Introduction - If you have any usage issues, please Google them yourself
DCT coefficients of a video-based watermark embedding program, read the yuv file, extract the number of frames, and then do each frame DCT transform coefficients in the low-frequency changes, then do IDCT, then each frame component YUV format player.