Multi-modal Stream Fusion for Skeleton-Based Action Recognition.
Ruixuan PangRongchang LiTianyang XuXiaoning SongXiaojun WuPublished in: ICIG (3) (2023)
Keyphrases
- multi modal
- action recognition
- multi modality
- single modality
- human actions
- fusing multiple
- bag of words
- activity recognition
- data streams
- computer vision
- action classification
- audio visual
- recognizing actions
- semantic concepts
- video search
- high dimensional
- view invariant
- uni modal
- image annotation
- shape analysis
- object recognition
- recognizing human actions
- recognition of human actions
- medical images
- feature vectors