Spatial Temporal Attention Graph Convolutional Networks with Mechanics-Stream for Skeleton-Based Action Recognition.
Katsutoshi ShirakiTsubasa HirakawaTakayoshi YamashitaHironobu FujiyoshiPublished in: ACCV (5) (2020)
Keyphrases
- spatial temporal
- action recognition
- human actions
- bag of words
- activity recognition
- computer vision
- action classification
- graph structure
- recognizing human actions
- recognition of human actions
- data streams
- multiscale
- bag of features
- space time interest points
- image classification
- low level
- high level
- video database
- human activities
- eye movements
- feature extraction