CPNet: Cross-Parallel Network for Efficient Anomaly Detection

TL;DR

This paper introduces CPNet, a parallel U-Net based architecture that reduces computational costs in video anomaly detection without sacrificing detection accuracy, by processing frames in parallel and capturing temporal relations.

Contribution

The paper proposes a novel Cross-Parallel Network with multiple parallel U-Nets and an inter-network shift module for efficient and accurate anomaly detection in videos.

Findings

Requires less computational cost than baseline U-Net

Maintains equivalent anomaly detection performance

Effectively captures temporal relationships among frames

Abstract

Anomaly detection in video streams is a challenging problem because of the scarcity of abnormal events and the difficulty of accurately annotating them. To alleviate these issues, unsupervised learning-based prediction methods have been previously applied. These approaches train the model with only normal events and predict a future frame from a sequence of preceding frames by use of encoder-decoder architectures so that they result in small prediction errors on normal events but large errors on abnormal events. The architecture, however, comes with the computational burden as some anomaly detection tasks require low computational cost without sacrificing performance. In this paper, Cross-Parallel Network (CPNet) for efficient anomaly detection is proposed here to minimize computations without performance drops. It consists of N smaller parallel U-Net, each of which is designed to handle a single input frame, to make the calculations significantly more efficient. Additionally, an inter-network shift module is incorporated to capture temporal relationships among sequential frames to enable more accurate future predictions.The quantitative results show that our model requires less computational cost than the baseline U-Net while delivering equivalent performance in anomaly detection.