Discrete neural representations for explainable anomaly detection

TL;DR

This paper introduces a novel approach for explainable video anomaly detection that does not rely on object or action classifiers, using saliency maps and discrete neural representations to improve explanation quality and outperform existing methods.

Contribution

It presents a classifier-independent method using saliency maps and a new neural architecture for discrete video representations, enhancing anomaly explanation accuracy.

Findings

Achieved 60% improvement over state-of-the-art on X-MAN dataset

Developed a neural architecture for learning discrete video representations

Demonstrated robustness in detecting and explaining anomalies without classifiers

Abstract

The aim of this work is to detect and automatically generate high-level explanations of anomalous events in video. Understanding the cause of an anomalous event is crucial as the required response is dependant on its nature and severity. Recent works typically use object or action classifier to detect and provide labels for anomalous events. However, this constrains detection systems to a finite set of known classes and prevents generalisation to unknown objects or behaviours. Here we show how to robustly detect anomalies without the use of object or action classifiers yet still recover the high level reason behind the event. We make the following contributions: (1) a method using saliency maps to decouple the explanation of anomalous events from object and action classifiers, (2) show how to improve the quality of saliency maps using a novel neural architecture for learning discrete representations of video by predicting future frames and (3) beat the state-of-the-art anomaly explanation methods by 60\% on a subset of the public benchmark X-MAN dataset.