. Now merge these

[FFmpeg-cvslog] avcodec/mpegvideo_enc: Unify the code style

ffmpeg | branch: master | Wenbin Chen | Mon Jul 4 11:17:35 2022 +0800| [bd7dae7daffc233bdb98b361f30170e1320ee0c6] | committer: Marton Balint avcodec/mpegvideo_enc: Unify the code style Change whitespace and add newline to unify the code style. Signed-off-by: Wenbin Chen Signed-off-by

[FFmpeg-cvslog] avcodec/mpegvideo_enc: Fix a chroma mb size error in sse_mb()

ffmpeg | branch: master | Wenbin Chen | Mon Jul 4 11:17:34 2022 +0800| [76b33704b628248d77ed61cbef9ecd9430a4b6c5] | committer: Marton Balint avcodec/mpegvideo_enc: Fix a chroma mb size error in sse_mb() For 422 frames we should not use hard coded 8 to calculate mb size for uv plane. Chroma

[FFmpeg-cvslog] libavcodec/qsvenc_hevc: Use default value from MSDK to set bf.

ffmpeg | branch: master | Wenbin Chen | Mon Jun 20 17:06:31 2022 +0800| [afa20e6db82245807bd0bb181dd22e1b714972cc] | committer: Haihao Xiang libavcodec/qsvenc_hevc: Use default value from MSDK to set bf. Change the default value of "bf" for hevc_qsv to -1. 8 isn't the best cho

[FFmpeg-cvslog] libavcodec/qsvenc: add ROI support to qsv encoder

ffmpeg | branch: master | Wenbin Chen | Wed Jun 8 13:02:48 2022 +0800| [97141ffeec803c448d81ee4a53cfa2355f79f7ec] | committer: Haihao Xiang libavcodec/qsvenc: add ROI support to qsv encoder Use The mfxEncoderCtrl parameter to enable ROI. Get side data "AVRegionOfInterest" and

[FFmpeg-cvslog] libavutil/hwcontext_qsv: Align width and heigh when download qsv frame

ffmpeg | branch: master | Wenbin Chen | Wed Apr 6 18:10:08 2022 +0800| [7e7b3a4c283a4d70357cb0d2cb3fad8ae8377e2a] | committer: Haihao Xiang libavutil/hwcontext_qsv: Align width and heigh when download qsv frame The width and height for qsv frame to download need to be aligned with 16. Add

[FFmpeg-cvslog] libavcodec/qsvenc: Add more pixel format support to qsvenc

ffmpeg | branch: master | Wenbin Chen | Wed Apr 6 16:48:04 2022 +0800| [fb8880906375a2a17ed94c2ef7b9048fd25a75ec] | committer: Haihao Xiang libavcodec/qsvenc: Add more pixel format support to qsvenc Qsv encoder only support input P010 and nv12 format directly from system memory. For other

[FFmpeg-cvslog] libavcodec/qsvdec: Add more pixel format support to qsvdec

ffmpeg | branch: master | Wenbin Chen | Wed Apr 6 16:48:03 2022 +0800| [e0ae810da3ed0b4489d0dc35c40aa239ba213ec6] | committer: Haihao Xiang libavcodec/qsvdec: Add more pixel format support to qsvdec Qsv decoder only supports directly output nv12 and p010 to system memory. For other format

[FFmpeg-cvslog] libavcodec/qsvdec: remove redundant decodeHeader()

ffmpeg | branch: master | Wenbin Chen | Fri Mar 18 14:25:10 2022 +0800| [4a633eb36fb078f600d6cbbd8f184204c61a96f9] | committer: Haihao Xiang libavcodec/qsvdec: remove redundant decodeHeader() Since ffmpeg-qsv uses return value to reinit decoder, it doesn't need to decode header each

[FFmpeg-cvslog] libavcodec/qsvdec: using suggested num to set init_pool_size

ffmpeg | branch: master | Wenbin Chen | Fri Mar 18 14:25:11 2022 +0800| [0a0847dbac9bdfd24607572886a0d42cdea1691a] | committer: Haihao Xiang libavcodec/qsvdec: using suggested num to set init_pool_size The init_pool_size is set to be 64 and it is too many. Use IOSurfQuery to get

[FFmpeg-cvslog] libavcodec/qsvdec: reinit decoder according to decode() return value

ffmpeg | branch: master | Wenbin Chen | Fri Mar 18 14:25:09 2022 +0800| [40f38659d777af3e0cf2487df81f4a22884e2806] | committer: Haihao Xiang libavcodec/qsvdec: reinit decoder according to decode() return value FFmpeg-qsv decoder reinit codec when width and height change, but there are not

[FFmpeg-cvslog] libavcodec/qsvenc: add mbbrc to hevc_qsv

ffmpeg | branch: master | Wenbin Chen | Mon Feb 21 16:56:56 2022 +0800| [342d4fb0563cafa4e444c83dc4ecfa0d69a71eda] | committer: Haihao Xiang libavcodec/qsvenc: add mbbrc to hevc_qsv Add mbbrc to hevc_qsv For detailed description, please see "mbbrc" part in: https://github.com/Intel

[FFmpeg-cvslog] libavcodec/vaapi_encode: Add async_depth to vaapi_encoder to increase performance

ffmpeg | branch: master | Wenbin Chen | Fri Feb 18 11:07:47 2022 +0800| [d165ce22a4a7cc4ed60238ce8f3d5dcbbad3e266] | committer: Haihao Xiang libavcodec/vaapi_encode: Add async_depth to vaapi_encoder to increase performance Fix: #7706. After commit 5fdcf85bbffe7451c2, vaapi encoder&#

[FFmpeg-cvslog] libavcodec/vaapi_encode: Add new API adaption to vaapi_encode

ffmpeg | branch: master | Wenbin Chen | Fri Feb 18 11:07:46 2022 +0800| [e0ff86993052b49a64d434bac345e92fc149f446] | committer: Haihao Xiang libavcodec/vaapi_encode: Add new API adaption to vaapi_encode Add vaSyncBuffer to VAAPI encoder. Old version API vaSyncSurface wait surface to complete

[FFmpeg-cvslog] libavcodec/qsvenc: add more ChromaFormat support to mjpeg_qsv

ffmpeg | branch: master | Wenbin Chen | Mon Feb 21 15:28:33 2022 +0800| [4869ccb3f38a606401a1c3e3a31f18d35eb7dc39] | committer: Haihao Xiang libavcodec/qsvenc: add more ChromaFormat support to mjpeg_qsv ChromaForamt for mjpeg-qsv is always set to yuv420, and this will be wrong when encode

[FFmpeg-cvslog] libavcodec/qsvenc_hevc: encode RGB format rawvideo

ffmpeg | branch: master | Wenbin Chen | Fri Feb 18 13:50:12 2022 +0800| [9da19c290960e976d9da42b5ee92f887a74dc8a5] | committer: Haihao Xiang libavcodec/qsvenc_hevc: encode RGB format rawvideo Add support for hevc_qsv to input RGB format frame. It will transform frame to yuv inside MediaSDK

[FFmpeg-cvslog] libavcodec/qsvdec: use the parameter from decodeHeader to configure surface

ffmpeg | branch: master | Wenbin Chen | Fri Feb 11 10:37:36 2022 +0800| [b1c26ce9fe4da93e4ca4108a32f478ded7cfd428] | committer: Haihao Xiang libavcodec/qsvdec: use the parameter from decodeHeader to configure surface MSDK recognizes both yuv420p10 and yuv420p9 as MFX_FOURCC_P010, but

[FFmpeg-cvslog] libavcodec/vaapi_decode: fix the problem that init_pool_size < nb_surface

ffmpeg | branch: master | Wenbin Chen | Tue Jan 11 14:55:37 2022 +0800| [a5b1e632c96727e485419748b4bd2241b5b22fe3] | committer: Haihao Xiang libavcodec/vaapi_decode: fix the problem that init_pool_size < nb_surface For vaapi if the init_pool_size is not zero, the pool size is fixed. T

[FFmpeg-cvslog] libavcodec/qsvenc: Add b_strategy option to hevc_qsv

ffmpeg | branch: master | Wenbin Chen | Tue Jan 25 15:12:36 2022 +0800| [2ecc4fde5f01289b4c2b8d58c905a408dbb96d8c] | committer: Haihao Xiang libavcodec/qsvenc: Add b_strategy option to hevc_qsv Add b_strategy option to hevc_qsv. By enabling this option, encoder can use b frames as reference

[FFmpeg-cvslog] libavcodec/qsvenc: Add intra refresh to hevc_qsv and add new intra refresh parameter

ffmpeg | branch: master | Wenbin Chen | Tue Jan 25 15:12:37 2022 +0800| [e102fd98d6f30d3cd41debafab30d29d5f72117c] | committer: Haihao Xiang libavcodec/qsvenc: Add intra refresh to hevc_qsv and add new intra refresh parameter Add intra refresh support to hevc_qsv as well. Add an new intra

[FFmpeg-cvslog] libavcodec/qsvenc: Add max slice size support to hevc_qsv

ffmpeg | branch: master | Wenbin Chen | Tue Jan 25 15:12:35 2022 +0800| [168c6e23a54a65447110e4033b11b5965fd560c1] | committer: Haihao Xiang libavcodec/qsvenc: Add max slice size support to hevc_qsv Add max_slice_size option to hevc_qsv as well. Signed-off-by: Wenbin Chen Signed-off-by

[FFmpeg-cvslog] libavcodec/qsvenc: fix a memory leak problem

ffmpeg | branch: master | Wenbin Chen | Thu Jan 20 15:13:43 2022 +0800| [05ba0ffca9b4c28e08fdd8086218c0075537125c] | committer: Haihao Xiang libavcodec/qsvenc: fix a memory leak problem "qf->frame" ref to input frame but it isn't released. av_frame_unref() is added before

[FFmpeg-cvslog] libavcodec/qsvenc: Add transform skip to hevc_qsv

ffmpeg | branch: master | Wenbin Chen | Mon Jan 17 13:11:22 2022 +0800| [1fe78f8419356fab4c0273bd78c0b376acdae52e] | committer: Haihao Xiang libavcodec/qsvenc: Add transform skip to hevc_qsv Add transform_skip option to hevc_qsv. By enabling this option, the transform_skip_enabled_flag in PPS

[FFmpeg-cvslog] libavcodec/qsvenc: Add low latency P-pyramid support to qsv

ffmpeg | branch: master | Wenbin Chen | Mon Jan 17 13:11:21 2022 +0800| [9d6cc6a5448c4464139c87ed05fb9a70ccbc7bac] | committer: Haihao Xiang libavcodec/qsvenc: Add low latency P-pyramid support to qsv Add low latency P-pyramid support to qsv. This feature relates to command line option

[FFmpeg-cvslog] libavcodec/qsvenc: Add DisableDeblockingIdc support to qsv

ffmpeg | branch: master | Wenbin Chen | Mon Jan 17 13:11:20 2022 +0800| [812b6ee6bdd1be7172af896088bd9a22eb538823] | committer: Haihao Xiang libavcodec/qsvenc: Add DisableDeblockingIdc support to qsv Add dblk_idc option to 264_qsv and hevc_qsv. Turining on this opion can disable deblocking

[FFmpeg-cvslog] libavcodec/qsvenc: Add max_frame_size support to hevc_qsv

ffmpeg | branch: master | Wenbin Chen | Mon Jan 17 13:11:19 2022 +0800| [ec8bc07104706a02d7a36a93b34844579f7fc6ec] | committer: Haihao Xiang libavcodec/qsvenc: Add max_frame_size support to hevc_qsv Add max_frame_size support to hevc_qsv as well. Signed-off-by: Wenbin Chen Signed-off-by

[FFmpeg-cvslog] doc/encoders.texi: Add doc for qsv

ffmpeg | branch: master | Wenbin Chen | Thu Jan 6 11:00:40 2022 +0800| [035d2cc46368a7cdbf2585acdb1a83885706e43f] | committer: Haihao Xiang doc/encoders.texi: Add doc for qsv Add doc for qsv decoder. Add more option's introduction to qsv encoder. Signed-off-by: Wenbin Chen Signed-o

[FFmpeg-cvslog] libavutil/hwcontext_qsv: clean padding when upload qsv frames

ffmpeg | branch: master | Wenbin Chen | Thu Dec 23 10:26:37 2021 +0800| [ed6c5c13b10930ea95c622d6ef6e32a6e2077018] | committer: Haihao Xiang libavutil/hwcontext_qsv: clean padding when upload qsv frames Fix #7830 When we upload a frame that is not padded as MSDK requires, we create a new

[FFmpeg-cvslog] hwcontext_vulkan: support creating DRM-tiled images and autodetecting modifiers

ffmpeg | branch: master | Wenbin Chen | Tue Dec 7 17:05:53 2021 +0800| [83fe28221e92921f74d99ed15c258a3c79f4441d] | committer: Lynne hwcontext_vulkan: support creating DRM-tiled images and autodetecting modifiers When vulkan image exports to drm, the tilling need to be

[FFmpeg-cvslog] hwcontext_vulkan: add support for mapping frames with planes in a single VkDeviceMemory

ffmpeg | branch: master | Wenbin Chen | Tue Dec 7 17:05:52 2021 +0800| [0d524b170ec4bf762e11c49743d813de650d409a] | committer: Lynne hwcontext_vulkan: add support for mapping frames with planes in a single VkDeviceMemory Add support to map vulkan frames to software frames when using

[FFmpeg-cvslog] libavutil/hwcontext_vaapi: Add a new nv12 format map to support vulkan frame

ffmpeg | branch: master | Wenbin Chen | Tue Dec 7 17:05:50 2021 +0800| [f3c9847c2754b7a43eb721c95e356a53085c2491] | committer: Lynne libavutil/hwcontext_vaapi: Add a new nv12 format map to support vulkan frame Vulkan will map nv12 to R8 and GR88, so add this map to vaapi to support vulkan

[FFmpeg-cvslog] hwcontext_vulkan: add support for allocating all planes in a single allocation

ffmpeg | branch: master | Wenbin Chen | Tue Dec 7 17:05:51 2021 +0800| [bd6ef73399f3f2fcd59ae2ff358954bb536540f2] | committer: Lynne hwcontext_vulkan: add support for allocating all planes in a single allocation VAAPI on Intel can import external frame, but the planes of the external frames

[FFmpeg-cvslog] libavutil/hwcontext_qsv: fix a bug for mapping qsv frame to vaapi

ffmpeg | branch: master | Wenbin Chen | Mon Sep 13 14:22:22 2021 +0800| [f2891fbdeddd9049ba03b600d6c93a1ab732df66] | committer: James Almer libavutil/hwcontext_qsv: fix a bug for mapping qsv frame to vaapi Command below failed. ffmpeg -v verbose -init_hw_device vaapi=va:/dev/dri/renderD128

[FFmpeg-cvslog] libavfilter/vf_vpp_qsv: fix uninitialized variable problem

ffmpeg | branch: master | Wenbin Chen | Wed Sep 15 13:15:18 2021 +0800| [a822e70becdec99363d1995a4fa44d3500d22c03] | committer: James Almer libavfilter/vf_vpp_qsv: fix uninitialized variable problem This two variables may be used below with uninitialized value. Now fix them. Signed-off-by