A MIL Approach for Anomaly Detection in Surveillance Videos from Multiple Camera Views

TL;DR

This paper introduces a novel multi-camera MIL approach for anomaly detection in surveillance videos, effectively addressing occlusion, clutter, and data scarcity issues, resulting in improved detection performance.

Contribution

The paper proposes the MC-MIL algorithm that combines multiple camera views with MIL to enhance anomaly detection in surveillance videos, a novel integration for this task.

Findings

Significant F1 score improvement over single-camera methods

Effective handling of occlusion and clutter through multiple camera views

Re-labeled PETS-2009 dataset for multi-camera anomaly detection

Abstract

Occlusion and clutter are two scene states that make it difficult to detect anomalies in surveillance video. Furthermore, anomaly events are rare and, as a consequence, class imbalance and lack of labeled anomaly data are also key features of this task. Therefore, weakly supervised methods are heavily researched for this application. In this paper, we tackle these typical problems of anomaly detection in surveillance video by combining Multiple Instance Learning (MIL) to deal with the lack of labels and Multiple Camera Views (MC) to reduce occlusion and clutter effects. In the resulting MC-MIL algorithm we apply a multiple camera combined loss function to train a regression network with Sultani's MIL ranking function. To evaluate the MC-MIL algorithm first proposed here, the multiple camera PETS-2009 benchmark dataset was re-labeled for the anomaly detection task from multiple camera views. The result shows a significant performance improvement in F1 score compared to the single-camera configuration.