Abnormal detection using interaction energy potentials

• Published in: CVPR 2011 pp. 3161 - 3167
• Main Authors: Xinyi Cui, Qingshan Liu, Mingchen Gao, Metaxas, D. N.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2011
• Subjects: Color; Computational modeling; Feature extraction; Humans; Solid modeling; Support vector machines; Tracking
• ISBN: 1457703947; 9781457703942
• ISSN: 1063-6919; 1063-6919
• DOI: 10.1109/CVPR.2011.5995558

A new method is proposed to detect abnormal behaviors in human group activities. This approach effectively models group activities based on social behavior analysis. Different from previous work that uses independent local features, our method explores the relationships between the current behavior state of a subject and its actions. An interaction energy potential function is proposed to represent the current behavior state of a subject, and velocity is used as its actions. Our method does not depend on human detection or segmentation, so it is robust to detection errors. Instead, tracked spatio-temporal interest points are able to provide a good estimation of modeling group interaction. SVM is used to find abnormal events. We evaluate our algorithm in two datasets: UMN and BEHAVE. Experimental results show its promising performance against the state-of-art methods.