Coarse-to-fine region selection and matching

Handle URI:
http://hdl.handle.net/10754/621243
Title:
Coarse-to-fine region selection and matching
Authors:
Yang, Yanchao; Lu, Zhaojin ( 0000-0003-1429-5033 ) ; Sundaramoorthi, Ganesh ( 0000-0003-3471-6384 )
Abstract:
We present a new approach to wide baseline matching. We propose to use a hierarchical decomposition of the image domain and coarse-to-fine selection of regions to match. In contrast to interest point matching methods, which sample salient regions to reduce the cost of comparing all regions in two images, our method eliminates regions systematically to achieve efficiency. One advantage of our approach is that it is not restricted to covariant salient regions, which is too restrictive under large viewpoint and leads to few corresponding regions. Affine invariant matching of regions in the hierarchy is achieved efficiently by a coarse-to-fine search of the affine space. Experiments on two benchmark datasets shows that our method finds more correct correspondence of the image (with fewer false alarms) than other wide baseline methods on large viewpoint change. © 2015 IEEE.
KAUST Department:
KAUST, Saudi Arabia
Citation:
Yang Y, Zhaojin Lu, Sundaramoorthi G (2015) Coarse-to-fine region selection and matching. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Available: http://dx.doi.org/10.1109/CVPR.2015.7299140.
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
Conference/Event name:
IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015
Issue Date:
15-Oct-2015
DOI:
10.1109/CVPR.2015.7299140
Type:
Conference Paper
Appears in Collections:
Conference Papers

Full metadata record

DC FieldValue Language
dc.contributor.authorYang, Yanchaoen
dc.contributor.authorLu, Zhaojinen
dc.contributor.authorSundaramoorthi, Ganeshen
dc.date.accessioned2016-11-03T06:56:15Z-
dc.date.available2016-11-03T06:56:15Z-
dc.date.issued2015-10-15en
dc.identifier.citationYang Y, Zhaojin Lu, Sundaramoorthi G (2015) Coarse-to-fine region selection and matching. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Available: http://dx.doi.org/10.1109/CVPR.2015.7299140.en
dc.identifier.doi10.1109/CVPR.2015.7299140en
dc.identifier.urihttp://hdl.handle.net/10754/621243-
dc.description.abstractWe present a new approach to wide baseline matching. We propose to use a hierarchical decomposition of the image domain and coarse-to-fine selection of regions to match. In contrast to interest point matching methods, which sample salient regions to reduce the cost of comparing all regions in two images, our method eliminates regions systematically to achieve efficiency. One advantage of our approach is that it is not restricted to covariant salient regions, which is too restrictive under large viewpoint and leads to few corresponding regions. Affine invariant matching of regions in the hierarchy is achieved efficiently by a coarse-to-fine search of the affine space. Experiments on two benchmark datasets shows that our method finds more correct correspondence of the image (with fewer false alarms) than other wide baseline methods on large viewpoint change. © 2015 IEEE.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.titleCoarse-to-fine region selection and matchingen
dc.typeConference Paperen
dc.contributor.departmentKAUST, Saudi Arabiaen
dc.identifier.journal2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)en
dc.conference.date7 June 2015 through 12 June 2015en
dc.conference.nameIEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015en
dc.contributor.institutionUniversity of California, Los Angeles, United Statesen
dc.contributor.institutionInstitute of Automation, Chinese Academy of Sciences, Chinaen
kaust.authorYang, Yanchaoen
kaust.authorLu, Zhaojinen
kaust.authorSundaramoorthi, Ganeshen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.