PAS-Net: Pose-based and Appearance-based Spatiotemporal Networks Fusion for Action Recognition.
Changzhen LiJie ZhangShiguang ShanXilin ChenPublished in: FG (2020)
Keyphrases
- action recognition
- human pose
- recognizing actions
- human actions
- activity recognition
- action detection
- human detection
- pose estimation
- computer vision
- body parts
- bag of words
- static images
- action classification
- recognition of human actions
- depth sensors
- space time
- spatial temporal
- bag of features
- recognizing human actions
- spatial and temporal
- moving objects
- independent subspace analysis
- view invariant
- object detection
- d objects
- action recognition in videos
- three dimensional