MTFD: Multi-Teacher Fusion Distillation for Compressed Video Action Recognition.
Jinxin GuoJiaqiang ZhangShaoJie LiXiaojing ZhangMing MaPublished in: ICASSP (2023)
Keyphrases
- action recognition
- compressed video
- human actions
- bag of words
- video quality
- human detection
- activity recognition
- video streams
- super resolution
- computer vision
- compressed domain
- spatial temporal
- bitstream
- video content
- action classification
- recognizing human actions
- data hiding
- body parts
- error correction
- video frames
- video coding
- bit rate
- quality assessment
- human pose
- video objects
- moving objects