Show simple item record

dc.contributor.authorSener, Fadime
dc.contributor.authorIkizler-Cinbis, Nazli
dc.date.accessioned2019-12-13T06:51:31Z
dc.date.available2019-12-13T06:51:31Z
dc.date.issued2014
dc.identifier.issn0262-8856
dc.identifier.urihttps://doi.org/10.1016/j.imavis.2014.02.014
dc.identifier.urihttp://hdl.handle.net/11655/18638
dc.description.abstractText-based image retrieval may perform poorly due to the irrelevant and/or incomplete text surrounding the images in the web pages. In such situations, visual content of the images can be leveraged to improve the image ranking performance. In this paper, we look into this problem of image re-ranking and propose a system that automatically constructs multiple candidate "multi-instance bags (MI-bags)", which are likely to contain relevant images. These automatically constructed bags are then utilized by ensembles of Multiple Instance Learning (MIL) classifiers and the images are re-ranked according to the final classification responses. Our method is unsupervised in the sense that, the only input to the system is the text query itself, without any user feedback or annotation. The experimental results demonstrate that constructing multiple instance bags based on the retrieval order and utilizing ensembles of MIL classifiers greatly enhance the retrieval performance, achieving on par or better results compared to the state-of-the-art. (C) 2014 Elsevier B.V. All rights reserved.
dc.language.isoen
dc.publisherElsevier
dc.relation.isversionof10.1016/j.imavis.2014.02.014
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subjectComputer Science
dc.subjectEngineering
dc.subjectOptics
dc.titleEnsemble Of Multiple Instance Classifiers For Image Re-Ranking
dc.typeinfo:eu-repo/semantics/article
dc.typeinfo:eu-repo/semantics/publishedVersion
dc.relation.journalImage And Vision Computing
dc.contributor.departmentBilgisayar Mühendisliği
dc.identifier.volume32
dc.identifier.issue5
dc.identifier.startpage348
dc.identifier.endpage362
dc.description.indexWoS


Files in this item

This item appears in the following Collection(s)

Show simple item record