Interact before Align: Leveraging Cross-Modal Knowledge for Domain Adaptive Action Recognition.

Lijin Yang Yifei Huang Yusuke Sugano Yoichi Sato

Published in: CVPR (2022)

Keyphrases

action recognition
cross modal
human actions
bag of words
computer vision
multi modal
activity recognition
higher level
recognizing human actions
human activities
visual data
multimedia retrieval
feature extraction
semantic information