[FFmpeg-trac] #5822(avfilter:new): filter "scale_npp" fails to select correct algorithm (Nvidia CUDA/NPP scaler)
FFmpeg
trac at avcodec.org
Mon Sep 5 22:40:58 EEST 2016
#5822: filter "scale_npp" fails to select correct algorithm (Nvidia CUDA/NPP
scaler)
-------------------------------------+-------------------------------------
Reporter: sdack | Owner:
Type: defect | Status: new
Priority: normal | Component: avfilter
Version: git-master | Resolution:
Keywords: scale_npp, | Blocked By:
interp_algo, Nvidia, CUDA, NPP | Reproduced by developer: 0
Blocking: |
Analyzed by developer: 0 |
-------------------------------------+-------------------------------------
Comment (by sdack):
Although one can influence the result with a different pixel shift and
thereby produce distinguishable images from the algorithms does this also
cause a minor shift in the image itself, which isn't acceptable.
After reading the Nvidia forums did I notice a dev saying there were
bugs...
''"We (NVIDIA) recently found a bug in the rotation code which at some
scale factors can cause the last source pixel in a row or column to not be
sampled. ResizeSqrPixel contains a similar test which probably also needs
fixed. It would be great if you could send us an example of a failure
case. In the meantime, a possible work around would be to increase
oSrcROI.width and oSrcROI.height by 1. To be safe in all cases however,
this may require that you increase the memory allocated for your source
image by 1 in both width and height." (Posted 07/26/2016)
''
This doesn't sound quite like the issue here, but I am hoping an Nvidia
dev will take a look at this since there is already one known bug. I have
posted the problem on the Nvidia forums.
[https://devtalk.nvidia.com/default/topic/962808/gpu-accelerated-libraries
/nppiresizesqrpixel_8u_c1r-produces-low-quality-images/]
--
Ticket URL: <https://trac.ffmpeg.org/ticket/5822#comment:4>
FFmpeg <https://ffmpeg.org>
FFmpeg issue tracker
More information about the FFmpeg-trac
mailing list