Detecting Contextual Anomalies by Discovering Consistent Spatial Regions

TL;DR

This paper introduces a simple yet effective method for detecting spatial context anomalies in videos by clustering object attributes, achieving state-of-the-art results and providing explainable normalcy maps without pre-trained models.

Contribution

The paper presents a novel clustering-based approach for spatial context modeling in video anomaly detection that requires fewer parameters and offers interpretability.

Findings

Achieves state-of-the-art performance on Street Scene dataset

Provides explainable normalcy maps without pre-trained segmentation models

Uses Gaussian mixture models for clustering object attributes

Abstract

We describe a method for modeling spatial context to enable video anomaly detection. The main idea is to discover regions that share similar object-level activities by clustering joint object attributes using Gaussian mixture models. We demonstrate that this straightforward approach, using orders of magnitude fewer parameters than competing models, achieves state-of-the-art performance in the challenging spatial-context-dependent Street Scene dataset. As a side benefit, the high-resolution discovered regions learned by the model also provide explainable normalcy maps for human operators without the need for any pre-trained segmentation model.