Support Vector Machines for visual gender classification

Ming Hsuan Yang, Baback Moghaddam

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Support Vector Machines (SVMs) are investigated for visual gender classification with low-resolution "thumbnail" faces (21-by-12 pixels) processed from 1,755 images from the FERET face database. The performance of SVMs (3.4% error) is shown to be superior to traditional pattern classifiers (Linear, Quadratic, Fisher Linear Discriminant, Nearest-Neighbor) as well as more modern techniques such as Radial Basis Function (RBF) classifiers and large ensemble-RBF networks. Surprisingly, SVMs also out-performed human test subjects at the same task: in an experimental study involving 30 human test subjects ranging in age from mid-20s to mid-40s, the average error rate was 32% for the same "thumbnails" and 6.7% with high-resolution images (still nearly twice the error rate of SVMs). The difference between low and high-resolution inputs with SVMs was only 1% thus demonstrating a degree of robustness and relative scale invariance.

Original languageEnglish
Pages (from-to)1115-1118
Number of pages4
JournalProceedings - International Conference on Pattern Recognition
Volume15
Issue number1
Publication statusPublished - 2000 Dec 1

Fingerprint

Support vector machines
Classifiers
Radial basis function networks
Image resolution
Invariance
Robustness (control systems)
Pixels

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

@article{c8fcfcc550b34e34a81a53afbee73794,
title = "Support Vector Machines for visual gender classification",
abstract = "Support Vector Machines (SVMs) are investigated for visual gender classification with low-resolution {"}thumbnail{"} faces (21-by-12 pixels) processed from 1,755 images from the FERET face database. The performance of SVMs (3.4{\%} error) is shown to be superior to traditional pattern classifiers (Linear, Quadratic, Fisher Linear Discriminant, Nearest-Neighbor) as well as more modern techniques such as Radial Basis Function (RBF) classifiers and large ensemble-RBF networks. Surprisingly, SVMs also out-performed human test subjects at the same task: in an experimental study involving 30 human test subjects ranging in age from mid-20s to mid-40s, the average error rate was 32{\%} for the same {"}thumbnails{"} and 6.7{\%} with high-resolution images (still nearly twice the error rate of SVMs). The difference between low and high-resolution inputs with SVMs was only 1{\%} thus demonstrating a degree of robustness and relative scale invariance.",
author = "Yang, {Ming Hsuan} and Baback Moghaddam",
year = "2000",
month = "12",
day = "1",
language = "English",
volume = "15",
pages = "1115--1118",
journal = "Proceedings - International Conference on Pattern Recognition",
issn = "1051-4651",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

Support Vector Machines for visual gender classification. / Yang, Ming Hsuan; Moghaddam, Baback.

In: Proceedings - International Conference on Pattern Recognition, Vol. 15, No. 1, 01.12.2000, p. 1115-1118.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Support Vector Machines for visual gender classification

AU - Yang, Ming Hsuan

AU - Moghaddam, Baback

PY - 2000/12/1

Y1 - 2000/12/1

N2 - Support Vector Machines (SVMs) are investigated for visual gender classification with low-resolution "thumbnail" faces (21-by-12 pixels) processed from 1,755 images from the FERET face database. The performance of SVMs (3.4% error) is shown to be superior to traditional pattern classifiers (Linear, Quadratic, Fisher Linear Discriminant, Nearest-Neighbor) as well as more modern techniques such as Radial Basis Function (RBF) classifiers and large ensemble-RBF networks. Surprisingly, SVMs also out-performed human test subjects at the same task: in an experimental study involving 30 human test subjects ranging in age from mid-20s to mid-40s, the average error rate was 32% for the same "thumbnails" and 6.7% with high-resolution images (still nearly twice the error rate of SVMs). The difference between low and high-resolution inputs with SVMs was only 1% thus demonstrating a degree of robustness and relative scale invariance.

AB - Support Vector Machines (SVMs) are investigated for visual gender classification with low-resolution "thumbnail" faces (21-by-12 pixels) processed from 1,755 images from the FERET face database. The performance of SVMs (3.4% error) is shown to be superior to traditional pattern classifiers (Linear, Quadratic, Fisher Linear Discriminant, Nearest-Neighbor) as well as more modern techniques such as Radial Basis Function (RBF) classifiers and large ensemble-RBF networks. Surprisingly, SVMs also out-performed human test subjects at the same task: in an experimental study involving 30 human test subjects ranging in age from mid-20s to mid-40s, the average error rate was 32% for the same "thumbnails" and 6.7% with high-resolution images (still nearly twice the error rate of SVMs). The difference between low and high-resolution inputs with SVMs was only 1% thus demonstrating a degree of robustness and relative scale invariance.

UR - http://www.scopus.com/inward/record.url?scp=10044238343&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=10044238343&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:10044238343

VL - 15

SP - 1115

EP - 1118

JO - Proceedings - International Conference on Pattern Recognition

JF - Proceedings - International Conference on Pattern Recognition

SN - 1051-4651

IS - 1

ER -