TRECVID is the leading global benchmark on video retrieval, organized by NIST, USA. I have been involved as a scientific co-coordinator (with Alan Smeaton) from 2003. TRECVID has been a very influential activity and resource for research in multimedia information retrieval, in particular for video. Through the years, TRECVID has programmed both core and supportive tasks, such as (video) ad hoc search, semantic indexing, summarization, copy detection, instance search and multimedia event detection. Find more information on TRECVID at the project website.
Some key publications of my work for TRECVID:
- Alan Smeaton, Paul Over, and Wessel Kraaij. Evaluation Campaigns and TRECVid. In Proceedings of the 8th ACM SIGMM International workshop on Multimedia Information Retrieval, 2006.
- A.F. Smeaton, P. Over, and W. Kraaij. High Level Feature Detection from Video in TRECVid: a 5-Year Retrospective of Achievements. In Ajay Divakaran, editor, Multimedia Content Analysis, Theory and Applications. Springer, 2008.
- J. Arlandis, P. Over, and W. Kraaij. Boundary error analysis and categorization in the TRECVID news story segmentation task. In W.-K. Leow et al.(Eds.), editor, CIVR 2005, Lecture Notes in Computer Science 3568, pages 103-112. Springer-Verlag, July 2005.
- George Awad, Paul Over, and Wessel Kraaij. Content-Based Video Copy Detection Benchmarking at TRECVID. ACM Transactions on Information Systems, 32(3):14:1-14:40, 2014.