Presentation 2015-02-19
Stereo Matching using Particle Swarm Belief Propagation
Ryosuke FURUTA, Toshihiko YAMASAKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Stereo matching is recently formulated as a continuous labeling problem using the slanted plane model, which improves the accuracy in a sub-pixel level order. In this model, label candidates are often generated by the PatchMatch strategy, and the energy minimization problem in the candidates is solved by discrete optimization techniques. In this paper, we propose particle swarm belief propagation, which combines particle swarm optimization (PSO) and belief propagation (BP). In our method, unlike the conventional methods using PatchMatch, label candidates are generated by PSO, based on the messages caluculated in the optimization process using BP. Experimental results show that our method achieves the solutions which have 2-5 times lower energy as compared to the conventional method in the Middlebury stereo benchmark.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Stereo Matching / Particle Swarm Optimization / Belief Propagation
Paper # PRMU2014-117,CNR2014-32
Date of Issue

Conference Information
Committee PRMU
Conference Date 2015/2/12(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) Stereo Matching using Particle Swarm Belief Propagation
Sub Title (in English)
Keyword(1) Stereo Matching
Keyword(2) Particle Swarm Optimization
Keyword(3) Belief Propagation
1st Author's Name Ryosuke FURUTA
1st Author's Affiliation Department of Information and Communication Engineering, The University of Tokyo()
2nd Author's Name Toshihiko YAMASAKI
2nd Author's Affiliation Department of Information and Communication Engineering, The University of Tokyo
Date 2015-02-19
Paper # PRMU2014-117,CNR2014-32
Volume (vol) vol.114
Number (no) 454
Page pp.pp.-
#Pages 6
Date of Issue