Hyperpointnet for Point Cloud Sequence-Based 3D Human Action Recognition.
Xing LiQian HuangTianjin YangQianhan WuPublished in: ICME (2022)
Keyphrases
- action recognition
- point cloud
- static images
- human activities
- human actions
- computer vision
- structure from motion
- point sets
- surface reconstruction
- motion history images
- activity recognition
- action classification
- body parts
- human object interactions
- recognition of human actions
- stereo camera
- view invariant
- recognizing human actions
- recognizing actions
- action recognition in videos
- atomic actions
- human pose
- laser scanner
- distance function
- human computer interaction
- high quality