A motion vector(MV) refinement network and a residual refinement network taking use of the multiple reference frames.
Q: reference frames? motion vector field?
Code(not released?): https://github.com/JianpingLin/M-LVC_CVPR2020
M-LVC: Multiple Frames Prediction for Learned Video Compression(CVPR 2020) - Video Compression Paper Reading
猜你喜欢
转载自www.cnblogs.com/hhhhhxh/p/13198687.html
今日推荐
周排行