Recherche avancée

Sur d’autres sites (10583)

• avcodec/frame_thread_encoder : Free AVCodecContext structure on error during init

  14 août 2021, par Michael Niedermayer

  avcodec/frame_thread_encoder : Free AVCodecContext structure on error during init
  Fixes : MemLeak
  
  Fixes : 8281
  
  Fixes : PoC_option158.jpg
  
  Fixes : CVE-2020-22037
  Reviewed-by : Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
  
  Signed-off-by : Michael Niedermayer <michael@niedermayer.cc>
  

  • [DH] libavcodec/frame_thread_encoder.c
  • [DH] libavcodec/frame_thread_encoder.h

• mp4 created by ffmpeg will not play with IPython.display.Video

  19 novembre 2020, par Austin

  I'm creating a mp4 video from jpegs with ffmpeg, using the following command :

  


  ffmpeg -y -threads 0 -f image2 -i jpegs/%05d.jpg -framerate 10 video.mp4

  


  The resulting video will play fine with VLC, but will not play in a Jupyter notebook via :

  


  from IPython.display import Video
Video('video.mp4')


  


  This notebook video functionality has worked for me in the past, so I assume there's something about the codec/mime type that the Video function can't resolve ?

  


  ---

  


  Requested update

  


  New command I've tried :

  


  ffmpeg -y -framerate 10 -i jpegs/%05d.jpg -vf format=yuv420p -movflags +faststart video.mp4

  


  Video still plays fine in VLC and still does not play in Jupyter. I've looked at the duplicates this is marked as and it does not actually seem to be the same issue..

  


  Here's the output (I've removed the framerate and threads flags to simplify) :

  


  algo-1-poqk5_1  | 2020-11-18 21:19:36 [INFO]: ffmpeg output:
algo-1-poqk5_1  | 2020-11-18 21:19:36 [INFO]: ffmpeg version 4.0 Copyright (c) 2000-2018 the FFmpeg developers
algo-1-poqk5_1  |   built with gcc 7.2.0 (crosstool-NG fa8859cb)
algo-1-poqk5_1  |   configuration: --prefix=/opt/conda --cc=/opt/conda/conda-bld/ffmpeg_1531088893642/_build_env/bin/x86_64-conda_cos6-linux-gnu-cc --disable-doc --enable-shared --enable-static --enable-zlib --enable-pic --enable-gpl --enable-version3 --disable-nonfree --enable-hardcoded-tables --enable-avresample --enable-libfreetype --disable-openssl --disable-gnutls --enable-libvpx --enable-pthreads --enable-libopus --enable-postproc --disable-libx264
algo-1-poqk5_1  |   libavutil      56. 14.100 / 56. 14.100
algo-1-poqk5_1  |   libavcodec     58. 18.100 / 58. 18.100
algo-1-poqk5_1  |   libavformat    58. 12.100 / 58. 12.100
algo-1-poqk5_1  |   libavdevice    58.  3.100 / 58.  3.100
algo-1-poqk5_1  |   libavfilter     7. 16.100 /  7. 16.100
algo-1-poqk5_1  |   libavresample   4.  0.  0 /  4.  0.  0
algo-1-poqk5_1  |   libswscale      5.  1.100 /  5.  1.100
algo-1-poqk5_1  |   libswresample   3.  1.100 /  3.  1.100
algo-1-poqk5_1  |   libpostproc    55.  1.100 / 55.  1.100
algo-1-poqk5_1  | Input #0, image2, from '/opt/ml/model/outputs/01-IR.mp4_1/%05d.jpg':
algo-1-poqk5_1  |   Duration: 00:00:12.00, start: 0.000000, bitrate: N/A
algo-1-poqk5_1  |     Stream #0:0: Video: mjpeg, yuvj420p(pc, bt470bg/unknown/unknown), 1920x1080 [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 25 tbn, 25 tbc
algo-1-poqk5_1  | Stream mapping:
algo-1-poqk5_1  |   Stream #0:0 -> #0:0 (mjpeg (native) -> mpeg4 (native))
algo-1-poqk5_1  | Press [q] to stop, [?] for help
algo-1-poqk5_1  | [swscaler @ 0x55b8cc6fcc40] deprecated pixel format used, make sure you did set range correctly
algo-1-poqk5_1  | Output #0, mp4, to '/opt/ml/model/outputs/01-IR.mp4_1/video.mp4':
algo-1-poqk5_1  |   Metadata:
algo-1-poqk5_1  |     encoder         : Lavf58.12.100
algo-1-poqk5_1  |     Stream #0:0: Video: mpeg4 (mp4v / 0x7634706D), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 200 kb/s, 25 fps, 12800 tbn, 25 tbc
algo-1-poqk5_1  |     Metadata:
algo-1-poqk5_1  |       encoder         : Lavc58.18.100 mpeg4
algo-1-poqk5_1  |     Side data:
algo-1-poqk5_1  |       cpb: bitrate max/min/avg: 0/0/200000 buffer size: 0 vbv_delay: -1
algo-1-poqk5_1  | frame=   31 fps=0.0 q=31.0 size=     768kB time=00:00:01.20 bitrate=5242.8kbits/s speed=2.33x    
algo-1-poqk5_1  | frame=   64 fps= 63 q=31.0 size=    1280kB time=00:00:02.52 bitrate=4161.0kbits/s speed=2.48x    
algo-1-poqk5_1  | frame=   98 fps= 64 q=31.0 size=    1792kB time=00:00:03.88 bitrate=3783.5kbits/s speed=2.53x    
algo-1-poqk5_1  | frame=  131 fps= 64 q=31.0 size=    2304kB time=00:00:05.20 bitrate=3629.7kbits/s speed=2.55x    
algo-1-poqk5_1  | frame=  164 fps= 64 q=31.0 size=    2560kB time=00:00:06.52 bitrate=3216.5kbits/s speed=2.56x    
algo-1-poqk5_1  | frame=  198 fps= 65 q=31.0 size=    3072kB time=00:00:07.88 bitrate=3193.6kbits/s speed=2.58x    
algo-1-poqk5_1  | frame=  232 fps= 65 q=31.0 size=    3328kB time=00:00:09.24 bitrate=2950.6kbits/s speed= 2.6x    
algo-1-poqk5_1  | frame=  266 fps= 66 q=31.0 size=    3840kB time=00:00:10.60 bitrate=2967.7kbits/s speed=2.62x    
algo-1-poqk5_1  | frame=  299 fps= 66 q=31.0 size=    4096kB time=00:00:11.92 bitrate=2815.0kbits/s speed=2.62x    
algo-1-poqk5_1  | [mp4 @ 0x55b8cc658900] Starting second pass: moving the moov atom to the beginning of the file
algo-1-poqk5_1  | frame=  300 fps= 65 q=31.0 Lsize=    4202kB time=00:00:11.96 bitrate=2877.9kbits/s speed=2.61x    
algo-1-poqk5_1  | video:4199kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.052020%


  


• ffmpeg "Underestimated required buffer size"

  10 avril 2021, par Kajuna

  I'm seeing this error on some Windows machine when converting an 8K png stream to ProRes 4444 with ffmpeg. The command looks like

  


  


  ffmpeg -f image2pipe -framerate 30000/1001 -i - -c:v prores_ks -y output.mov

  


  


  And the output

  


  


  ffmpeg version git-2020-06-17-0b3bd00 Copyright (c) 2000-2020 the
FFmpeg developers built with gcc 9.3.1 (GCC) 20200523
  
configuration : —enable-gpl —enable-version3 —enable-sdl2
—enable-fontconfig —enable-gnutls —enable-iconv —enable-libass —enable-libdav1d —enable-libbluray —enable-libfreetype —enable-libmp3lame —enable-libopencore-amrnb —enable-libopencore-amrwb —enable-libopenjpeg —enable-libopus —enable-libshine —enable-libsnappy —enable-libsoxr —enable-libsrt —enable-libtheora —enable-libtwolame —enable-libvpx —enable-libwavpack —enable-libwebp —enable-libx264 —enable-libx265 —enable-libxml2 —enable-libzimg —enable-lzma —enable-zlib —enable-gmp —enable-libvidstab —enable-libvmaf —enable-libvorbis —enable-libvo-amrwbenc —enable-libmysofa —enable-libspeex —enable-libxvid —enable-libaom —disable-w32threads —enable-libmfx —enable-ffnvcodec —enable-cuda-llvm —enable-cuvid —enable-d3d11va —enable-nvenc —enable-nvdec —enable-dxva2 —enable-avisynth —enable-libopenmpt —enable-amf libavutil 56. 55.100 / 56. 55.100 libavcodec 58. 92.100 / 58. 92.100 libavformat 58. 46.101 / 58. 46.101 libavdevice 58. 11.100 / 58. 11.100 libavfilter 7. 86.100 / 7. 86.100 libswscale 5. 8.100 /
5. 8.100 libswresample 3. 8.100 / 3. 8.100 libpostproc 55. 8.100 / 55. 8.100 [image2pipe @ 000002245a96da40] Stream #0 : not enough frames to estimate rate ; consider increasing probesize Input
#0, image2pipe, from 'pipe :' : Duration : N/A, bitrate : N/A
Stream #0:0 : Video : png, rgba(pc), 7680x3840, 29.97 tbr, 29.97 tbn, 29.97 tbc Stream mapping : Stream #0:0 -> #0:0 (png (native) ->
prores (prores_ks)) frame= 0 fps=0.0 q=0.0 size= 0kB
time=-577014:32:22.77 bitrate= -0.0kbits/s speed=N/A frame= 0
fps=0.0 q=0.0 size= 0kB time=-577014:32:22.77 bitrate=
-0.0kbits/s speed=N/A frame= 0 fps=0.0 q=0.0 size= 0kB time=-577014:32:22.77 bitrate= -0.0kbits/s speed=N/A frame= 0
fps=0.0 q=0.0 size= 0kB time=-577014:32:22.77 bitrate=
-0.0kbits/s speed=N/A frame= 0 fps=0.0 q=0.0 size= 0kB time=-577014:32:22.77 bitrate= -0.0kbits/s speed=N/A [prores_ks @
000002245aaaf080] Autoselected 4:4:4:4 profile because of the used
input colorspace. It can be overridden through -profile option.
[prores_ks @ 000002245aaaf700] Autoselected 4:4:4:4 profile because of
the used input colorspace. It can be overridden through -profile
option. [prores_ks @ 000002245aa1e7c0] Autoselected 4:4:4:4 profile
because of the used input colorspace. It can be overridden through
-profile option. [prores_ks @ 000002245aa1d580] Autoselected 4:4:4:4 profile because of the used input colorspace. It can be overridden
through -profile option. [prores_ks @ 000002245aa1ec40] Autoselected
4:4:4:4 profile because of the used input colorspace. It can be
overridden through -profile option. [prores_ks @ 000002245aa1da00]
Autoselected 4:4:4:4 profile because of the used input colorspace. It
can be overridden through -profile option. [prores_ks @
000002245aa1f540] Autoselected 4:4:4:4 profile because of the used
input colorspace. It can be overridden through -profile option.
[prores_ks @ 000002245aa1de80] Autoselected 4:4:4:4 profile because of
the used input colorspace. It can be overridden through -profile
option. [prores_ks @ 000002245a972b40] Autoselected 4:4:4:4 profile
because of the used input colorspace. It can be overridden through
-profile option. Output #0, mov, to 'output.mov' : Metadata :
encoder : Lavf58.46.101
Stream #0:0 : Video : prores (prores_ks) (ap4h / 0x68347061), yuva444p10le, 7680x3840, q=2-31, 200 kb/s, 29.97 fps, 30k tbn, 29.97
tbc
Metadata :
encoder : Lavc58.92.100 prores_ks frame= 1 fps=0.2 q=0.0 size= 0kB time=00:00:00.00 bitrate=N/A speed= 0x
  
frame= 2 fps=0.3 q=0.0 size= 0kB time=00:00:00.00 bitrate=N/A
speed= 0x frame= 3 fps=0.4 q=0.0 size= 0kB
time=00:00:00.00 bitrate=N/A speed= 0x frame= 4 fps=0.4 q=0.0
size= 0kB time=00:00:00.00 bitrate=N/A speed= 0x frame=
  
5 fps=0.5 q=0.0 size= 0kB time=00:00:00.00 bitrate=N/A speed=
  
0x [prores_ks @ 000002245aaaf080] Underestimated required buffer
size. Video encoding failed [prores_ks @ 000002245aaaf700]
Underestimated required buffer size. [prores_ks @ 000002245aa1e7c0]
Underestimated required buffer size. [prores_ks @ 000002245aa1d580]
Underestimated required buffer size. [prores_ks @ 000002245aa1ec40]
Underestimated required buffer size. [prores_ks @ 000002245aa1da00]
Underestimated required buffer size. Conversion failed !

  


  


  How could I debug this considering I don't have access to the problematic machine (a user does) ? Could the "not enough frames to estimate rate ; consider increasing probesize" message be related to the problem ?

  


  Thanks

  


  Edit : I increased probesize and that warning disappeared, but the conversion keeps failing with the same error message "Underestimated required buffer size". I can now reproduce this on one of my Windows machines.

  


  Edit 2 : Not a solution but I found out that this problem is gone if using ffmpeg 4.2.3 and was introduced with 4.3