Learning gender with support faces

Baback Moghaddam, Ming Hsuan Yang

Research output: Contribution to journalArticle

453 Citations (Scopus)

Abstract

Nonlinear Support Vector Machines (SVMs) are investigated for appearance-based gender classification with low-resolution "thumbnail" faces processed from 1,755 images from the FERET face database. The performance of SVMs (3.4 percent 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. Furthermore, the difference in classification performance with low-resolution "thumbnails" (21-by-12 pixels) and the corresponding higher resolution images (84-by-48 pixels) was found to be only 1 percent, thus demonstrating robustness and stability with respect to scale and degree of facial detail.

Original languageEnglish
Pages (from-to)707-711
Number of pages5
JournalIEEE transactions on pattern analysis and machine intelligence
Volume24
Issue number5
DOIs
Publication statusPublished - 2002 May 1

Fingerprint

Percent
Support vector machines
Support Vector Machine
Classifiers
Pixel
Pixels
Classifier
Face
Radial basis function networks
Radial Basis Function Network
Image resolution
Radial Functions
Discriminant
Basis Functions
Nearest Neighbor
Ensemble
High Resolution
Robustness
Gender
Learning

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Cite this

@article{2e0d34a21487437aa1230a98462a7f9b,
title = "Learning gender with support faces",
abstract = "Nonlinear Support Vector Machines (SVMs) are investigated for appearance-based gender classification with low-resolution {"}thumbnail{"} faces processed from 1,755 images from the FERET face database. The performance of SVMs (3.4 percent 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. Furthermore, the difference in classification performance with low-resolution {"}thumbnails{"} (21-by-12 pixels) and the corresponding higher resolution images (84-by-48 pixels) was found to be only 1 percent, thus demonstrating robustness and stability with respect to scale and degree of facial detail.",
author = "Baback Moghaddam and Yang, {Ming Hsuan}",
year = "2002",
month = "5",
day = "1",
doi = "10.1109/34.1000244",
language = "English",
volume = "24",
pages = "707--711",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "5",

}

Learning gender with support faces. / Moghaddam, Baback; Yang, Ming Hsuan.

In: IEEE transactions on pattern analysis and machine intelligence, Vol. 24, No. 5, 01.05.2002, p. 707-711.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Learning gender with support faces

AU - Moghaddam, Baback

AU - Yang, Ming Hsuan

PY - 2002/5/1

Y1 - 2002/5/1

N2 - Nonlinear Support Vector Machines (SVMs) are investigated for appearance-based gender classification with low-resolution "thumbnail" faces processed from 1,755 images from the FERET face database. The performance of SVMs (3.4 percent 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. Furthermore, the difference in classification performance with low-resolution "thumbnails" (21-by-12 pixels) and the corresponding higher resolution images (84-by-48 pixels) was found to be only 1 percent, thus demonstrating robustness and stability with respect to scale and degree of facial detail.

AB - Nonlinear Support Vector Machines (SVMs) are investigated for appearance-based gender classification with low-resolution "thumbnail" faces processed from 1,755 images from the FERET face database. The performance of SVMs (3.4 percent 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. Furthermore, the difference in classification performance with low-resolution "thumbnails" (21-by-12 pixels) and the corresponding higher resolution images (84-by-48 pixels) was found to be only 1 percent, thus demonstrating robustness and stability with respect to scale and degree of facial detail.

UR - http://www.scopus.com/inward/record.url?scp=0036565806&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036565806&partnerID=8YFLogxK

U2 - 10.1109/34.1000244

DO - 10.1109/34.1000244

M3 - Article

AN - SCOPUS:0036565806

VL - 24

SP - 707

EP - 711

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 5

ER -