Robust and efficient anomaly detection using heterogeneous representations

Abstract

Various approaches have been proposed for video anomaly detection. Yet these approaches typically suffer from one or more limitations: they often characterize the pattern using its internal information, but ignore its external relationship which is important for local anomaly detection. Moreover, the high-dimensionality and the lack of robustness of pattern representation may lead to problems, including overfitting, increased computational cost and memory requirements, and high false alarm rate. We propose a video anomaly detection framework which relies on a heterogeneous representation to account for both the pattern's internal information and external relationship. The internal information is characterized by slow features learned by slow feature analysis from low-level representations, and the external relationship is characterized by the spatial contextual distances. The heterogeneous representation is compact, robust, efficient, and discriminative for anomaly detection. Moreover, both the pattern's internal information and external relationship can be taken into account in the proposed framework. Extensive experiments demonstrate the robustness and efficiency of our approach by comparison with the state-of-the-art approaches on the widely used benchmark datasets.

Abnormal event detection in crowded scenes using histogram of oriented contextual gradient descriptor

“…Then, the volume is further divided into a set of cuboids with a size of 4 × 4 × 5. We extract the slow features proposed in our previous works [34] from each cuboid as the regional feature, which is robust and discriminative. We construct a 2D HOCG descriptor for each event, i.e., the spatial direction range is quantized into 8 directions with each direction being 45°.…”

“…We utilize four types of Fig . 7 Examples of detection results for global abnormal event detection from the UNM dataset regional features, i.e., the gray value, 3D gradient, GCM [34], and slow features (SF) [43] descriptors. The comparisons demonstrate that after constructing the HORG descriptors, not only the performance of AED is improved but also the dimensionality is also reduced.…”

“…On the other hand, to reduce the influence of noise, we select the four nearest neighbors in the feature space from the eight neighbors. The contextual gradient can be computed based on different types of features, such as the gray values of raw pixels, HOG, HOF, and gradient central moments (GCM) [34]. We adopted the commonly used Euclidean distance as the distance measure between two features, defined as…”

Learning Spatio-Temporal Features for Detecting Anomalies in Videos using Convolutional Autoencoder

“…images or videos) is important to enable a good classification performance for different types of anomalies. In an attempt to classify pattern representation approaches, Hu et al [15] suggested that the methods for representing patterns can be grouped into four main categories: trajectory-based, spatio-temporal-interest-point-based, foreground-blob-based and volume-based. However, methods that can be grouped into one or more categories are frequently observed in the literature and, usually, they have in common the use of hand-crafted features.…”