Plug-and-Play Anomaly Detection with Expectation Maximization Filtering

TL;DR

This paper introduces a novel unsupervised neural network for crowd anomaly detection that operates efficiently with minimal training and integrates an Expectation Maximization filter for sample selection, enhancing detection accuracy.

Contribution

The paper presents a plug-and-play anomaly detection framework combining a core neural network with an EM filter, enabling effective unsupervised learning in constrained surveillance scenarios.

Findings

CAD improves AUC by 4.66% and 4.9% over existing methods.

Single-epoch training yields significant performance gains.

Overall framework surpasses future frame prediction methods by 24.87% AUC.

Abstract

Anomaly detection in crowds enables early rescue response. A plug-and-play smart camera for crowd surveillance has numerous constraints different from typical anomaly detection: the training data cannot be used iteratively; there are no training labels; and training and classification needs to be performed simultaneously. We tackle all these constraints with our approach in this paper. We propose a Core Anomaly-Detection (CAD) neural network which learns the motion behavior of objects in the scene with an unsupervised method. On average over standard datasets, CAD with a single epoch of training shows a percentage increase in Area Under the Curve (AUC) of 4.66% and 4.9% compared to the best results with convolutional autoencoders and convolutional LSTM-based methods, respectively. With a single epoch of training, our method improves the AUC by 8.03% compared to the convolutional LSTM-based approach. We also propose an Expectation Maximization filter which chooses samples for training the core anomaly-detection network. The overall framework improves the AUC compared to future frame prediction-based approach by 24.87% when crowd anomaly detection is performed on a video stream. We believe our work is the first step towards using deep learning methods with autonomous plug-and-play smart cameras for crowd anomaly detection.