A Modular and Unified Framework for Detecting and Localizing Video Anomalies

TL;DR

This paper introduces MOVAD, a modular, real-time framework for detecting and localizing anomalies in videos, addressing limitations of existing methods by enhancing interpretability, cross-domain adaptivity, and evaluation metrics.

Contribution

The paper presents a novel transfer learning-based architecture and a comprehensive framework for online video anomaly detection and localization, improving upon current state-of-the-art methods.

Findings

Significantly outperforms existing approaches on benchmark datasets.

Provides a modular, interpretable, and real-time capable detection system.

Introduces a new performance metric suitable for real-time video anomaly detection.

Abstract

Anomaly detection in videos has been attracting an increasing amount of attention. Despite the competitive performance of recent methods on benchmark datasets, they typically lack desirable features such as modularity, cross-domain adaptivity, interpretability, and real-time anomalous event detection. Furthermore, current state-of-the-art approaches are evaluated using the standard instance-based detection metric by considering video frames as independent instances, which is not ideal for video anomaly detection. Motivated by these research gaps, we propose a modular and unified approach to the online video anomaly detection and localization problem, called MOVAD, which consists of a novel transfer learning based plug-and-play architecture, a sequential anomaly detector, a mathematical framework for selecting the detection threshold, and a suitable performance metric for real-time anomalous event detection in videos. Extensive performance evaluations on benchmark datasets show that the proposed framework significantly outperforms the current state-of-the-art approaches.