Structured Evidence Selection for Weakly Supervised Video Anomaly Detection
2026-07-11 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the problem of detecting unusual events in videos using only overall video labels, which makes pinpointing exact anomalies challenging. They propose a method called SESAD that breaks down videos into smaller parts and selects important clues by considering both the scene and actions, reducing confusion from background details. Additionally, they introduce a simple geometric tool to better distinguish normal and abnormal patterns based on their relationships in space. Their approach performs well on multiple benchmark datasets while remaining efficient.
Weakly supervised learningVideo anomaly detectionClip-level featuresSemantic entanglementContext-conditioned selectionGeometric discriminationDual-prototype embeddingAUC (Area Under Curve)UBnormal datasetShanghaiTech dataset
Authors
Chenglizhao Chen, Tianxiang Nan, Wen Li, Xinyu Liu, Guisheng Zhang, Mengke Song, Xiaomin Yu
Abstract
Weakly supervised video anomaly detection relies solely on video-level labels for training, making it difficult to accurately localize anomalous events in complex scenes. In real-world videos, anomalous behaviors exhibit large variations in appearance and temporal duration, while scene appearance and action dynamics are often tightly entangled. Consequently, existing models tend to rely on scene-related statistical cues rather than true behavioral deviations, resulting in unstable detection performance. To address this challenge, we propose a Structured Evidence Selection framework (SESAD) that reformulates anomaly detection as a structured reasoning process over clip-level visual evidence. Instead of directly mapping aggregated features to anomaly scores, SESAD reorganizes clip representations into semantically structured candidate evidence and performs context-conditioned selection under scene and action constraints. This mechanism adaptively emphasizes anomaly-relevant semantics while suppressing scene interference, thereby alleviating semantic entanglement under weak supervision. Furthermore, we introduce a lightweight geometric discrimination module that constructs a dual-prototype structure in the embedding space, enabling anomaly decisions through relative geometric relations. Extensive experiments on UBnormal, ShanghaiTech, and UCF-Crime show that SESAD achieves 67.92, 97.99, and 88.46 AUC, respectively, while maintaining high computational efficiency and overall consistently stable anomaly discrimination.