Promote colorspace information from decoder to swscale in FFmpeg backend.

This commit is contained in:
Alexander Smorkalov 2023-09-18 17:14:26 +03:00
parent 157b0e7760
commit 7c740f555a

View File

@ -1636,6 +1636,7 @@ bool CvCapture_FFMPEG::retrieveFrame(int flag, unsigned char** data, int* step,
return false;
CV_LOG_DEBUG(NULL, "Input picture format: " << av_get_pix_fmt_name((AVPixelFormat)sw_picture->format));
CV_LOG_DEBUG(NULL, "Input picture colorspace: " << av_get_colorspace_name(sw_picture->colorspace));
const AVPixelFormat result_format = convertRGB ? AV_PIX_FMT_BGR24 : (AVPixelFormat)sw_picture->format;
switch (result_format)
{
@ -1695,6 +1696,12 @@ bool CvCapture_FFMPEG::retrieveFrame(int flag, unsigned char** data, int* step,
frame.height = video_st->CV_FFMPEG_CODEC_FIELD->height;
frame.data = rgb_picture.data[0];
frame.step = rgb_picture.linesize[0];
const int* colorspace_coeffs = sws_getCoefficients(sw_picture->colorspace);
sws_setColorspaceDetails(img_convert_ctx,
colorspace_coeffs, sw_picture->color_range,
colorspace_coeffs, convertRGB? 0: sw_picture->color_range,
0, 1<<16, 1<<16);
}
sws_scale(