Bounding-Box Trajectories Matter for Video Anomaly Detection

TL;DR

This paper introduces TrajVAD, a novel framework that leverages bounding-box trajectories with normalizing flows for improved video anomaly detection, outperforming pose-based methods on key datasets.

Contribution

The paper presents TrajVAD, a new trajectory-only model that effectively uses bounding-box trajectories for anomaly detection, surpassing existing pose-based approaches.

Findings

TrajVAD-T achieves 87.7% AP on ShanghaiTech, outperforming pose-based methods.

TrajVAD-P, incorporating pose info, reaches 88.6% AUROC on ShanghaiTech.

Bounding-box trajectories are shown to be an effective modality for anomaly detection.

Abstract

Video anomaly detection is critical for public safety and security, yet remains highly challenging despite extensive research due to large variations in appearance, viewpoint, and scene dynamics. Among existing approaches, human pose-based methods have emerged as a major line of research, showing strong performance since many anomalies in public datasets involve humans and pose representations are robust to appearance changes while providing compact motion descriptions. However, these methods often overlook bounding-box trajectories, although such information is inherently available in pose-based pipelines. In this paper, we explicitly leverage these trajectories as a primary anomaly cue. We present TrajVAD, a framework that models multi-class bounding-box trajectories using normalizing flows to learn normal kinematic patterns. Its trajectory-only variant (TrajVAD-T) eliminates pose estimation and surpasses all compared pose-based methods on ShanghaiTech in AP (87.7%), while achieving the best results on MSAD. An extended version (TrajVAD-P) incorporates pose information and further improves performance to 88.6% AUROC and 90.9% AP on ShanghaiTech, highlighting bounding-box trajectories as an effective yet underexplored modality for video anomaly detection.