Optimization without minimization search

Constraint satisfaction by orthogonal projection with applications to multiview triangulation

Kenichi Kanatani, Yasuyuki Sugaya, Hirotaka Niitsuma

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We present an alternative approach to what we call the "standard optimization", which minimizes a cost function by searching a parameter space. Instead, our approach "projects" in the joint observation space onto the manifold defined by the "consistency constraint", which de-mands that any minimal subset of observations produce the same result. This approach avoids many difficulties encountered in the standard opti-mization. As typical examples, we apply it to line fitting and multiview triangulation. The latter produces a new algorithm far more efficient than existing methods. We also discuss the optimality of our approach.

Original languageEnglish
Pages (from-to)2836-2845
Number of pages10
JournalIEICE Transactions on Information and Systems
VolumeE93-D
Issue number10
DOIs
Publication statusPublished - Oct 2010

Fingerprint

Triangulation
Cost functions

Keywords

  • Consistency constraint satisfaction
  • Line fitting
  • Multiview triangulation
  • Orthogonal projection
  • Trifocal tensor

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Software
  • Artificial Intelligence
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition

Cite this

Optimization without minimization search : Constraint satisfaction by orthogonal projection with applications to multiview triangulation. / Kanatani, Kenichi; Sugaya, Yasuyuki; Niitsuma, Hirotaka.

In: IEICE Transactions on Information and Systems, Vol. E93-D, No. 10, 10.2010, p. 2836-2845.

Research output: Contribution to journalArticle

@article{950189419f5d43afab2a61df5e612b61,
title = "Optimization without minimization search: Constraint satisfaction by orthogonal projection with applications to multiview triangulation",
abstract = "We present an alternative approach to what we call the {"}standard optimization{"}, which minimizes a cost function by searching a parameter space. Instead, our approach {"}projects{"} in the joint observation space onto the manifold defined by the {"}consistency constraint{"}, which de-mands that any minimal subset of observations produce the same result. This approach avoids many difficulties encountered in the standard opti-mization. As typical examples, we apply it to line fitting and multiview triangulation. The latter produces a new algorithm far more efficient than existing methods. We also discuss the optimality of our approach.",
keywords = "Consistency constraint satisfaction, Line fitting, Multiview triangulation, Orthogonal projection, Trifocal tensor",
author = "Kenichi Kanatani and Yasuyuki Sugaya and Hirotaka Niitsuma",
year = "2010",
month = "10",
doi = "10.1587/transinf.E93.D.2836",
language = "English",
volume = "E93-D",
pages = "2836--2845",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "10",

}

TY - JOUR

T1 - Optimization without minimization search

T2 - Constraint satisfaction by orthogonal projection with applications to multiview triangulation

AU - Kanatani, Kenichi

AU - Sugaya, Yasuyuki

AU - Niitsuma, Hirotaka

PY - 2010/10

Y1 - 2010/10

N2 - We present an alternative approach to what we call the "standard optimization", which minimizes a cost function by searching a parameter space. Instead, our approach "projects" in the joint observation space onto the manifold defined by the "consistency constraint", which de-mands that any minimal subset of observations produce the same result. This approach avoids many difficulties encountered in the standard opti-mization. As typical examples, we apply it to line fitting and multiview triangulation. The latter produces a new algorithm far more efficient than existing methods. We also discuss the optimality of our approach.

AB - We present an alternative approach to what we call the "standard optimization", which minimizes a cost function by searching a parameter space. Instead, our approach "projects" in the joint observation space onto the manifold defined by the "consistency constraint", which de-mands that any minimal subset of observations produce the same result. This approach avoids many difficulties encountered in the standard opti-mization. As typical examples, we apply it to line fitting and multiview triangulation. The latter produces a new algorithm far more efficient than existing methods. We also discuss the optimality of our approach.

KW - Consistency constraint satisfaction

KW - Line fitting

KW - Multiview triangulation

KW - Orthogonal projection

KW - Trifocal tensor

UR - http://www.scopus.com/inward/record.url?scp=77957847697&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957847697&partnerID=8YFLogxK

U2 - 10.1587/transinf.E93.D.2836

DO - 10.1587/transinf.E93.D.2836

M3 - Article

VL - E93-D

SP - 2836

EP - 2845

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 10

ER -