Show simple item record

dc.contributor.authorArbelaez, Pablo
dc.date.accessioned2014-06-16T12:29:32Z
dc.date.available2014-06-16T12:29:32Z
dc.date.issued2006
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/13468
dc.language.isoenen
dc.subjectsegmentation techniquesen
dc.subjectComputer visionen
dc.subject.ddc006.3en
dc.titleBoundary Extraction in Natural Images Using Ultrametric Contour Mapsen
dc.typeCommunication / Conférence
dc.description.abstractenThis paper presents a low-level system for boundary extraction and segmentation of natural images and the evaluation of its performance. We study the problem in the framework of hierarchical classification, where the geometric structure of an image can be represented by an ultrametric contour map, the soft boundary image associated to a family of nested segmentations. We define generic ultrametric distances by integrating local contour cues along the regions boundaries and combining this information with region attributes. Then, we evaluate quantitatively our results with respect to ground-truth segmentation data, proving that our system outperforms significantly two widely used hierarchical segmentation techniques, as well as the state of the art in local edge detection.en
dc.identifier.citationpages182en
dc.relation.ispartoftitleConference on Computer Vision and Pattern Recognition Workshop, 2006. CVPRW '06. Proceedingsen
dc.relation.ispartofpublnameIEEEen
dc.relation.ispartofdate2006
dc.subject.ddclabelIntelligence artificielleen
dc.relation.ispartofisbn0-7695-2646-2en
dc.relation.conftitleConference on Computer Vision and Pattern Recognition Workshop, 2006. CVPRW '06en
dc.relation.confdate2006-06
dc.relation.confcityNew Yorken
dc.relation.confcountryÉtats-Unisen
dc.relation.forthcomingnonen
dc.identifier.doihttp://dx.doi.org/10.1109/CVPRW.2006.48en


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record