Multi-view graph convolution network for the recognition of human action with spatial and temporal occlusion problems.
Yang ChenLing WangDekun HuHong ChengPublished in: J. Vis. Commun. Image Represent. (2023)
Keyphrases
- multi view
- spatial and temporal
- spatio temporal
- space time
- human actions
- view invariant
- spatial temporal
- action recognition
- single view
- multiple views
- dynamic scenes
- d objects
- activity recognition
- three dimensional
- object recognition
- depth map
- human activities
- multiple cameras
- feature extraction
- motion recognition
- computer vision
- video sequences
- visual hull
- image sequences
- semi supervised
- low level
- surface reconstruction
- optical flow
- pedestrian detection