A Light-weight Relevance Feedback Solution for Large Scale Content-Based Video Retrieval
This paper addresses the problem of large scale content-based video retrieval with relevance feedback. We analyze the common methods which leverage local feature detectors to extract feature descriptors from video collections and perform multi-level matching after indexing and retrieval of feature vectors. Instead of learning similarity-preserving codes, we introduce the relevance feedback approach in a light-weight way. A relevance model is proposed to merge semantic similarity with the original distance matching at descriptor level. By learning several weights using canonical correlation analysis (CCA), the resulting candidate list of similar videos changes according to relevance feedback. Finally, we demonstrate the improvement of the proposed method by experiments on a standard real world dataset.
Keywords: Content-based Video Retrieval, Relevance Feedback, CCA
Download Full-Text
ABOUT THE AUTHORS
Zimian Li
Zimian Li received the B.S. degree from the University of Science and Technology of China (USTC), Hefei, China, in 2006. He is currently pursuing the Ph.D. degree in the School of Information Science and Technology of USTC. He already published three EI-indexed papers in international conferences and one in Chinese domestic journal. His research interests include self-healing multimedia system, multimedia communication.
Ming Zhu
Ming Zhu is currently a professor of University of Science and Technology of China. He received B.S., M.S. and Ph.D. degrees in Computer Science from University of Science and Technology of China in 1986, 1989 and 2001, respectively. He became an assistant professor in 1989 and a professor of USTC in 2004. He worked as a visiting scholar in Department of Computing, The Hong Kong Polytechnic University from 1997 to 1998. He is the Director of the Key Lab of Network Communication System & Control, Chinese Academy of Sciences and the Director of the Key Lab of Network Communication System & Control, Anhui. His research interests include intelligent software systems, data mining and network security.
Zimian Li
Zimian Li received the B.S. degree from the University of Science and Technology of China (USTC), Hefei, China, in 2006. He is currently pursuing the Ph.D. degree in the School of Information Science and Technology of USTC. He already published three EI-indexed papers in international conferences and one in Chinese domestic journal. His research interests include self-healing multimedia system, multimedia communication.
Ming Zhu
Ming Zhu is currently a professor of University of Science and Technology of China. He received B.S., M.S. and Ph.D. degrees in Computer Science from University of Science and Technology of China in 1986, 1989 and 2001, respectively. He became an assistant professor in 1989 and a professor of USTC in 2004. He worked as a visiting scholar in Department of Computing, The Hong Kong Polytechnic University from 1997 to 1998. He is the Director of the Key Lab of Network Communication System & Control, Chinese Academy of Sciences and the Director of the Key Lab of Network Communication System & Control, Anhui. His research interests include intelligent software systems, data mining and network security.
