1 paper
Le Yang, Junwei Han, Tao Zhao +3
Weakly supervised temporal action localization aims at learning the instance-level action pattern from the video-level labels, where a significant challenge is action-context confu…