Deep learning-based Anomaly Detection on Surveillance Videos: Recent Advances
Akmalazakia Fatan Derana Marsiano, Indah Soesanti, Igi Ardiyanto · 2019
Surveillance cameras number keep increasing from year to year, in 2017 alone estimated 106 million new CCTV installed. Surveillance cameras are installed to widen surveillance coverage in order to detect anomalies. Conventionally, detecting anomalies through surveillance cameras are using human observers. The increasing number of surveillance cameras installed is increasing the need for an automatic system to replace unreliable human observer. Thus an intelligent computer vision-based system to detect anomalies need to be developed. The goal of this system is to provide a warning to the first responder accurately and as fast as possible while the system is running all the time. The faster the response of the first responder, the better chance of an anomaly to be resolved. Deep learning based method has been proposed to solve anomaly detection on videos. Anomaly detected is composed of a set of violent and non-violent crimes with high impact in society. To increase the accuracy of previous system to detect anomalies in condition that resembles real life situation, multiple processes need to be combined. This paper will review various method used to increase the performance of a deep learning based action recognition on videos. To capture temporal information effectively the model need to be able to use multimodalities to detect motion, incorporating long-range temporal structure, and numerous deep learning architectures with various characteristics.