Local linear discriminant analysis with composite kernel for face recognition

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

7 Citations (Scopus)

Abstract

This paper presents a method for nonlinear discriminant analysis utilizing a composite kernel which is derived from a combination of local linear models with interpolation. The underlying idea is to decompose a complex nonlinear problem into a set of simpler local linear problems. Combining with the theory of nonlinear classification based on kernels, the local linear models with interpolation can be formulated as a composite kernel based discriminant analysis form. In face recognition, linear discriminant analysis (LDA) has been widely adopted owing to its efficiency, but it fails to solve nonlinear problems. Conventional kernel based approaches such as generalized discriminant analysis (GDA) has been successfully applied to extend LDA to nonlinear pattern recognition tasks. However, selecting an appropriate kernel function is usually difficult. Utilizing an implicit kernel mapping may face potential over-training problems for some complex and noised tasks. Our proposed method gives an alternative solution for nonlinear discriminant analysis while the conventional linear and nonlinear approaches are difficult to achieve a satisfactory results. Experiments on both synthetic data and face data set show the effectiveness of the proposed methods.

Original languageEnglish
Title of host publicationProceedings of the International Joint Conference on Neural Networks
DOIs
Publication statusPublished - 2012
Event2012 Annual International Joint Conference on Neural Networks, IJCNN 2012, Part of the 2012 IEEE World Congress on Computational Intelligence, WCCI 2012 - Brisbane, QLD
Duration: 2012 Jun 102012 Jun 15

Other

Other2012 Annual International Joint Conference on Neural Networks, IJCNN 2012, Part of the 2012 IEEE World Congress on Computational Intelligence, WCCI 2012
CityBrisbane, QLD
Period12/6/1012/6/15

Fingerprint

Discriminant analysis
Face recognition
Composite materials
Interpolation
Pattern recognition
Experiments

Keywords

  • composite kernel
  • dimensionality reduction
  • generalized discriminant analysis
  • Linear discriminant analysis
  • local linear model
  • support vector machine

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Shi, Z., & Furuzuki, T. (2012). Local linear discriminant analysis with composite kernel for face recognition. In Proceedings of the International Joint Conference on Neural Networks [6252385] https://doi.org/10.1109/IJCNN.2012.6252385

Local linear discriminant analysis with composite kernel for face recognition. / Shi, Zhan; Furuzuki, Takayuki.

Proceedings of the International Joint Conference on Neural Networks. 2012. 6252385.

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

Shi, Z & Furuzuki, T 2012, Local linear discriminant analysis with composite kernel for face recognition. in Proceedings of the International Joint Conference on Neural Networks., 6252385, 2012 Annual International Joint Conference on Neural Networks, IJCNN 2012, Part of the 2012 IEEE World Congress on Computational Intelligence, WCCI 2012, Brisbane, QLD, 12/6/10. https://doi.org/10.1109/IJCNN.2012.6252385
Shi Z, Furuzuki T. Local linear discriminant analysis with composite kernel for face recognition. In Proceedings of the International Joint Conference on Neural Networks. 2012. 6252385 https://doi.org/10.1109/IJCNN.2012.6252385
Shi, Zhan ; Furuzuki, Takayuki. / Local linear discriminant analysis with composite kernel for face recognition. Proceedings of the International Joint Conference on Neural Networks. 2012.
@inproceedings{102bbabffc964e118b7b7a9c88dc3b24,
title = "Local linear discriminant analysis with composite kernel for face recognition",
abstract = "This paper presents a method for nonlinear discriminant analysis utilizing a composite kernel which is derived from a combination of local linear models with interpolation. The underlying idea is to decompose a complex nonlinear problem into a set of simpler local linear problems. Combining with the theory of nonlinear classification based on kernels, the local linear models with interpolation can be formulated as a composite kernel based discriminant analysis form. In face recognition, linear discriminant analysis (LDA) has been widely adopted owing to its efficiency, but it fails to solve nonlinear problems. Conventional kernel based approaches such as generalized discriminant analysis (GDA) has been successfully applied to extend LDA to nonlinear pattern recognition tasks. However, selecting an appropriate kernel function is usually difficult. Utilizing an implicit kernel mapping may face potential over-training problems for some complex and noised tasks. Our proposed method gives an alternative solution for nonlinear discriminant analysis while the conventional linear and nonlinear approaches are difficult to achieve a satisfactory results. Experiments on both synthetic data and face data set show the effectiveness of the proposed methods.",
keywords = "composite kernel, dimensionality reduction, generalized discriminant analysis, Linear discriminant analysis, local linear model, support vector machine",
author = "Zhan Shi and Takayuki Furuzuki",
year = "2012",
doi = "10.1109/IJCNN.2012.6252385",
language = "English",
isbn = "9781467314909",
booktitle = "Proceedings of the International Joint Conference on Neural Networks",

}

TY - GEN

T1 - Local linear discriminant analysis with composite kernel for face recognition

AU - Shi, Zhan

AU - Furuzuki, Takayuki

PY - 2012

Y1 - 2012

N2 - This paper presents a method for nonlinear discriminant analysis utilizing a composite kernel which is derived from a combination of local linear models with interpolation. The underlying idea is to decompose a complex nonlinear problem into a set of simpler local linear problems. Combining with the theory of nonlinear classification based on kernels, the local linear models with interpolation can be formulated as a composite kernel based discriminant analysis form. In face recognition, linear discriminant analysis (LDA) has been widely adopted owing to its efficiency, but it fails to solve nonlinear problems. Conventional kernel based approaches such as generalized discriminant analysis (GDA) has been successfully applied to extend LDA to nonlinear pattern recognition tasks. However, selecting an appropriate kernel function is usually difficult. Utilizing an implicit kernel mapping may face potential over-training problems for some complex and noised tasks. Our proposed method gives an alternative solution for nonlinear discriminant analysis while the conventional linear and nonlinear approaches are difficult to achieve a satisfactory results. Experiments on both synthetic data and face data set show the effectiveness of the proposed methods.

AB - This paper presents a method for nonlinear discriminant analysis utilizing a composite kernel which is derived from a combination of local linear models with interpolation. The underlying idea is to decompose a complex nonlinear problem into a set of simpler local linear problems. Combining with the theory of nonlinear classification based on kernels, the local linear models with interpolation can be formulated as a composite kernel based discriminant analysis form. In face recognition, linear discriminant analysis (LDA) has been widely adopted owing to its efficiency, but it fails to solve nonlinear problems. Conventional kernel based approaches such as generalized discriminant analysis (GDA) has been successfully applied to extend LDA to nonlinear pattern recognition tasks. However, selecting an appropriate kernel function is usually difficult. Utilizing an implicit kernel mapping may face potential over-training problems for some complex and noised tasks. Our proposed method gives an alternative solution for nonlinear discriminant analysis while the conventional linear and nonlinear approaches are difficult to achieve a satisfactory results. Experiments on both synthetic data and face data set show the effectiveness of the proposed methods.

KW - composite kernel

KW - dimensionality reduction

KW - generalized discriminant analysis

KW - Linear discriminant analysis

KW - local linear model

KW - support vector machine

UR - http://www.scopus.com/inward/record.url?scp=84865076502&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865076502&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2012.6252385

DO - 10.1109/IJCNN.2012.6252385

M3 - Conference contribution

AN - SCOPUS:84865076502

SN - 9781467314909

BT - Proceedings of the International Joint Conference on Neural Networks

ER -