Show simple item record

dc.contributor.authorGinsca, A.L.en_US
dc.contributor.authorPopescu, A.en_US
dc.contributor.authorIonescu, B.en_US
dc.contributor.authorArmagan, A.en_US
dc.contributor.authorKanellos I.en_US
dc.date.accessioned2016-02-08T12:00:37Z
dc.date.available2016-02-08T12:00:37Z
dc.date.issued2014en_US
dc.identifier.urihttp://hdl.handle.net/11693/27741
dc.description.abstractExisting image retrieval systems exploit textual or/and visual information to return results. Retrieval is mostly focused on data themselves and disregards the data sources. In Web 2.0 platforms, the quality of annotations provided by different users can vary strongly. To account for this variability, we complement existing methods by introducing user tagging credibility in the retrieval process. Tagging credibility is automatically estimated by leveraging a large set of visual concept classifiers learned with Overfeat, a convolutional neural network (CNN) feature. A good image retrieval system should return results that are both relevant and diversified and here we tackle both challenges. Classically, we diversify results by using a k-Means algorithm and increase relevance by favoring images uploaded by users with good credibility estimates. Evaluation is performed on DIV400, a publicly available social image retrieval dataset and shows that our method is competitive with existing approaches.en_US
dc.language.isoEnglishen_US
dc.source.titleMM 2014 - Proceedings of the 2014 ACM Conference on Multimediaen_US
dc.relation.isversionofhttp://dx.doi.org/10.1145/2647868.2655033en_US
dc.subjectImage retrievalen_US
dc.subjectNeural networksen_US
dc.subjectUser interfacesen_US
dc.subjectConvolutional neural networken_US
dc.subjectData-sourcesen_US
dc.subjectImage retrieval systemsen_US
dc.subjectk-Means algorithmen_US
dc.subjectRetrieval processen_US
dc.subjectSocial image retrievalsen_US
dc.subjectVisual concepten_US
dc.subjectVisual informationen_US
dc.subjectSearch enginesen_US
dc.titleToward an estimation of user tagging credibility for social image retrievalen_US
dc.typeConference Paperen_US
dc.departmentBilkent Universityen_US
dc.citation.spage1021en_US
dc.citation.epage1024en_US
dc.identifier.doi10.1145/2647868.2655033en_US
dc.publisherAssociation for Computing Machinery, Incen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record