Representation Learning for Compressed Video Action Recognition via Attentive Cross-modal Interaction with Motion Enhancement.
Bing LiJiaxin ChenDongming ZhangXiuguo BaoDi HuangPublished in: IJCAI (2022)
Keyphrases
- action recognition
- recognizing human actions
- perceptual information
- recognition of human actions
- cross modal
- compressed video
- action recognition in videos
- human actions
- bag of words
- activity recognition
- multi modal
- motion estimation
- computer vision
- object tracking
- human computer interaction
- spatio temporal
- multiscale