Recherche avancée

Sur d’autres sites (7625)

• AVFrame confusion between width, height and linesize

  27 août 2019, par Lucas Zanella

  I’m trying to understand AVFrame, specially the linesize property.

  Here’s my decoded AVFrame properties :

  width = 640 
  
  height= 360
  
  
  
  linesize[0] = 640
  
  linesize[1] = 320
  
  linesize[2] = 320

  It’s an YUV 420 planar image (AV_PIX_FMT_YUVJ420P)

  I’m reading this code and here’s the part that deals with AVFrame properties :

  int linesize = qAbs(m_format.renderFrame->linesize[i]);
  
  AVRational widthRational = params.yuvwidths[i];
  
  AVRational heightRational = params.yuvheights[i];
  
  int width = linesize * widthRational.num / widthRational.den;
  
  int height = m_format.renderFrame->height * heightRational.num / heightRational.den;
  
  glTexImage2D ( GL_TEXTURE_2D, 0, params.yuvInternalformat[i],width ,height, 0, params.yuvGlFormat[i], params.dataType, NULL);

  Where, for YUV420P, widthRational and heightRational are 1/1 and 1/1 for i=0, and 1/2 and 1/2 for i = 1,2. And yuvInternalformat and yuvGlFormat are always GL_RED.

  There are a few things that I can’t understand in ths code :

  Why he takes an absolute value in linesize ? Can linesize be negative ? There’s nothing about negative values in the documentation. I understand why he makes the fraction multiplication in height, but why in linesize ? Shouldn’t linesize be the actual width of the planar image and thus require no multiplication ?

  So what is linesize, how should width and height be calculated in order to use glTexImage2D ?

• avcodec/v4l2_buffers : read height/width from the proper context

  30 août 2019, par Aman Gupta

  avcodec/v4l2_buffers : read height/width from the proper context
  Frames are generally dequeued into capture buffers, so using
  
  the output properties here was incorrect. It happened to work
  
  fine for decoding, since the output/capture buffers have the same
  
  dimensions.
  For the v4l2 scaler, the dimensions can be different between output
  
  and capture. Using the buffer's associated context makes this code
  
  work correctly regardless of where the frame is coming from.
  Signed-off-by : Aman Gupta <aman@tmm1.net>
  

  • [DH] libavcodec/v4l2_buffers.c

• FFMPEG - height not divisible by 2

  27 septembre 2019, par iSaBo

  i have an problem with ffmpeg. i would like to format a image sequence into a video. I use the followed command for this :

  ffmpeg -framerate 24 -i image%04d.jpeg Project.mp4 -vf "pad=ceil(iw/2)*2:ceil(ih/2)*2"

  i have 4 pictures :

  • image0001.jpeg
  • image0002.jpeg
  • image0003.jpeg
  • image0004.jpeg

  With this command, i get the following error :

  [libx264 @ 000001f12e7a0540] height not divisible by 2 (1200x1599) 
  
  Error initializing output stream 0:0 -- Error while opening encoder for output stream #0:0 - maybe 
  
  incorrect parameters such as bit_rate, rate, width or height
  
  Conversion failed!

  Can someone tell me why this mistake comes and how can I fix it ?

  Thanks