Recherche avancée

Sur d’autres sites (8494)

• Invalid data found when processing input. No start code is found

  23 septembre 2020, par Alex Blokha

  I am trying to concat video from many files.
But sometimes I receive an error : "Invalid data found when processing input".

  


  So, I am trying to iterate over files with ffprobe and find invalid files.
Here is the error.

  


  ffprobe started on 2020-06-28 at 16:06:39
Report written to "ffreport.log"
Log level: 32
Command line:
"f:\\prog\\ffmpeg\\bin\\ffprobe.exe" -i Rec496_20200423123337_A_1.avi
ffprobe version git-2020-03-23-ba698a2 Copyright (c) 2007-2020 the FFmpeg developers
  built with gcc 9.2.1 (GCC) 20200122
  configuration: --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --e
nable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrn
b --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-lib
soxr --enable-libsrt --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --e
nable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-l
ibvidstab --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-
libaom --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. 42.101 / 56. 42.101
  libavcodec     58. 76.100 / 58. 76.100
  libavformat    58. 42.100 / 58. 42.100
  libavdevice    58.  9.103 / 58.  9.103
  libavfilter     7. 77.100 /  7. 77.100
  libswscale      5.  6.101 /  5.  6.101
  libswresample   3.  6.100 /  3.  6.100
  libpostproc    55.  6.100 / 55.  6.100
[avi @ 000002510398e1c0] non-interleaved AVI
[h264 @ 00000251039a0480] missing picture in access unit with size 256
[extract_extradata @ 0000025103996740] No start code is found.
Rec496_20200423123337_A_1.avi: Invalid data found when processing input


  


  Questions :

  


  

  1. How can this be fixed, so I don't need to iterate with ffprobe and just use concat (command line : f :\prog\ffmpeg\bin\ffmpeg.exe -f concat -safe 0 -i mylist.txt -c copy output-xx.avi) ?

  


  2. If not, how can I receive an error in my .bat file and move "invalid" file to "Error" folder ?

  


  


• FFmpeg image2pipe write buffer wait until done

  20 décembre 2020, par Michael Joseph Aubry

  I'm extracting frames from an external source and passing it in as a buffer to FFMPEG using image2pipe and -i -

  


  const childProcess = spawn(ffmpeg, [
  "-y",
  "-f",
  "image2pipe",
  "-i",
  "-",
  "-vcodec",
  "libx264",
  "-pix_fmt",
  "yuv420p",
  output
]);


  


  Then I have a loop that does the job.

  


  for (let i = 0; i < 250; i++) {
  // ...await
}


  


  Inside the promise

  


  // ... do the job to get buffer

childProcess.stdin.write(frame); // frame === buffer

// frame done
resolve("success!");


  


  The problem is in some videos the frames jump and is janky. This is because FFmpeg is not fully done writing to the file before moving onto the next frame.

  


  Is there a way to write a buffer to a file through FFmpeg and make sure the frame is done writing before moving on ?

  


  Some more information

  


  Here is the source file https://s3.us-west-2.amazonaws.com/storycreator.v2.uploads/ckigi4kro00160vlfjmt74afp

  


  Here is the rendered file https://s3.us-west-2.amazonaws.com/storycreator.testing/607715f0-3ab9-11eb-a139-3bb84618c6c5.mp4?t=1607585343922

  


  Here are logs

  


  2020-12-10T07:28:48.942Z    0ae0c435-54d3-416f-9d1a-8ddf595a7e83    INFO    frame=  130 fps= 16 q=-1.0 Lsize=      83kB time=00:00:05.08 bitrate= 133.6kbits/s speed=0.641x    video:80kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 2.928688%


  


  2020-12-10T07:28:48.942Z    0ae0c435-54d3-416f-9d1a-8ddf595a7e83    INFO    [libx264 @ 0x640d6c0] frame I:1     Avg QP:15.47  size: 68227[libx264 @ 0x640d6c0] frame P:33    Avg QP:15.07  size:   246[libx264 @ 0x640d6c0] frame B:96    Avg QP:18.75  size:    56[libx264 @ 0x640d6c0] consecutive B-frames:  1.5%  0.0%  0.0% 98.5%


  


  2020-12-10T07:28:48.943Z    0ae0c435-54d3-416f-9d1a-8ddf595a7e83    INFO    [libx264 @ 0x640d6c0] mb I  I16..4: 24.6% 60.5% 14.9%[libx264 @ 0x640d6c0] mb P  I16..4:  0.0%  0.2%  0.0%  P16..4:  0.8%  0.0%  0.0%  0.0%  0.0%    skip:98.9%[libx264 @ 0x640d6c0] mb B  I16..4:  0.0%  0.0%  0.0%  B16..8:  0.3%  0.0%  0.0%  direct: 0.0%  skip:99.7%  L0:26.7% L1:73.3% BI: 0.0%[libx264 @ 0x640d6c0] 8x8 transform intra:62.0% inter:83.6%[libx264 @ 0x640d6c0] coded y,uvDC,uvAC intra: 49.2% 44.3% 32.1% inter: 0.0% 0.2% 0.0%[libx264 @ 0x640d6c0] i16 v,h,dc,p: 53% 38%  7%  2%[libx264 @ 0x640d6c0] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 26% 24% 29%  3%  3%  3%  5%  3%  4%[libx264 @ 0x640d6c0] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 37% 26% 14%  3%  4%  5%  4%  4%  3%[libx264 @ 0x640d6c0] i8c dc,h,v,p: 57% 27% 13%  3%[libx264 @ 0x640d6c0] Weighted P-Frames: Y:0.0% UV:0.0%[libx264 @ 0x640d6c0] ref P L0: 95.1%  1.2%  3.1%  0.6%[libx264 @ 0x640d6c0] ref B L0: 48.1% 51.3%  0.7%[libx264 @ 0x640d6c0] ref B L1: 97.3%  2.7%[libx264 @ 0x640d6c0] kb/s:125.75


  


  2020-12-10T07:28:48.944Z    0ae0c435-54d3-416f-9d1a-8ddf595a7e83    INFO    [cache @ 0x5f57940] Statistics, cache hits:0 cache misses:3551


  


• Writing linear float range to openEXR turns out non linear

  3 avril 2022, par Chryfi

  I am writing the linearized depth buffer of a game to openEXR using FFmpeg. Unfortunately, FFmpeg does not adhere to the openEXR file specification fully (like allowing unsigned integer for one channel) so I am writing one float channel to openEXR, which is put into the green channel with this command -f rawvideo -pix_fmt grayf32be -s %WIDTH%x%HEIGHT% -r %FPS% -i - -vf %DEFVF% -preset ultrafast -tune zerolatency -qp 6 -compression zip1 -pix_fmt gbrpf32le %NAME%_depth_%d.exr.

  


  The float range is from 0F to 1F and it is linear. I can confirm that the calculation and linearization is correct by testing 16 bit integer (per pixel component) PNG in Blender compositor. The 16 bit integer data is written like this short s = (short) (linearzieDepth(depth) * (Math.pow(2,16) - 1)) whereas for float the linearized value is directly written to OpenEXR without multiplying with a value.

  


  However, when viewing the openEXR file it doesn't have the same "gradient" as the 16 bit png... when viewing them side by side, it appears as if the values near 0 are not linear, and they are not as dark as they should be like in the 16 bit png.
(And yes, I set the image node to linear), and comparing it with 3d tracking data from the game I cant reproduce the depth and cant mask things using the depth buffer where as with the png I can.

  


  How is it possible for a linear float range to turn out so different to a linear integer range in an image ?