Recherche avancée

Sur d’autres sites (7777)

• avcodec/nvenc : only unregister input resources when absolutely needed

  24 avril 2019, par Timo Rothenpieler

  avcodec/nvenc : only unregister input resources when absolutely needed
  This reverts nvenc to old behaviour, which in some super rare edge cases
  
  performs better.
  
  The implication of this is that any potential API user who relies on
  
  nvenc cleaning up every frames device resources after it's done using
  
  them will have to change their usage pattern.
  That should not be a problem, since pretty much every normal usage
  
  pattern automatically implies that surfaces are reused from a common
  
  pool, since constant re-allocation is also very expensive.
  

  • [DH] libavcodec/nvenc.c

• Revision 3d655805f2 : Use same intra prediction for all block size The commmit changed to use same in

  23 avril 2013, par Yaowu Xu

  Changed Paths :
   Modify /vp9/common/vp9_reconintra.c

  
   Modify /vp9/common/vp9_reconintra4x4.c

  
  
  Use same intra prediction for all block size

  The commmit changed to use same intra prediction function for all
  block sizes.

  Some details on the changes :
  1. All directional modes except DC/TM/V/H now have built-in filtering
  for all pixels with filter taps either (1, 2, 1)/4 or (1, 1)/2.
  2. Above edge get automatic extended to double width (bw*2), which
  makes a lot of the prediciton mode computation simpler.
  3. Same intra prediction function is called with different size
  for i4x4_pred and all other larger size.

  Overall, the change helped keyframe only coding for both cif size
  and std-hd size test sets by .5% consistently on all encodings.
  For normal coding with single/auto key frame, the change now also
  is consistently net positive for all encodings. The overall gains
  is about .15% on std-hd set.

  Change-Id : I01ceb31fbc73d49776262e6bdc06853b03bbd1d1

• convert 16bit grayscale raw video by ffmpeg for analysis in openCV

  15 février 2015, par user3730845

  I have 16bit grayscale stream of scientific data which I want to analyse in openCV. I tried to look for some conversion of data, however it seems that grayscale is a trouble for ffmpeg and available codecs.

  My idea is to use 3 colours of video wisely and stream my grayscales into 3 colours, one by one and then in openCV access each colour and do my analysis there. Is such a thing possible ?

  When I convert stream by : ffmpeg -f rawvideo -y -s 256x256 -pix_fmt gray16be -i rawfile.bin -an -vcodec ffv1 video.avi, I get weird colour space but I can open the result in openCV.

  If there is any other way to do such a thing can you point me towards libraries/howtos, ultimately I need to do thresholding, cross-correlation, edge filtering, centre of gravity types of functions and I need to do this as quickly as possible for datasets of >10GB

  My raw files/streams are set of 256x256 16bit b&w images, with 256 bytes header and gaps