IntegralAction: Pose-driven Feature Integration for Robust Human Action Recognition in Videos.

Gyeongsik Moon Heeseung Kwon Kyoung Mu Lee Minsu Cho

Published in: CoRR (2020)

Keyphrases

pose invariant face recognition
partial occlusion
illumination variations
data driven
feature descriptors
computer vision
three dimensional
data integration
action recognition
face recognition
spatio temporal
d objects
pose estimation
robust recognition
action recognition in videos