[Libav-user] Converting AV_PIX_FMT_CUDA image (AVFrame) to AV_PIX_FMT_BGRA image

Omar Álvarez omar.alvarez at cinfo.es
Thu Feb 21 16:35:42 EET 2019

You could try:

sws_getContext(src_width, src_height, (AVPixelFormat) AV_PIX_FMT_RGB24,
                                      src_width, src_height,
(AVPixelFormat) AV_PIX_FMT_NV12,
                                      SWS_FAST_BILINEAR, NULL, NULL, NULL);

But for me it was too slow due to resizing code after the Meltdown bugs.

El jue., 21 feb. 2019 a las 14:12, Ахмед Анам (<
ahmedanam-at-mail.ru at ffmpeg.org>) escribió:

> Hello!
> I am trying to use libav API with hardware accelaration (CUDA). I need to
> transform an AVFrame with AV_PIX_FMT_CUDA pixel format to the frame
> with AV_PIX_FMT_BGRA pixel format. I tried to create Sws_Context for this
> purpose ( *my_sws_context = sws_getContext(width, height,
> NULL, NULL) ) *but the function always return NULL.
> What is the best way to do this? Are there some examples?
> --
> С уважением, Ахмед Анам.
> Best regards, Akhmed Anam.
> _______________________________________________
> Libav-user mailing list
> Libav-user at ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/libav-user
> To unsubscribe, visit link above, or email
> libav-user-request at ffmpeg.org with subject "unsubscribe".
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ffmpeg.org/pipermail/libav-user/attachments/20190221/fc005bb2/attachment.html>

More information about the Libav-user mailing list