Dissimilarity measures for content-based image retrieval.

Hu, Rui; R�ger, Stefan; Song, Dawei; Liu, Haiming; Huang, Zi

Abstract

Dissimilarity measurement plays a crucial role in content-based image retrieval. In this paper, 16 core dissimilarity measures are introduced and evaluated. We carry out a systematic performance comparison on three image collections, Corel, Getty and Trecvid2003, with 7 different feature spaces. Two search scenarios are considered: single image queries based on the Vector Space Model, and multi-image queries based on k-Nearest Neighbours search. A number of observations are drawn, which will lay a foundation for developing more effective image search technologies.

Citation

HU, R., RUGER, S., SONG, D., LIU, H. and HUANG, Z. 2008. Dissimilarity measures for content-based image retrieval. In Proceedings of the 2008 IEEE international conference on multimedia and expo (ICME 2008), 23-26 June 2008, Hannover, Germany. New York: IEEE [online], article number 4607697, pages 1365-1368. Available from: https://doi.org/10.1109/ICME.2008.4607697

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 2008 IEEE international conference on multimedia and expo (ICME 2008)
• Start Date: Jun 23, 2008
• End Date: Jun 26, 2008
• Acceptance Date: Jun 23, 2008
• Online Publication Date: Aug 26, 2008
• Publication Date: Dec 31, 2008
• Deposit Date: Jun 3, 2009
• Publicly Available Date: Jun 3, 2009
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Article Number: 4607697
• Pages: 1365-1368
• Series Title: Proceedings of the IEEE international conference on multimedia and expo
• ISBN: 9781424425709
• DOI: https://doi.org/10.1109/ICME.2008.4607697
• Keywords: Dissimilarity measure; Feature space; Content based image retrieval
• Public URL: http://hdl.handle.net/10059/354
• Contract Date: Jun 3, 2009