Robust Projection onto Normalized Eigenspace Using Relative Residual Analysis and Optimal Partial Projection

Fumihiko Sakaue, Takeshi Shakunaga

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

The present paper reports a robust projection onto eigenspace that is based on iterative projection. The fundamental method proposed in Shakunaga and Sakaue [11] and involves iterative analysis of relative residual and projection. The present paper refines the projection method by solving linear equations while taking noise ratio into account. The refinement improves both the efficiency and robustness of the projection. Experimental results indicate that the proposed method works well for various kinds of noise, including shadows, reflections and occlusions. The proposed method can be applied to a wide variety of computer vision problems, which include object/face recognition and image-based rendering.

Original languageEnglish
Pages (from-to)34-41
Number of pages8
JournalIEICE Transactions on Information and Systems
VolumeE87-D
Issue number1
Publication statusPublished - Jan 2004

Fingerprint

Face recognition
Linear equations
Computer vision

Keywords

  • Eigenspace
  • Face recognition
  • Relative residual
  • Robust projection

ASJC Scopus subject areas

  • Information Systems
  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Robust Projection onto Normalized Eigenspace Using Relative Residual Analysis and Optimal Partial Projection. / Sakaue, Fumihiko; Shakunaga, Takeshi.

In: IEICE Transactions on Information and Systems, Vol. E87-D, No. 1, 01.2004, p. 34-41.

Research output: Contribution to journalArticle

@article{9e089475f10f45febb30e76c3766eaf3,
title = "Robust Projection onto Normalized Eigenspace Using Relative Residual Analysis and Optimal Partial Projection",
abstract = "The present paper reports a robust projection onto eigenspace that is based on iterative projection. The fundamental method proposed in Shakunaga and Sakaue [11] and involves iterative analysis of relative residual and projection. The present paper refines the projection method by solving linear equations while taking noise ratio into account. The refinement improves both the efficiency and robustness of the projection. Experimental results indicate that the proposed method works well for various kinds of noise, including shadows, reflections and occlusions. The proposed method can be applied to a wide variety of computer vision problems, which include object/face recognition and image-based rendering.",
keywords = "Eigenspace, Face recognition, Relative residual, Robust projection",
author = "Fumihiko Sakaue and Takeshi Shakunaga",
year = "2004",
month = "1",
language = "English",
volume = "E87-D",
pages = "34--41",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "1",

}

TY - JOUR

T1 - Robust Projection onto Normalized Eigenspace Using Relative Residual Analysis and Optimal Partial Projection

AU - Sakaue, Fumihiko

AU - Shakunaga, Takeshi

PY - 2004/1

Y1 - 2004/1

N2 - The present paper reports a robust projection onto eigenspace that is based on iterative projection. The fundamental method proposed in Shakunaga and Sakaue [11] and involves iterative analysis of relative residual and projection. The present paper refines the projection method by solving linear equations while taking noise ratio into account. The refinement improves both the efficiency and robustness of the projection. Experimental results indicate that the proposed method works well for various kinds of noise, including shadows, reflections and occlusions. The proposed method can be applied to a wide variety of computer vision problems, which include object/face recognition and image-based rendering.

AB - The present paper reports a robust projection onto eigenspace that is based on iterative projection. The fundamental method proposed in Shakunaga and Sakaue [11] and involves iterative analysis of relative residual and projection. The present paper refines the projection method by solving linear equations while taking noise ratio into account. The refinement improves both the efficiency and robustness of the projection. Experimental results indicate that the proposed method works well for various kinds of noise, including shadows, reflections and occlusions. The proposed method can be applied to a wide variety of computer vision problems, which include object/face recognition and image-based rendering.

KW - Eigenspace

KW - Face recognition

KW - Relative residual

KW - Robust projection

UR - http://www.scopus.com/inward/record.url?scp=0842331883&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0842331883&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0842331883

VL - E87-D

SP - 34

EP - 41

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 1

ER -