Videos represent the primary source of information for surveillanceapplications and are available in large amounts but in most cases containlittle or no annotation for supervised learning. This article reviews thestate-of-the-art deep learning based methods for video anomaly detection andcategorizes them based on the type of model and criteria of detection. We alsoperform simple studies to understand the different approaches and provide thecriteria of evaluation for spatio-temporal anomaly detection.
展开▼
