Face recognition: A convolutional neural-network approach

Steve Lawrence, C. Lee Giles, Ah Chung Tsoi, Andrew D. Back

Research output: Contribution to journalArticle

1195 Citations (Scopus)

Abstract

Faces represent complex multidimensional meaningful visual stimuli and developing a computational model for face recognition is difficult. We present a hybrid neural-network solution which compares favorably with other methods. The system combines local image sampling, a self-organizing map (SOM) neural network, and a convolutional neural network. The SOM provides a quantization of the image samples into a topological space where inputs that are nearby in the original space are also nearby in the output space, thereby providing dimensionality reduction and invariance to minor changes in the image sample, and the convolutional neural network provides for partial invariance to translation, rotation, scale, and deformation. The convolutional network extracts successively larger features in a hierarchical set of layers. We present results using the Karhunen-Loève (KL) transform in place of the SOM, and a multilayer perceptron (MLP) in place of the convolutional network. The KL transform performs almost as well (5.3% error versus 3.8%). The MLP performs very poorly (40% error versus 3.8%). The method is capable of rapid classification, requires only fast approximate normalization and preprocessing, and consistently exhibits better classification performance than the eigenfaces approach on the database considered as the number of images per person in the training database is varied from one to five. With five images per person the proposed method and eigenfaces result in 3.8% and 10.5% error, respectively. The recognizer provides a measure of confidence in its output and classification error approaches zero when rejecting as few as 10% of the examples. We use a database of 400 images of 40 individuals which contains quite a high degree of variability in expression, pose, and facial details. We analyze computational complexity and discuss how new classes could be added to the trained recognizer.

Original languageEnglish (US)
Pages (from-to)98-113
Number of pages16
JournalIEEE Transactions on Neural Networks
Volume8
Issue number1
DOIs
StatePublished - Jan 1 1997

Fingerprint

Face recognition
Self organizing maps
Neural networks
Multilayer neural networks
Invariance
Image sampling
Computational complexity

All Science Journal Classification (ASJC) codes

  • Software
  • Computer Science Applications
  • Computer Networks and Communications
  • Artificial Intelligence

Cite this

Lawrence, Steve ; Giles, C. Lee ; Tsoi, Ah Chung ; Back, Andrew D. / Face recognition : A convolutional neural-network approach. In: IEEE Transactions on Neural Networks. 1997 ; Vol. 8, No. 1. pp. 98-113.
@article{da03d85810254455902d00974aae5f9a,
title = "Face recognition: A convolutional neural-network approach",
abstract = "Faces represent complex multidimensional meaningful visual stimuli and developing a computational model for face recognition is difficult. We present a hybrid neural-network solution which compares favorably with other methods. The system combines local image sampling, a self-organizing map (SOM) neural network, and a convolutional neural network. The SOM provides a quantization of the image samples into a topological space where inputs that are nearby in the original space are also nearby in the output space, thereby providing dimensionality reduction and invariance to minor changes in the image sample, and the convolutional neural network provides for partial invariance to translation, rotation, scale, and deformation. The convolutional network extracts successively larger features in a hierarchical set of layers. We present results using the Karhunen-Lo{\`e}ve (KL) transform in place of the SOM, and a multilayer perceptron (MLP) in place of the convolutional network. The KL transform performs almost as well (5.3{\%} error versus 3.8{\%}). The MLP performs very poorly (40{\%} error versus 3.8{\%}). The method is capable of rapid classification, requires only fast approximate normalization and preprocessing, and consistently exhibits better classification performance than the eigenfaces approach on the database considered as the number of images per person in the training database is varied from one to five. With five images per person the proposed method and eigenfaces result in 3.8{\%} and 10.5{\%} error, respectively. The recognizer provides a measure of confidence in its output and classification error approaches zero when rejecting as few as 10{\%} of the examples. We use a database of 400 images of 40 individuals which contains quite a high degree of variability in expression, pose, and facial details. We analyze computational complexity and discuss how new classes could be added to the trained recognizer.",
author = "Steve Lawrence and Giles, {C. Lee} and Tsoi, {Ah Chung} and Back, {Andrew D.}",
year = "1997",
month = "1",
day = "1",
doi = "10.1109/72.554195",
language = "English (US)",
volume = "8",
pages = "98--113",
journal = "IEEE Transactions on Neural Networks and Learning Systems",
issn = "2162-237X",
publisher = "IEEE Computational Intelligence Society",
number = "1",

}

Face recognition : A convolutional neural-network approach. / Lawrence, Steve; Giles, C. Lee; Tsoi, Ah Chung; Back, Andrew D.

In: IEEE Transactions on Neural Networks, Vol. 8, No. 1, 01.01.1997, p. 98-113.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Face recognition

T2 - A convolutional neural-network approach

AU - Lawrence, Steve

AU - Giles, C. Lee

AU - Tsoi, Ah Chung

AU - Back, Andrew D.

PY - 1997/1/1

Y1 - 1997/1/1

N2 - Faces represent complex multidimensional meaningful visual stimuli and developing a computational model for face recognition is difficult. We present a hybrid neural-network solution which compares favorably with other methods. The system combines local image sampling, a self-organizing map (SOM) neural network, and a convolutional neural network. The SOM provides a quantization of the image samples into a topological space where inputs that are nearby in the original space are also nearby in the output space, thereby providing dimensionality reduction and invariance to minor changes in the image sample, and the convolutional neural network provides for partial invariance to translation, rotation, scale, and deformation. The convolutional network extracts successively larger features in a hierarchical set of layers. We present results using the Karhunen-Loève (KL) transform in place of the SOM, and a multilayer perceptron (MLP) in place of the convolutional network. The KL transform performs almost as well (5.3% error versus 3.8%). The MLP performs very poorly (40% error versus 3.8%). The method is capable of rapid classification, requires only fast approximate normalization and preprocessing, and consistently exhibits better classification performance than the eigenfaces approach on the database considered as the number of images per person in the training database is varied from one to five. With five images per person the proposed method and eigenfaces result in 3.8% and 10.5% error, respectively. The recognizer provides a measure of confidence in its output and classification error approaches zero when rejecting as few as 10% of the examples. We use a database of 400 images of 40 individuals which contains quite a high degree of variability in expression, pose, and facial details. We analyze computational complexity and discuss how new classes could be added to the trained recognizer.

AB - Faces represent complex multidimensional meaningful visual stimuli and developing a computational model for face recognition is difficult. We present a hybrid neural-network solution which compares favorably with other methods. The system combines local image sampling, a self-organizing map (SOM) neural network, and a convolutional neural network. The SOM provides a quantization of the image samples into a topological space where inputs that are nearby in the original space are also nearby in the output space, thereby providing dimensionality reduction and invariance to minor changes in the image sample, and the convolutional neural network provides for partial invariance to translation, rotation, scale, and deformation. The convolutional network extracts successively larger features in a hierarchical set of layers. We present results using the Karhunen-Loève (KL) transform in place of the SOM, and a multilayer perceptron (MLP) in place of the convolutional network. The KL transform performs almost as well (5.3% error versus 3.8%). The MLP performs very poorly (40% error versus 3.8%). The method is capable of rapid classification, requires only fast approximate normalization and preprocessing, and consistently exhibits better classification performance than the eigenfaces approach on the database considered as the number of images per person in the training database is varied from one to five. With five images per person the proposed method and eigenfaces result in 3.8% and 10.5% error, respectively. The recognizer provides a measure of confidence in its output and classification error approaches zero when rejecting as few as 10% of the examples. We use a database of 400 images of 40 individuals which contains quite a high degree of variability in expression, pose, and facial details. We analyze computational complexity and discuss how new classes could be added to the trained recognizer.

UR - http://www.scopus.com/inward/record.url?scp=0030737097&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030737097&partnerID=8YFLogxK

U2 - 10.1109/72.554195

DO - 10.1109/72.554195

M3 - Article

C2 - 18255614

AN - SCOPUS:0030737097

VL - 8

SP - 98

EP - 113

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 1

ER -