LAE-Net: Light and Efficient Network for Compressed Video Action Recognition.
Jinxin GuoJiaqiang ZhangXiaojing ZhangMing MaPublished in: MMM (2) (2023)
Keyphrases
- action recognition
- compressed video
- human actions
- activity recognition
- recognizing human actions
- action classification
- computer vision
- spatial temporal
- human detection
- body parts
- super resolution
- compressed domain
- recognition of human actions
- human activities
- video quality
- data hiding
- video content
- video streams
- video frames
- low level