Combining Gaussian Mixture models and Support Vector Machines for relevance feedback in content based image retrieval

Marakakis, A; Galatsanos, N; Likas, A; Stafylopatis, A

dc.contributor.author	Marakakis, A	en
dc.contributor.author	Galatsanos, N	en
dc.contributor.author	Likas, A	en
dc.contributor.author	Stafylopatis, A	en
dc.date.accessioned	2014-03-01T01:29:59Z
dc.date.available	2014-03-01T01:29:59Z
dc.date.issued	2009	en
dc.identifier.issn	15715736	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/19439
dc.subject	Content Based Image Retrieval	en
dc.subject	Distance Measure	en
dc.subject	Gaussian Mixture	en
dc.subject	Gaussian Mixture Model	en
dc.subject	Image Features	en
dc.subject	Image Modeling	en
dc.subject	Kernel Function	en
dc.subject	Model Specification	en
dc.subject	Numerical Experiment	en
dc.subject	Probability Density Function	en
dc.subject	Relevance Feedback	en
dc.subject	Support Vector Machine	en
dc.title	Combining Gaussian Mixture models and Support Vector Machines for relevance feedback in content based image retrieval	en
heal.type	journalArticle	en
heal.identifier.primary	10.1007/978-1-4419-0221-4_30	en
heal.identifier.secondary	http://dx.doi.org/10.1007/978-1-4419-0221-4_30	en
heal.publicationDate	2009	en
heal.abstract	A relevance feedback (RF) approach for content based image retrieval (CBIR) is proposed, which combines Support Vector Machines (SVMs) with Gaussian Mixture (GM) models. Specifically, it constructs GM models of the image features distribution to describe the image content and trains an SVM classifier to distinguish between the relevant and irrelevant images according to the preferences of the user. The method is based on distance measures between probability density functions (pdfs), which can be computed in closed form for GM models. In particular, these distance measures are used to define a new SVM kernel function expressing the similarity between the corresponding images modeled as GMs. Using this kernel function and the user provided feedback examples, an SVM classifier is trained in each RF round, resulting in an updated ranking of the database images. Numerical experiments are presented that demonstrate the merits of the proposed relevance feedback methodology and the advantages of using GMs for image modeling in the RF framework. © 2009 International Federation for Information Processing.	en
heal.journalName	IFIP International Federation for Information Processing	en
dc.identifier.doi	10.1007/978-1-4419-0221-4_30	en
dc.identifier.volume	296	en
dc.identifier.spage	249	en
dc.identifier.epage	258	en