Currently we are using the libav API to: 1) decode a 4K mp4 file 2) apply GLSL shader 3) encode to new 4K mp4 file
It works now using AVFrame but the copies to/from GPU is really slow. So how I make the frames stay in GPU memory and get a texture reference to them to apply a GLSL shader? If not possible with ffmpeg, can it be done using nvidia VideoCodecSDK? Regards, Morten
_______________________________________________ Libav-user mailing list [email protected] http://ffmpeg.org/mailman/listinfo/libav-user
