Content-based video copy detection - A survey

Abstract

In the presence of overwhelming amount of digital video data, the need for automated procedures to protect owners against unauthorized use of their content. to enable content publishers and distributors to monitor broadcast usage of videos and to help businesses manage storage of videos in large databases has become ever more critical Till now, two technical approaches have been proposed toward these goals, namely, watermarking-based systems and content-based copy detection (CBCD). Among them, the former one has been more thoroughly studied and explored, while the latter one is still in its early stages. This chapter reviews existing video copy detection systems, investigates some typical CBCD algorithms, compares the algorithms through appropriate performance metrics, and lists some challenges and open issues in this research field. © 2010 Springer-Verlag Berlin Heidelberg.