Presentation 1996/2/3
Stereo matching network from transparent and ambiguous surfaces
Tomoya Tsukahara, Yuzo Hirai,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper a stereo matching network which can reconstruct 3-D surfaces portrayed in random-dot stereograms is proposed. Forbidden cone with anisotropic boundary is used as constraints. Because matching elements in this network is arrayed differently between horizontal axis and vertical one, disparity gradient changes between 2.0 and l.0 respectively. Around monocular region excitatory interaction is employed to discover correct matching elements. This interaction also facilitate a reconstruction of transparent or ambiguous surfaces. A series of simulation studies shows that the present network can recover various surfaces successfully.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) stereo matching / forbidden cone / disparity gradient limit / transparent surface / ambiguous surface
Paper # NC95-102
Date of Issue

Conference Information
Committee NC
Conference Date 1996/2/3(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 Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Stereo matching network from transparent and ambiguous surfaces
Sub Title (in English)
Keyword(1) stereo matching
Keyword(2) forbidden cone
Keyword(3) disparity gradient limit
Keyword(4) transparent surface
Keyword(5) ambiguous surface
1st Author's Name Tomoya Tsukahara
1st Author's Affiliation Doctoral Programs in Engineering, University of Tsukuba()
2nd Author's Name Yuzo Hirai
2nd Author's Affiliation Institute of Information Sciences and Electronics, University of Tsukuba
Date 1996/2/3
Paper # NC95-102
Volume (vol) vol.95
Number (no) 506
Page pp.pp.-
#Pages 8
Date of Issue