Recent review of crowd anomaly detection

With the population and crowd activities increased, large-scale crowd events make the probability of hazard incidents larger than ever. In the past decades, automated crowd scene analysis has been of great interest in computer vision. As a key aspect of crowd scene analysis, a number of work on video-based crowd anomaly detection in dense scenes have been proposed. This article presents a survey on crowd analysis in dense crowd scenes using computer vision techniques, covering two aspects: crowd density estimation and abnormal event detection. Some existing problems and perspectives are also discussed at the end.