Show simple item record

Files in this item

Thumbnail

Item metadata

dc.contributor.authorArandelovic, Ognjen
dc.date.accessioned2016-09-16T15:30:11Z
dc.date.available2016-09-16T15:30:11Z
dc.date.issued2016-06-26
dc.identifier245211630
dc.identifierb4ac1e0e-ba61-4cfb-822e-d6b9536b6a76
dc.identifier84986286603
dc.identifier000400012304102
dc.identifier.citationArandelovic , O 2016 , Learnt quasi-transitive similarity for retrieval from large collections of faces . in Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . , 7780897 , IEEE Computer Society Conference on Computer Vision and Pattern Recognition , IEEE Computer Society , pp. 4883-4892 , IEEE Conference on Computer Vision and Pattern Recognition , Las Vegas , United States , 26/06/16 . https://doi.org/10.1109/CVPR.2016.528en
dc.identifier.citationconferenceen
dc.identifier.isbn9781467388511
dc.identifier.issn1063-6919
dc.identifier.urihttps://hdl.handle.net/10023/9516
dc.description.abstractWe are interested in identity-based retrieval of face sets from large unlabelled collections acquired in uncontrolled environments. Given a baseline algorithm for measuring the similarity of two face sets, the meta-algorithm introduced in this paper seeks to leverage the structure of the data corpus to make the best use of the available baseline. In particular, we show how partial transitivity of inter-personal similarity can be exploited to improve the retrieval of particularly challenging sets which poorly match the query under the baseline measure. We: (i) describe the use of proxy sets as a means of computing the similarity between two sets, (ii) introduce transitivity meta-features based on the similarity of salient modes of appearance variation between sets, (iii) show how quasi-transitivity can be learnt from such features without any labelling or manual intervention, and (iv) demonstrate the effectiveness of the proposed methodology through experiments on the notoriously challenging YouTube database.
dc.format.extent10
dc.format.extent1666888
dc.language.isoeng
dc.publisherIEEE Computer Society
dc.relation.ispartofProceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)en
dc.relation.ispartofseriesIEEE Computer Society Conference on Computer Vision and Pattern Recognitionen
dc.subjectQA75 Electronic computers. Computer scienceen
dc.subjectNDASen
dc.subjectBDCen
dc.subjectR2Cen
dc.subject~DC~en
dc.subject.lccQA75en
dc.titleLearnt quasi-transitive similarity for retrieval from large collections of facesen
dc.typeConference itemen
dc.contributor.institutionUniversity of St Andrews. School of Computer Scienceen
dc.identifier.doi10.1109/CVPR.2016.528
dc.identifier.urlhttp://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Arandjelovic_Learnt_Quasi-Transitive_Similarity_CVPR_2016_paper.pdfen


This item appears in the following Collection(s)

Show simple item record