Sign in

Self-supervised Video Representation Learning with Motion-Aware Masked Autoencoders.

Haosen YangDeng HuangBin WenJiannan WuHongxun YaoYi JiangXiatian ZhuZehuan Yuan
Published in: CoRR (2022)
Keyphrases
  • video representation
  • spatio temporal
  • space time
  • prior knowledge
  • motion estimation
  • feature points
  • motion model