Weakly Supervised Video Anomaly Detection with Anomaly-Connected Components and Intention Reasoning

TL;DR

This paper introduces LAS-VAD, a novel weakly supervised video anomaly detection framework that uses anomaly-connected components and intention reasoning to improve detection accuracy without dense annotations.

Contribution

LAS-VAD integrates semantic grouping and intention awareness mechanisms, along with anomaly attribute information, to enhance weakly supervised video anomaly detection.

Findings

Outperforms state-of-the-art on XD-Violence and UCF-Crime datasets.

Achieves significant accuracy improvements in anomaly localization.

Effectively models anomaly semantics with attribute guidance.

Abstract

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. The former is designed to assign video frames into distinct semantic groups within a video, and frame segments within the same group are deemed to share identical semantic information. The latter leverages an intention-aware strategy to distinguish between similar normal and abnormal behaviors (e.g., taking items and stealing). To further model the semantic information of anomalies, as anomaly occurrence is accompanied by distinct characteristic attributes (i.e., explosions are characterized by flames and thick smoke), we additionally incorporate anomaly attribute information to guide accurate detection. Extensive experiments on two benchmark datasets, XD-Violence and UCF-Crime, demonstrate that our LAS-VAD outperforms current state-of-the-art methods with remarkable gains.