Exploiting Completeness and Uncertainty of Pseudo Labels for Weakly Supervised Video Anomaly Detection

TL;DR

This paper introduces a novel framework for weakly supervised video anomaly detection that leverages completeness and uncertainty properties of pseudo labels to enhance self-training, resulting in improved detection performance.

Contribution

It proposes a multi-head classification module with diversity loss and an iterative uncertainty refinement strategy to better generate and improve pseudo labels for anomaly detection.

Findings

Outperforms state-of-the-art on UCF-Crime, TAD, XD-Violence datasets

Effective pseudo label coverage of abnormal events

Enhanced pseudo label quality through uncertainty refinement

Abstract

Weakly supervised video anomaly detection aims to identify abnormal events in videos using only video-level labels. Recently, two-stage self-training methods have achieved significant improvements by self-generating pseudo labels and self-refining anomaly scores with these labels. As the pseudo labels play a crucial role, we propose an enhancement framework by exploiting completeness and uncertainty properties for effective self-training. Specifically, we first design a multi-head classification module (each head serves as a classifier) with a diversity loss to maximize the distribution differences of predicted pseudo labels across heads. This encourages the generated pseudo labels to cover as many abnormal events as possible. We then devise an iterative uncertainty pseudo label refinement strategy, which improves not only the initial pseudo labels but also the updated ones obtained by the desired classifier in the second stage. Extensive experimental results demonstrate the proposed method performs favorably against state-of-the-art approaches on the UCF-Crime, TAD, and XD-Violence benchmark datasets.