Decomposed eigenface for face recognition under various lighting conditions

Takeshi Shakunaga, Kazuma Shigenari

Research output: Chapter in Book/Report/Conference proceedingConference contribution

40 Citations (Scopus)

Abstract

Face recognition under various lighting conditions is discussed to cover cases when too few images are available for registration. This paper proposes decomposition of an eigenface into two orthogonal eigenspaces for realizing robust face recognition under such conditions. The decomposed eigenfaces consisting of two eigenspaces are constructed for each person even if only one image is available. A universal eigenspace called the canonical space (CS) plays an important role in creating the eigenspaces by way of decomposition, where CS is constructed a priori by principal component analysis (PCA) over face images of many people under many lighting conditions. In the registration stage, an input face image is decomposed to a projection image in CS and the residual of the projection. Then two eigenspaces are created independently in CS and in the orthogonal complement CS. Some refinements of the two eigenspaces are also discussed. By combining the two eigenspaces, we can easily realize face identification that is robust to illumination change, even when too few images are registered. Through experiments, we show the effectiveness of the decomposed eigenfaces as compared with conventional methods.

Original languageEnglish
Title of host publicationProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume1
Publication statusPublished - 2001
Event2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Kauai, HI, United States
Duration: Dec 8 2001Dec 14 2001

Other

Other2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
CountryUnited States
CityKauai, HI
Period12/8/0112/14/01

Fingerprint

Face recognition
Lighting
Decomposition
Principal component analysis
Experiments

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Software
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Shakunaga, T., & Shigenari, K. (2001). Decomposed eigenface for face recognition under various lighting conditions. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Vol. 1)

Decomposed eigenface for face recognition under various lighting conditions. / Shakunaga, Takeshi; Shigenari, Kazuma.

Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Vol. 1 2001.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Shakunaga, T & Shigenari, K 2001, Decomposed eigenface for face recognition under various lighting conditions. in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. vol. 1, 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Kauai, HI, United States, 12/8/01.
Shakunaga T, Shigenari K. Decomposed eigenface for face recognition under various lighting conditions. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Vol. 1. 2001
Shakunaga, Takeshi ; Shigenari, Kazuma. / Decomposed eigenface for face recognition under various lighting conditions. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Vol. 1 2001.
@inproceedings{8a7cf4bec0b24c6f9906e7de6cf10887,
title = "Decomposed eigenface for face recognition under various lighting conditions",
abstract = "Face recognition under various lighting conditions is discussed to cover cases when too few images are available for registration. This paper proposes decomposition of an eigenface into two orthogonal eigenspaces for realizing robust face recognition under such conditions. The decomposed eigenfaces consisting of two eigenspaces are constructed for each person even if only one image is available. A universal eigenspace called the canonical space (CS) plays an important role in creating the eigenspaces by way of decomposition, where CS is constructed a priori by principal component analysis (PCA) over face images of many people under many lighting conditions. In the registration stage, an input face image is decomposed to a projection image in CS and the residual of the projection. Then two eigenspaces are created independently in CS and in the orthogonal complement CS⊥. Some refinements of the two eigenspaces are also discussed. By combining the two eigenspaces, we can easily realize face identification that is robust to illumination change, even when too few images are registered. Through experiments, we show the effectiveness of the decomposed eigenfaces as compared with conventional methods.",
author = "Takeshi Shakunaga and Kazuma Shigenari",
year = "2001",
language = "English",
volume = "1",
booktitle = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

}

TY - GEN

T1 - Decomposed eigenface for face recognition under various lighting conditions

AU - Shakunaga, Takeshi

AU - Shigenari, Kazuma

PY - 2001

Y1 - 2001

N2 - Face recognition under various lighting conditions is discussed to cover cases when too few images are available for registration. This paper proposes decomposition of an eigenface into two orthogonal eigenspaces for realizing robust face recognition under such conditions. The decomposed eigenfaces consisting of two eigenspaces are constructed for each person even if only one image is available. A universal eigenspace called the canonical space (CS) plays an important role in creating the eigenspaces by way of decomposition, where CS is constructed a priori by principal component analysis (PCA) over face images of many people under many lighting conditions. In the registration stage, an input face image is decomposed to a projection image in CS and the residual of the projection. Then two eigenspaces are created independently in CS and in the orthogonal complement CS⊥. Some refinements of the two eigenspaces are also discussed. By combining the two eigenspaces, we can easily realize face identification that is robust to illumination change, even when too few images are registered. Through experiments, we show the effectiveness of the decomposed eigenfaces as compared with conventional methods.

AB - Face recognition under various lighting conditions is discussed to cover cases when too few images are available for registration. This paper proposes decomposition of an eigenface into two orthogonal eigenspaces for realizing robust face recognition under such conditions. The decomposed eigenfaces consisting of two eigenspaces are constructed for each person even if only one image is available. A universal eigenspace called the canonical space (CS) plays an important role in creating the eigenspaces by way of decomposition, where CS is constructed a priori by principal component analysis (PCA) over face images of many people under many lighting conditions. In the registration stage, an input face image is decomposed to a projection image in CS and the residual of the projection. Then two eigenspaces are created independently in CS and in the orthogonal complement CS⊥. Some refinements of the two eigenspaces are also discussed. By combining the two eigenspaces, we can easily realize face identification that is robust to illumination change, even when too few images are registered. Through experiments, we show the effectiveness of the decomposed eigenfaces as compared with conventional methods.

UR - http://www.scopus.com/inward/record.url?scp=0035680150&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035680150&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0035680150

VL - 1

BT - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

ER -