Deep-Learning Algorithms for Video Anomaly Detection: A Mini-Review

With the increasing demand for security in the day-to-day lives of people, especially when in public spaces, surveillance system helps to monitor human behavior, prevent illegal activities and detect anomalous events. Anomaly detection in video involves the process of detecting anomalous behaviors that occur in such video. Due to the inefficiency of humans when it comes to surveillance, various algorithms can help to automatically detect these anomalous events in videos. Deep learning is fast rising, which is very useful as it imitates how humans gain knowledge and this helps in the area of surveillance and detection. This study aims to provide a comprehensive review of deep learning algorithms used for the detection of anomalous events captured in videos. This study also captures the datasets available for anomaly detection, the applications and future research directions.