Abnormal Event Detection in Videos using Generative Adversarial Nets

TL;DR

This paper introduces a novel abnormal event detection method in crowded scenes using GANs trained solely on normal data, which effectively identifies anomalies by comparing real data with generated normal representations.

Contribution

The paper presents a GAN-based approach trained only on normal data for abnormal event detection in videos, improving detection accuracy over existing methods.

Findings

Outperforms state-of-the-art in frame-level abnormality detection

Effective at pixel-level anomaly localization

Demonstrates robustness on challenging datasets

Abstract

In this paper we address the abnormality detection problem in crowded scenes. We propose to use Generative Adversarial Nets (GANs), which are trained using normal frames and corresponding optical-flow images in order to learn an internal representation of the scene normality. Since our GANs are trained with only normal data, they are not able to generate abnormal events. At testing time the real data are compared with both the appearance and the motion representations reconstructed by our GANs and abnormal areas are detected by computing local differences. Experimental results on challenging abnormality detection datasets show the superiority of the proposed method compared to the state of the art in both frame-level and pixel-level abnormality detection tasks.