Yuzhi Huang, Chenxin Li, Haitao Zhang +9

Fei Li, Wenxuan Liu, Jingjing Chen +4

Snehashis Majhi, Giacomo D'Amicantonio, Antitza Dantcheva +5

Han Hu, Wenli Du, Peng Liao +2

Muchao Ye, Weiyang Liu, Pan He

Yu Wang, Shengjie Zhao

Weakly supervised video anomaly detection (WS-VAD) involves identifying the temporal intervals that contain anomalous events in untrimmed videos, where only video-level annotations are provided as supervisory signals. However, a key limitation persists in WS-VAD, as dense frame-level annotations are absent, which often leaves existing methods struggling to learn anomaly semantics effectively. To address this issue, we propose a novel framework named LAS-VAD, short for Learning Anomaly Semantics for WS-VAD, which integrates anomaly-connected component mechanism and intention awareness mechanism