Partially Occluded Skeleton Action Recognition Based on Multi-stream Fusion Graph Convolutional Networks.

Dan Li Wuzhen Shi

Published in: CGI (2021)

Keyphrases

action recognition
partially occluded
multi stream
single image
bag of words
activity recognition
human actions
audio visual
computer vision
human activities
partial occlusion
spatio temporal
hidden markov models
multi modal
image content
foreground objects