Video: Masked Motion Modeling for Self-Supervised Video Representation Learning.
Xinyu SunPeihao ChenLiangwei ChenThomas H. LiMingkui TanChuang GanPublished in: CoRR (2022)
Keyphrases
- video representation
- space time
- video streams
- spatio temporal
- video analysis
- video database
- motion estimation
- video objects
- spatial and temporal
- image sequences
- key frames
- moving camera
- video content
- prior knowledge
- motion model
- human motion
- video retrieval
- camera motion
- video data
- dynamic textures
- motion trajectories
- feature vectors
- visual vocabulary
- video sequences