Abstract

    Surveillance has rapidly gained increasing popularity as a modern technology, which can be used to ensure life safety and break the wall of security mistrust. Anomaly detection is of great significance for intelligent surveillance systems. An abnormal or anomalous event refers to an activity that raises suspicions by differing from the majority of the activities. The challenges of anomaly detection relate to complex and crowded scenes, the anomaly localization, small anomaly datasets, and many false-positive detection results. The existing solutions for effective anomaly detection are often limited to specific scenarios. There is a room for an improvement regarding the detection and localization problems and time complexity. This special session will bring together experts from academia and industry across different countries to discuss and showcase the latest advancements in anomaly detection. We encourage submissions on topics such as object detection and localization and real-world anomaly detection applications. The core topics of this special section include but are not limited to: image and video anomaly detection, anomalous detection in digital signal and network, abnormal event localization, pattern recognition and data analysis, human-computer interaction.