Multi-class support vector machines with case-based combination for face recognition

Jaepil Ko, Hyeran Byun

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

The support vector machine is basically to deal with a two-class classification problem. To get M-class classifiers for face recognition, it is common to construct a set of binary classifiers f1,....fM, each trained to separate one class from the rest. The multi-class classification method has a main shortcoming that the binary classifiers used are obtained by training on different binary classification problems, and thus it is unclear whether their real-valued outputs are on comparable scales. In this paper, we try to use additional information, relative outputs of the machines, for final decision. We propose case-based combination with reject option to use the information. The experiments on the ORL face database shows that the proposed method achieves a slight better performance than the previous multi-class support vector machines.

Original languageEnglish
Pages (from-to)623-629
Number of pages7
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2756
Publication statusPublished - 2003 Dec 1

Fingerprint

Multi-class
Face recognition
Face Recognition
Support vector machines
Support Vector Machine
Classifiers
Classifier
Classification Problems
Binary
Multi-class Classification
Binary Classification
Output
Information use
Face
Experiment
Class
Experiments

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{05f9577b2e724ea785c8c591b7ca2f02,
title = "Multi-class support vector machines with case-based combination for face recognition",
abstract = "The support vector machine is basically to deal with a two-class classification problem. To get M-class classifiers for face recognition, it is common to construct a set of binary classifiers f1,....fM, each trained to separate one class from the rest. The multi-class classification method has a main shortcoming that the binary classifiers used are obtained by training on different binary classification problems, and thus it is unclear whether their real-valued outputs are on comparable scales. In this paper, we try to use additional information, relative outputs of the machines, for final decision. We propose case-based combination with reject option to use the information. The experiments on the ORL face database shows that the proposed method achieves a slight better performance than the previous multi-class support vector machines.",
author = "Jaepil Ko and Hyeran Byun",
year = "2003",
month = "12",
day = "1",
language = "English",
volume = "2756",
pages = "623--629",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Multi-class support vector machines with case-based combination for face recognition

AU - Ko, Jaepil

AU - Byun, Hyeran

PY - 2003/12/1

Y1 - 2003/12/1

N2 - The support vector machine is basically to deal with a two-class classification problem. To get M-class classifiers for face recognition, it is common to construct a set of binary classifiers f1,....fM, each trained to separate one class from the rest. The multi-class classification method has a main shortcoming that the binary classifiers used are obtained by training on different binary classification problems, and thus it is unclear whether their real-valued outputs are on comparable scales. In this paper, we try to use additional information, relative outputs of the machines, for final decision. We propose case-based combination with reject option to use the information. The experiments on the ORL face database shows that the proposed method achieves a slight better performance than the previous multi-class support vector machines.

AB - The support vector machine is basically to deal with a two-class classification problem. To get M-class classifiers for face recognition, it is common to construct a set of binary classifiers f1,....fM, each trained to separate one class from the rest. The multi-class classification method has a main shortcoming that the binary classifiers used are obtained by training on different binary classification problems, and thus it is unclear whether their real-valued outputs are on comparable scales. In this paper, we try to use additional information, relative outputs of the machines, for final decision. We propose case-based combination with reject option to use the information. The experiments on the ORL face database shows that the proposed method achieves a slight better performance than the previous multi-class support vector machines.

UR - http://www.scopus.com/inward/record.url?scp=33745925925&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745925925&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:33745925925

VL - 2756

SP - 623

EP - 629

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -