Presentation 1998/6/18
Shape Reconstruction by Random Sampling and Voting
Kazuhiko Kawamoto, Tomoki Ueno, Atsushi Imiya,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper we proposes an algorithm for a three-dimensional object reconstruction which dose not assume any predetermination of point-correspondences among frames. We randomly select points from a series of images and check the epipolar constraint. Furthermore, if a pair of points hold the epipolar constraint, a pair of points determine a spatial position. However, this procedure also determines artifact points. These artifact points are yield appling the epipolar equation to non-corresponding points. Therefor, we extract true points using voting procedure which infure solutions by collecting many evidence. This idea is an extension of the randomized Hough transform for the extraction of lines on an image.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Random Sampling / Voting / Epipolar Constraint / Randomized Hough Transform
Paper # PRMU98-28
Date of Issue

Conference Information
Committee PRMU
Conference Date 1998/6/18(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) Shape Reconstruction by Random Sampling and Voting
Sub Title (in English)
Keyword(1) Random Sampling
Keyword(2) Voting
Keyword(3) Epipolar Constraint
Keyword(4) Randomized Hough Transform
1st Author's Name Kazuhiko Kawamoto
1st Author's Affiliation Dept. of Information and Image Sciences, Chiba University()
2nd Author's Name Tomoki Ueno
2nd Author's Affiliation Dept. of Information and Image Sciences, Chiba University
3rd Author's Name Atsushi Imiya
3rd Author's Affiliation Dept. of Information and Image Sciences, Chiba University
Date 1998/6/18
Paper # PRMU98-28
Volume (vol) vol.98
Number (no) 126
Page pp.pp.-
#Pages 8
Date of Issue