AED-Net: An Abnormal Event Detection Network

TL;DR

AED-Net is a self-supervised abnormal event detection framework using PCAnet and kPCA, effectively identifying anomalies in crowded scenes without requiring labeled abnormal data, and improves with local response normalization.

Contribution

The paper introduces AED-Net, a novel self-supervised framework combining PCAnet and kPCA for anomaly detection in surveillance videos, with an enhancement using LRN layers.

Findings

Achieves higher EER and AUC on UMN and UCSD datasets.

Self-supervised approach eliminates need for abnormal training data.

Adding LRN layer improves generalization and detection performance.

Abstract

It is challenging to detect the anomaly in crowded scenes for quite a long time. In this paper, a self-supervised framework, abnormal event detection network (AED-Net), which is composed of PCAnet and kernel principal component analysis (kPCA), is proposed to address this problem. Using surveillance video sequences of different scenes as raw data, PCAnet is trained to extract high-level semantics of crowd's situation. Next, kPCA,a one-class classifier, is trained to determine anomaly of the scene. In contrast to some prevailing deep learning methods,the framework is completely self-supervised because it utilizes only video sequences in a normal situation. Experiments of global and local abnormal event detection are carried out on UMN and UCSD datasets, and competitive results with higher EER and AUC compared to other state-of-the-art methods are observed. Furthermore, by adding local response normalization (LRN) layer, we propose an improvement to original AED-Net. And it is proved to perform better by promoting the framework's generalization capacity according to the experiments.