Presentation 2014-03-13
Study of Clustering Feature Points of Moving Objects From Dynamic RGB-D Images : Proposing a Method that Combines Supervoxel and Multi Label Graph-Cut
Naotomo Tatematsu, Jun Ohya, Larry S. Davis,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a method that can discriminate feature points of each moving object from the still background by utilizing supervoxel and Multi-label Graph-Cut. The ordinal clustering method based on multi label graph-cut has a problem that can not classify small movement object. To solve this problem, our method utilize initial cluster based on supervoxel and construct multi-labeled-graph by the unit of group that belongs same supervoxel. This improvement enable accurate clustering feature points even if the amount of movement is small. Experiments using multiple moving objects and real stereo sequences demonstrate the effectiveness of our proposed method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English)
Paper # PRMU2013-189
Date of Issue

Conference Information
Committee PRMU
Conference Date 2014/3/6(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Study of Clustering Feature Points of Moving Objects From Dynamic RGB-D Images : Proposing a Method that Combines Supervoxel and Multi Label Graph-Cut
Sub Title (in English)
Keyword(1)
1st Author's Name Naotomo Tatematsu
1st Author's Affiliation Graduate School of Global Information and Telecommunication Studies, Waseda University()
2nd Author's Name Jun Ohya
2nd Author's Affiliation Department of Computer Science, University of Maryland College Park
3rd Author's Name Larry S. Davis
3rd Author's Affiliation Department of Computer Science, University of Maryland College Park
Date 2014-03-13
Paper # PRMU2013-189
Volume (vol) vol.113
Number (no) 493
Page pp.pp.-
#Pages 6
Date of Issue