Geometric Features Informed Multi-person Human-Object Interaction Recognition in Videos.
Tanqiu QiaoQianhui MenFrederick W. B. LiYoshiki KubotaniShigeo MorishimaHubert P. H. ShumPublished in: ECCV (4) (2022)
Keyphrases
- geometric features
- human object interactions
- action recognition
- human actions
- human pose
- human activities
- caption text
- complex events
- shape descriptors
- shape features
- geometric structure
- human motion
- activity recognition
- motion features
- object recognition
- video sequences
- video frames
- computer vision
- human body
- body parts
- spatio temporal
- feature extraction
- feature space