EVAL: Explainable Video Anomaly Localization

TL;DR

This paper introduces a novel, explainable framework for localizing anomalies in single-scene videos, providing human-understandable reasons for anomaly detection and demonstrating significant improvements over existing methods.

Contribution

It presents a new approach that combines deep learned representations with high-level, location-dependent models for explainable video anomaly localization.

Findings

Significant performance improvements on standard datasets

The model provides human-understandable explanations for anomalies

Effective detection of anomalies in various real-world scenes

Abstract

We develop a novel framework for single-scene video anomaly localization that allows for human-understandable reasons for the decisions the system makes. We first learn general representations of objects and their motions (using deep networks) and then use these representations to build a high-level, location-dependent model of any particular scene. This model can be used to detect anomalies in new videos of the same scene. Importantly, our approach is explainable - our high-level appearance and motion features can provide human-understandable reasons for why any part of a video is classified as normal or anomalous. We conduct experiments on standard video anomaly detection datasets (Street Scene, CUHK Avenue, ShanghaiTech and UCSD Ped1, Ped2) and show significant improvements over the previous state-of-the-art.