Dimensionality reduction based on nonparametric discriminant analysis with kernels for feature extraction and recognition

Jun Bao Li; Shu Chuan Chu; Jeng Shyang Pan

doi:10.1145/2108616.2108652

Dimensionality reduction based on nonparametric discriminant analysis with kernels for feature extraction and recognition

Jun Bao Li, Shu Chuan Chu, Jeng Shyang Pan

Research output: Contribution to conference › Paper › peer-review

Abstract

Dimensionality reduction is the most popular method for feature extraction and recognition. Recently, Li et al. (IEEE PAMI, 2009) proposed Nonparametric Discriminant Analysis (NDA) based dimensionality reduction for face recognition and reported an excellent recognition performance. However, NDA has its limitations on extracting the nonlinear features of face images for recognition, and owing to the highly nonlinear and complex distribution of face images under a perceivable variation in viewpoint, illumination or facial expression. In order to increase the NDA, we extend the NDA with kernel trick to propose Nonparametric Kernel Discriminant Analysis (NKDA) for feature extraction and recognition. Experimental results on ORL, YALE and UMIST face databases show that NKDA outperforms NDA on recognition, which demonstrates that it is feasible to improve NDA with kernel trick for feature extraction.

Original language	English
Number of pages	8
DOIs	https://doi.org/10.1145/2108616.2108652
Publication status	Published - Jan 2010
Externally published	Yes
Event	4th International Conference on Ubiquitous Information Management and Communication, ICUIMC'10 - Suwon, Korea, Republic of Duration: 14 Jan 2010 → 15 Jan 2010

Conference

Conference	4th International Conference on Ubiquitous Information Management and Communication, ICUIMC'10
Country/Territory	Korea, Republic of
City	Suwon
Period	14/01/10 → 15/01/10

Keywords

face recognition
kernel method
nonparametric discriminant analysis
nonparametric kernel discriminant analysis

Access to Document

10.1145/2108616.2108652Licence: In Copyright

Cite this

@conference{e1bb1b22f5524893afa374de48cb1bbd,

title = "Dimensionality reduction based on nonparametric discriminant analysis with kernels for feature extraction and recognition",

abstract = "Dimensionality reduction is the most popular method for feature extraction and recognition. Recently, Li et al. (IEEE PAMI, 2009) proposed Nonparametric Discriminant Analysis (NDA) based dimensionality reduction for face recognition and reported an excellent recognition performance. However, NDA has its limitations on extracting the nonlinear features of face images for recognition, and owing to the highly nonlinear and complex distribution of face images under a perceivable variation in viewpoint, illumination or facial expression. In order to increase the NDA, we extend the NDA with kernel trick to propose Nonparametric Kernel Discriminant Analysis (NKDA) for feature extraction and recognition. Experimental results on ORL, YALE and UMIST face databases show that NKDA outperforms NDA on recognition, which demonstrates that it is feasible to improve NDA with kernel trick for feature extraction.",

keywords = "face recognition, kernel method, nonparametric discriminant analysis, nonparametric kernel discriminant analysis",

author = "Li, {Jun Bao} and Chu, {Shu Chuan} and Pan, {Jeng Shyang}",

year = "2010",

month = jan,

doi = "10.1145/2108616.2108652",

language = "English",

note = "4th International Conference on Ubiquitous Information Management and Communication, ICUIMC'10 ; Conference date: 14-01-2010 Through 15-01-2010",

}

Dimensionality reduction based on nonparametric discriminant analysis with kernels for feature extraction and recognition. / Li, Jun Bao; Chu, Shu Chuan; Pan, Jeng Shyang.
2010. Paper presented at 4th International Conference on Ubiquitous Information Management and Communication, ICUIMC'10, Suwon, Korea, Republic of.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Dimensionality reduction based on nonparametric discriminant analysis with kernels for feature extraction and recognition

AU - Li, Jun Bao

AU - Chu, Shu Chuan

AU - Pan, Jeng Shyang

PY - 2010/1

Y1 - 2010/1

N2 - Dimensionality reduction is the most popular method for feature extraction and recognition. Recently, Li et al. (IEEE PAMI, 2009) proposed Nonparametric Discriminant Analysis (NDA) based dimensionality reduction for face recognition and reported an excellent recognition performance. However, NDA has its limitations on extracting the nonlinear features of face images for recognition, and owing to the highly nonlinear and complex distribution of face images under a perceivable variation in viewpoint, illumination or facial expression. In order to increase the NDA, we extend the NDA with kernel trick to propose Nonparametric Kernel Discriminant Analysis (NKDA) for feature extraction and recognition. Experimental results on ORL, YALE and UMIST face databases show that NKDA outperforms NDA on recognition, which demonstrates that it is feasible to improve NDA with kernel trick for feature extraction.

AB - Dimensionality reduction is the most popular method for feature extraction and recognition. Recently, Li et al. (IEEE PAMI, 2009) proposed Nonparametric Discriminant Analysis (NDA) based dimensionality reduction for face recognition and reported an excellent recognition performance. However, NDA has its limitations on extracting the nonlinear features of face images for recognition, and owing to the highly nonlinear and complex distribution of face images under a perceivable variation in viewpoint, illumination or facial expression. In order to increase the NDA, we extend the NDA with kernel trick to propose Nonparametric Kernel Discriminant Analysis (NKDA) for feature extraction and recognition. Experimental results on ORL, YALE and UMIST face databases show that NKDA outperforms NDA on recognition, which demonstrates that it is feasible to improve NDA with kernel trick for feature extraction.

KW - face recognition

KW - kernel method

KW - nonparametric discriminant analysis

KW - nonparametric kernel discriminant analysis

UR - http://www.scopus.com/inward/record.url?scp=84863288571&partnerID=8YFLogxK

U2 - 10.1145/2108616.2108652

DO - 10.1145/2108616.2108652

M3 - Paper

AN - SCOPUS:84863288571

T2 - 4th International Conference on Ubiquitous Information Management and Communication, ICUIMC'10

Y2 - 14 January 2010 through 15 January 2010

ER -

Dimensionality reduction based on nonparametric discriminant analysis with kernels for feature extraction and recognition

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this