A Review of Object Classification for Video Surveillance Systems

摘要

Object classification in video surveillance systems is a major research field for robust event detection. It is an important step for improving the reliability of various smart applications in visual sensor networks. This paper presents a comprehensive review of the state-of-the-art methodologies for object classification. It proposes various evaluation techniques regarding robust performance estimations in video surveillance systems. It illustrates a large variation of dynamics contexts and their effects in visual scenes.