activity
20222025
most citedTwo-stream joint matching method based on contrastive learning for few-shot action recognition

1 citations · 1 across the 3 of their papers we have counts for

collaborators
Showing 2022 · cs.CVShow all

1 paper · 2 filters

cs.CV2022

Forcing the Whole Video as Background: An Adversarial Learning Strategy for Weakly Temporal Action Localization

Ziqiang Li, Yongxin Ge, Jiaruo Yu +1

With video-level labels, weakly supervised temporal action localization (WTAL) applies a localization-by-classification paradigm to detect and classify the action in untrimmed vide…