Bounding Boxes and Probabilistic Graphical Models: Video Anomaly Detection Simplified

TL;DR

This paper presents a probabilistic approach to video anomaly detection using only object bounding boxes, enabling faster, more privacy-preserving models suitable for edge devices, with competitive and superior results on benchmark datasets.

Contribution

The study introduces a bounding box-based probabilistic model for video anomaly detection that reduces feature space and computational requirements while maintaining high performance.

Findings

Achieves training in less than 7 seconds on a standard CPU.

Performs competitively on CUHK Avenue and ShanghaiTech datasets.

Outperforms state-of-the-art on the challenging StreetScene dataset.

Abstract

In this study, we formulate the task of Video Anomaly Detection as a probabilistic analysis of object bounding boxes. We hypothesize that the representation of objects via their bounding boxes only, can be sufficient to successfully identify anomalous events in a scene. The implied value of this approach is increased object anonymization, faster model training and fewer computational resources. This can particularly benefit applications within video surveillance running on edge devices such as cameras. We design our model based on human reasoning which lends itself to explaining model output in human-understandable terms. Meanwhile, the slowest model trains within less than 7 seconds on a 11th Generation Intel Core i9 Processor. While our approach constitutes a drastic reduction of problem feature space in comparison with prior art, we show that this does not result in a reduction in performance: the results we report are highly competitive on the benchmark datasets CUHK Avenue and ShanghaiTech, and significantly exceed on the latest State-of-the-Art results on StreetScene, which has so far proven to be the most challenging VAD dataset.