Motion-Driven Spatial and Temporal Adaptive High-Resolution Graph Convolutional Networks for Skeleton-Based Action Recognition.
Zengxi HuangYusong QinXiaobing LinTianlin LiuZhen-Hua FengYiguang LiuPublished in: IEEE Trans. Circuits Syst. Video Technol. (2023)
Keyphrases
- spatial and temporal
- action recognition
- high resolution
- space time
- human actions
- spatial temporal
- spatio temporal
- recognizing human actions
- recognition of human actions
- static images
- human movements
- body parts
- bag of words
- human detection
- computer vision
- action classification
- activity recognition
- low resolution
- motion features
- temporal domain
- video frames
- super resolution
- action recognition in videos
- action primitives
- view invariant
- motion patterns
- mid level
- dynamic textures
- motion history images
- spatial and temporal information
- video sequences
- pedestrian detection
- human activities
- three dimensional