Face recognition using extended isomap

Research output: Contribution to conferencePaper

49 Citations (Scopus)

Abstract

The Isomap method has demonstrated promising results in finding low dimensional manifolds from samples in the high dimensional input space. While conventional subspace methods compute L1 or L2 metrics to represent distances between samples and apply Principal Component Analysis or alike to induce linear manifolds, the Isomap method estimates geodesic distance between samples and then uses multidimensional scaling to induce a low dimensional manifold. Since the Isomap method is developed based on reconstruction principle, it may not be optimal from the classification viewpoint. In this paper, we present an extended Isomap method that utilizes-Fisher Linear Discriminant for pattern classification. Numerous experiments on image data sets show that our extension is more effective than the original Isomap method for pattern classification. Furthermore, the extended Isomap method shows promising results compared with best classification methods in the literature.

Original languageEnglish
PagesII/117-II/120
Publication statusPublished - 2002 Jan 1
EventInternational Conference on Image Processing (ICIP'02) - Rochester, NY, United States
Duration: 2002 Sep 222002 Sep 25

Other

OtherInternational Conference on Image Processing (ICIP'02)
CountryUnited States
CityRochester, NY
Period02/9/2202/9/25

Fingerprint

Face recognition
Pattern recognition
Principal component analysis
Experiments

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Electrical and Electronic Engineering

Cite this

Yang, M. H. (2002). Face recognition using extended isomap. II/117-II/120. Paper presented at International Conference on Image Processing (ICIP'02), Rochester, NY, United States.
Yang, Ming Hsuan. / Face recognition using extended isomap. Paper presented at International Conference on Image Processing (ICIP'02), Rochester, NY, United States.
@conference{90761f4787784be39552edcb7ad2b11d,
title = "Face recognition using extended isomap",
abstract = "The Isomap method has demonstrated promising results in finding low dimensional manifolds from samples in the high dimensional input space. While conventional subspace methods compute L1 or L2 metrics to represent distances between samples and apply Principal Component Analysis or alike to induce linear manifolds, the Isomap method estimates geodesic distance between samples and then uses multidimensional scaling to induce a low dimensional manifold. Since the Isomap method is developed based on reconstruction principle, it may not be optimal from the classification viewpoint. In this paper, we present an extended Isomap method that utilizes-Fisher Linear Discriminant for pattern classification. Numerous experiments on image data sets show that our extension is more effective than the original Isomap method for pattern classification. Furthermore, the extended Isomap method shows promising results compared with best classification methods in the literature.",
author = "Yang, {Ming Hsuan}",
year = "2002",
month = "1",
day = "1",
language = "English",
pages = "II/117--II/120",
note = "International Conference on Image Processing (ICIP'02) ; Conference date: 22-09-2002 Through 25-09-2002",

}

Yang, MH 2002, 'Face recognition using extended isomap' Paper presented at International Conference on Image Processing (ICIP'02), Rochester, NY, United States, 02/9/22 - 02/9/25, pp. II/117-II/120.

Face recognition using extended isomap. / Yang, Ming Hsuan.

2002. II/117-II/120 Paper presented at International Conference on Image Processing (ICIP'02), Rochester, NY, United States.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Face recognition using extended isomap

AU - Yang, Ming Hsuan

PY - 2002/1/1

Y1 - 2002/1/1

N2 - The Isomap method has demonstrated promising results in finding low dimensional manifolds from samples in the high dimensional input space. While conventional subspace methods compute L1 or L2 metrics to represent distances between samples and apply Principal Component Analysis or alike to induce linear manifolds, the Isomap method estimates geodesic distance between samples and then uses multidimensional scaling to induce a low dimensional manifold. Since the Isomap method is developed based on reconstruction principle, it may not be optimal from the classification viewpoint. In this paper, we present an extended Isomap method that utilizes-Fisher Linear Discriminant for pattern classification. Numerous experiments on image data sets show that our extension is more effective than the original Isomap method for pattern classification. Furthermore, the extended Isomap method shows promising results compared with best classification methods in the literature.

AB - The Isomap method has demonstrated promising results in finding low dimensional manifolds from samples in the high dimensional input space. While conventional subspace methods compute L1 or L2 metrics to represent distances between samples and apply Principal Component Analysis or alike to induce linear manifolds, the Isomap method estimates geodesic distance between samples and then uses multidimensional scaling to induce a low dimensional manifold. Since the Isomap method is developed based on reconstruction principle, it may not be optimal from the classification viewpoint. In this paper, we present an extended Isomap method that utilizes-Fisher Linear Discriminant for pattern classification. Numerous experiments on image data sets show that our extension is more effective than the original Isomap method for pattern classification. Furthermore, the extended Isomap method shows promising results compared with best classification methods in the literature.

UR - http://www.scopus.com/inward/record.url?scp=18344411530&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=18344411530&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:18344411530

SP - II/117-II/120

ER -

Yang MH. Face recognition using extended isomap. 2002. Paper presented at International Conference on Image Processing (ICIP'02), Rochester, NY, United States.