A content-based image retrieval scheme allowing for robust automatic personalization

Chatzis, S; Doulamis, A; Varvarigou, T

dc.contributor.author	Chatzis, S	en
dc.contributor.author	Doulamis, A	en
dc.contributor.author	Varvarigou, T	en
dc.date.accessioned	2014-03-01T02:44:22Z
dc.date.available	2014-03-01T02:44:22Z
dc.date.issued	2007	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/31785
dc.subject	Mixture models	en
dc.subject	Personalization	en
dc.subject	T distributions	en
dc.subject.other	Feedback	en
dc.subject.other	Robust control	en
dc.subject.other	Semantics	en
dc.subject.other	Sensory perception	en
dc.subject.other	User interfaces	en
dc.subject.other	Content based image retrieval systems	en
dc.subject.other	Human perception	en
dc.subject.other	Mixture models	en
dc.subject.other	Relevance feedback	en
dc.subject.other	Robust automatic personalization	en
dc.subject.other	Content based retrieval	en
dc.title	A content-based image retrieval scheme allowing for robust automatic personalization	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1145/1282280.1282281	en
heal.identifier.secondary	http://dx.doi.org/10.1145/1282280.1282281	en
heal.publicationDate	2007	en
heal.abstract	The retrieval performance of content-based image retrieval (CBIR) systems is often disappointingly low, mainly due to the subjectivity of human perception. Relevance feedback (RF) has been widely considered as a powerful tool to enhance CBIR systems by incorporating human perception subjectivity into the retrieval procedure. However, usually, the obtained feedback logs are scarce and contain lots of outliers, undermining the RF adaptation effectiveness. In this paper, we tackle these shortcomings exploiting the inherent outlier downweighting capabilities mixtures of Student's t distributions offer. Each semantic class is modeled by a mixture of t distributions fitted to data provided by the system operators. Further, the semantic class models get personalized by application of a novel, efficient RF algorithm allowing for the robust adaptation of the semantic class models to the accumulated feedback of each user. The efficacy of our approach is validated through a series of experiments using objective performance criteria. Copyright 2007 ACM.	en
heal.journalName	Proceedings of the 6th ACM International Conference on Image and Video Retrieval, CIVR 2007	en
dc.identifier.doi	10.1145/1282280.1282281	en
dc.identifier.spage	1	en
dc.identifier.epage	8	en