Abnormal Event Detection in Videos using Generative Adversarial Nets
This addresses the problem of automated surveillance for public safety, though it is incremental as it applies an existing GAN framework to a specific domain.
The paper tackles abnormal event detection in crowded videos by training Generative Adversarial Nets (GANs) on normal frames and optical-flow images to learn scene normality, achieving state-of-the-art results in frame-level and pixel-level detection tasks.
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.