Anomaly detection in crowded scenes

Large-scale surveillance system: detection and tracking of suspicious motion patterns in crowded traffic scenes

“…In [8], a framework is proposed to identify multiple crowd behaviors through stability analysis for dynamical systems, without the need for object tracking. Abnormal behavior detection is also studied in [9] using spatiotemporal cuboids, in [10] by modelling appearance and dynamics based on dynamic textures, and by learning normal behavior in [11].…”

Abnormal event detection in crowded scenes using sparse representation

“…We evaluate our method in three different abnormal event detection datasets, including the UMN dataset [31], the UCSD dataset [23], and the subway dataset [1]. The main contributions are as below:…”

“…In [30], a graph-based nonlinear dimensionality reduction method using motion cues is applied for abnormality detection. Mahadevan et al [23] model the normal crowd behavior by mixtures of dynamic textures. Mehran et al [27] present a new way to formulate the abnormal crowd behavior by adopting the social force model [9,35].…”

“…According to the definition above, the abnormal events can be identified as irregular events from normal ones. Depending on the scale of interests, previous work in abnormal video event detection, such as [39,3,1,23,27,4,7], can be categorized into two classes, as shown in Fig. 1 (each ellipse stands for a moving pedestrian):…”

“…Some other methods consider the spatial-temporal information, such as Histogram of Optical Flow (HOF) [1], spatial-temporal gradient [17], social force model [27], chaotic invariant [34], mixtures of dynamic textures [23]. There are also saliency feature [14] and graphbased nonlinear dimensionality reduction method [30].…”