Dimension-specific search for multimedia retrieval.

Huang, Zi; Shen, Heng Tao; Song, Dawei; Li, Xue; R�eger, Stefan

Abstract

Observing that current Global Similarity Measures (GSM) which average the effect of few significant differences on all dimensions may cause possible performance limitation, we propose the first Dimension-specific Similarity Measure (DSM) to take local dimensionspecific constraints into consideration. The rationale for DSM is that significant differences on some individual dimensions may lead to different semantics. An efficient search algorithm is proposed to achieve fast Dimension-specific KNN (DKNN) retrieval. Experiment results show that our methods outperform traditional methods by large gaps.

Citation

HUANG, Z., SHEN, H.T., SONG, D., LI, X. and RUEGER, S. 2009. Dimension-specific search for multimedia retrieval. In Zhou, X., Yokota, H., Deng, K. and Liu, Q. (eds.) Proceedings of the 14th International conference on database systems for advanced applications (DASFAA 2009), 21-23 April 2009, Brisbane, Australia. Lecture notes in computer science, 5463. Berlin: Springer [online], pages 693-698. Available from: https://doi.org/10.1007/978-3-642-00887-0_60

• Conference Name: 14th International conference on database systems for advanced applications (DASFAA 2009)
• Conference Location: Brisbane, Australia
• Start Date: Apr 21, 2009
• End Date: Apr 23, 2009
• Acceptance Date: Mar 31, 2009
• Online Publication Date: Mar 31, 2009
• Publication Date: Dec 31, 2009
• Deposit Date: Aug 14, 2009
• Publicly Available Date: Aug 14, 2009
• Publisher: Springer
• Pages: 693-698
• Series Title: Lecture notes in computer science
• Series Number: 5463
• ISBN: 9783642008863
• DOI: https://doi.org/10.1007/978-3-642-00887-0_60
• Keywords: Global similarity measures (GSM); Performance limitation; Dimension-specific similarity measure (DSM) ; Dimension-specific constraints; Dimension-specific KNN (DKNN) retrieval
• Public URL: http://hdl.handle.net/10059/397