Show simple item record

Using Point Correspondences Without Projective Deformation For Multi-View Stereo Reconstruction

dc.contributor.authorVincent, Nicole
dc.contributor.authorCohen, Laurent D.
HAL ID: 738939
dc.contributor.authorAuclair, Adrien
dc.date.accessioned2012-05-09T14:37:27Z
dc.date.available2012-05-09T14:37:27Z
dc.date.issued2008
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/9160
dc.language.isoenen
dc.subjectLucas–Kanade methoden
dc.subjectPattern recognition systemsen
dc.subjectScale Invariant Features Transformen
dc.subject.ddc006.3en
dc.titleUsing Point Correspondences Without Projective Deformation For Multi-View Stereo Reconstructionen
dc.typeCommunication / Conférence
dc.contributor.editoruniversityotherCentre de Recherche en Informatique de Paris 5 (CRIP5) http://www.math-info.univ-paris5.fr/crip5/ Université Paris V - Paris Descartes;France
dc.description.abstractenThis paper proposes a novel algorithm to reconstruct a 3D surface from a calibrated set of images. In a first pass, it uses Scale Invariant Features Transform (SIFT) descriptor correspondences to drive the deformation of a mesh toward the true object surface. We introduce a method to handle the fact that these local descriptors are computed at positions that are not projections of mesh vertices in the images. In order to avoid projective deformations due to the large windows of interest of this descriptor, correspondences are only computed between images from the same viewpoint. This is used in a first pass to recover large concavities of the object. In a second pass, a one dimensional Lucas-Kanade tracker is used to recover small scale details. Using publicly available benchmarks, our algorithm obtains high accuracy while being among the fastest ones.en
dc.identifier.citationpages193-196en
dc.relation.ispartoftitleConference on Image Processing, 2008. ICIP 2008. 15th IEEE Internationalen
dc.relation.ispartofpublnameIEEEen
dc.relation.ispartofdate2008
dc.description.sponsorshipprivateouien
dc.subject.ddclabelIntelligence artificielleen
dc.relation.ispartofisbnE-ISBN : 978-1-4244-1764-3 Print ISBN: 978-1-4244-1765-0en
dc.relation.conftitleICIP 2008en
dc.relation.confdate2008-09
dc.relation.confcitySan Diegoen
dc.relation.confcountryÉtats-Unisen
dc.identifier.doihttp://dx.doi.org/10.1109/ICIP.2008.4711724


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record