Holistic Features For Real-Time Crowd Behaviour Anomaly Detection
This work addresses real-time anomaly detection in crowds, which is important for security and surveillance applications, but it is incremental as it builds on existing features and methods.
The paper tackles crowd behavior anomaly detection by proposing a low-dimensional descriptor with four holistic features, including two new ones, and evaluates it using Gaussian Mixture Model and Support Vector Machine approaches. It achieves state-of-the-art classification on the violent-flows dataset and processes at 40 frames per second, enabling real-time performance.
This paper presents a new approach to crowd behaviour anomaly detection that uses a set of efficiently computed, easily interpretable, scene-level holistic features. This low-dimensional descriptor combines two features from the literature: crowd collectiveness [1] and crowd conflict [2], with two newly developed crowd features: mean motion speed and a new formulation of crowd density. Two different anomaly detection approaches are investigated using these features. When only normal training data is available we use a Gaussian Mixture Model (GMM) for outlier detection. When both normal and abnormal training data is available we use a Support Vector Machine (SVM) for binary classification. We evaluate on two crowd behaviour anomaly detection datasets, achieving both state-of-the-art classification performance on the violent-flows dataset [3] as well as better than real-time processing performance (40 frames per second).