video - 如何使用 'overlay_cuda'、ffmpeg 视频过滤器在视频上刻录基于字幕的图像

Question

我必须使用 ffmpeg(v4.3) 和硬件加速器 cuda 在视频上刻录基于字幕的图像。我想在过滤器复合体中使用“overlay_cuda”。
这是我的命令。

./ffmpeg -init_hw_device cuda=cuda:1 -hwaccel cuda -filter_hw_device cuda -hwaccel_output_format cuda -i 2_dump.ts -filter_complex "[v:0]scale_npp=1920:1080,format=yuv420p[base_video];[base_video][s:3]overlay_cuda[resurlt_video]" -map "[resurlt_video]" -map 0:a -c:v h264_nvenc -f mpegts subtitle_test.ts

但它失败了。
这些是包含错误的输出消息。

ffmpeg version 4.3 Copyright (c) 2000-2020 the FFmpeg developers
built with gcc 4.8.5 (GCC) 20150623 (Red Hat 4.8.5-39)
configuration: --prefix=/usr/local --pkg-config-flags=--static --extra-cflags=-I/usr/local/include --extra-ldflags='-L/usr/local/lib -L/usr/local/lib64' --extra-libs='-lm -lpthread -lc' --bindir=/usr/local/bin --enable-cross-compile --enable-pic --enable-gpl --enable-libfdk_aac --enable-libfreetype --enable-libmp3lame --enable-libopus --enable-libvpx --enable-libx264 --enable-libx265 --enable-nonfree --enable-cuda-nvcc --enable-cuvid --enable-nvenc --enable-libnpp --extra-cflags=-I/usr/local/cuda/include --extra-ldflags=-L/usr/local/cuda/lib64 --enable-libass --enable-static --disable-shared
libavutil      56. 51.100 / 56. 51.100
libavcodec     58. 91.100 / 58. 91.100
libavformat    58. 45.100 / 58. 45.100
libavdevice    58. 10.100 / 58. 10.100
libavfilter     7. 85.100 /  7. 85.100
libswscale      5.  7.100 /  5.  7.100
libswresample   3.  7.100 /  3.  7.100
libpostproc    55.  7.100 / 55.  7.100
[mpegts @ 0x4971500] start time for stream 5 is not set in estimate_timings_from_pts
[mpegts @ 0x4971500] start time for stream 6 is not set in estimate_timings_from_pts
[mpegts @ 0x4971500] start time for stream 7 is not set in estimate_timings_from_pts
[mpegts @ 0x4971500] start time for stream 8 is not set in estimate_timings_from_pts
[mpegts @ 0x4971500] start time for stream 9 is not set in estimate_timings_from_pts
[mpegts @ 0x4971500] start time for stream 10 is not set in estimate_timings_from_pts
[mpegts @ 0x4971500] start time for stream 11 is not set in estimate_timings_from_pts
[mpegts @ 0x4971500] start time for stream 12 is not set in estimate_timings_from_pts
Input #0, mpegts, from '2_dump.ts':
Duration: 00:05:09.80, start: 1.400000, bitrate: 12595 kb/s
Program 1 
Metadata:
  service_name    : Service01
  service_provider: AMUZLAB
Stream #0:0[0x100](eng): Audio: ac3 ([129][0][0][0] / 0x0081), 48000 Hz, 5.1(side), fltp, 384 kb/s
Stream #0:1[0x101](ind): Audio: ac3 ([129][0][0][0] / 0x0081), 48000 Hz, stereo, fltp, 192 kb/s
Stream #0:2[0x102](zho): Audio: ac3 ([129][0][0][0] / 0x0081), 48000 Hz, stereo, fltp, 192 kb/s
Stream #0:3[0x103](kho): Audio: ac3 ([129][0][0][0] / 0x0081), 48000 Hz, stereo, fltp, 192 kb/s
Stream #0:4[0x104]: Video: h264 (High) ([27][0][0][0] / 0x001B), yuv420p(top first), 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 50 tbr, 90k tbn, 50 tbc
Stream #0:5[0x105](CHI): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
Stream #0:6[0x106](CHS): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
Stream #0:7[0x107](IND): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
Stream #0:8[0x108](THA): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
Stream #0:9[0x109](MAN): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
Stream #0:10[0x10a](MON): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
Stream #0:11[0x10b](BUR): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
Stream #0:12[0x10c](ENG): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
File 'subtitle_test.ts' already exists. Overwrite? [y/N] y
Stream mapping:
Stream #0:4 (h264) -> scale_npp (graph 0)
Stream #0:8 (dvbsub) -> overlay_cuda:overlay (graph 0)
overlay_cuda (graph 0) -> Stream #0:0 (h264_nvenc)
Stream #0:0 -> #0:1 (ac3 (native) -> mp2 (native))
Stream #0:1 -> #0:2 (ac3 (native) -> mp2 (native))
Stream #0:2 -> #0:3 (ac3 (native) -> mp2 (native))
Stream #0:3 -> #0:4 (ac3 (native) -> mp2 (native))
Press [q] to stop, [?] for help
[mpegts @ 0x4971500] sub2video: using 1920x1080 canvas
Impossible to convert between the formats supported by the filter 'Parsed_scale_npp_0' and the filter 'auto_scaler_0'
Error reinitializing filters!
Failed to inject frame into filter network: Function not implemented
Error while processing the decoded data for stream #0:4
Conversion failed!

并且，命令行成功使用了“覆盖”、软件视频过滤器、非硬件加速器。

./ffmpeg -hwaccel cuda -re -i 2_dump.ts -filter_complex "[0:v:0][0:s:3]overlay[v]" -map "[v]" -map 0:a -c:v h264_nvenc -f mpegts subtitle_test.ts

如何使用ffmpeg(v4.3)和cuda，硬件加速器在视频上刻录基于字幕的图像？
谢谢你。

Answer 1

你可以像下面这样，ffmpeg -y -hwaccel cuvid -hwaccel_output_format cuda -i c:\outfile.mp4 -i C:\overlay.png -filter_complex "[0:v]scale_cuda=1280:720,format=cuda [base]; [1:v]format=nv12,hwupload_cuda,scale_cuda=1280:720,format=cuda [ovly1]; [base][ovly1] overlay_cuda,scale_cuda=1280:720,hwdownload,format=nv12[marked]" -map "[marked]" -vcodec h264_nvenc -map 0:1 -acodec copy testfile.mp4特别是，overlay_cuda 仅支持以下覆盖格式:NV12,YUV420P,YUVA420P来自 https://github.com/FFmpeg/FFmpeg/blob/4976b102d89787d59a3aaa438b76c62aec86ad5a/libavfilter/vf_overlay_cuda.c