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

Chen Xu, Chunguo Li, Hongjie Xing

Jash Dalvi, Ali Dabouei, Gunjan Dhanuka +1

Ali Karami, Thi Kieu Khanh Ho, Narges Armanfard

Anindya Sundar Das, Guansong Pang, Monowar Bhuyan

Sanggeon Yun, Ryozo Masukawa, Minhyoung Na +1