Video Anomaly Detection with Compact Feature Sets for Online Performance

Abstract

Over the past decade, video anomaly detection has been explored with remarkable results. However, research on methodologies suitable for online performance is still very limited. In this paper, we present an online framework for video anomaly detection. The key aspect of our framework is a compact set of highly descriptive features, which is extracted from a novel cell structure that helps to define support regions in a coarse-to-fine fashion. Based on the scene’s activity, only a limited number of support regions are processed, thus limiting the size of the feature set. Specifically, we use foreground occupancy and optical flow features. The framework uses an inference mechanism that evaluates the compact feature set via Gaussian Mixture Models, Markov Chains and Bag-of-Words in order to detect abnormal events. Our framework also considers the joint response of the models in the local spatio-temporal neighborhood to increase detection accuracy. We test our framework on popular existing datasets and on a new dataset comprising a wide variety of realistic videos captured by surveillance cameras. This particular dataset includes surveillance videos depicting criminal activities, car accidents and other dangerous situations.

Existing System

Spatiotemporal Compositions, Gaussian Process Regression.

Proposed System

In this paper, we proposed an online framework for video anomaly detection. Our framework extracts a compact set of features based on foreground occupancy and optical flow information. The framework employs a novel variable-sized cell structure which allows extracting features from a limited number of different support regions in a fine-to-coarse fashion. The extracted compact set of features comprises features from foreground occupancy and optical flow. Features from foreground occupancy help to efficiently capture events associated with weak motion, such as loitering or the abnormal presence of subjects; while features from optical flow are useful to detect events associated with sudden motion, such as panic or fights. Multiple inference models are employed to accurately describe the activity of challenging scenes, where anomalous events can be due to sudden motion, weak motion, or both. This is particularly useful to attain a good performance on scenes depicting realistic events.

Architecture

BLOCK DIAGRAM