Detection of Fights in Videos: A Comparison Study of Anomaly Detection and Action Recognition

TL;DR

This study compares anomaly detection and action recognition methods for fight detection in videos, showing that anomaly detection can outperform action recognition and can be used to generate training data iteratively.

Contribution

The paper demonstrates the effectiveness of anomaly detection in fight detection and introduces an iterative approach to generate training data for action recognition.

Findings

Anomaly detection performs as well or better than action recognition.

Iterative use of anomaly detection improves training data quality.

Achieved state-of-the-art results on three datasets.

Abstract

Detection of fights is an important surveillance application in videos. Most existing methods use supervised binary action recognition. Since frame-level annotations are very hard to get for anomaly detection, weakly supervised learning using multiple instance learning is widely used. This paper explores the detection of fights in videos as one special type of anomaly detection and as binary action recognition. We use the UBI-Fight and NTU-CCTV-Fight datasets for most of the study since they have frame-level annotations. We find that the anomaly detection has similar or even better performance than the action recognition. Furthermore, we study to use anomaly detection as a toolbox to generate training datasets for action recognition in an iterative way conditioned on the performance of the anomaly detection. Experiment results should show that we achieve state-of-the-art performance on three fight detection datasets.