100% Accuracy in Automatic Face Recognition

R. Jenkins, A. M. Burton

Research output: Contribution to journalArticle

98 Citations (Scopus)

Abstract

Accurate face recognition is critical for many security applications. Current automatic face-recognition systems are defeated by natural changes in lighting and pose, which often affect face images more profoundly than changes in identity. The only system that can reliably cope with such variability is a human observer who is familiar with the faces concerned. We modeled human familiarity by using image averaging to derive stable face representations from naturally varying photographs. This simple procedure increased the accuracy of an industry standard face-recognition algorithm from 54% to 100%, bringing the robust performance of a familiar human to an automated system.
Original languageEnglish
Pages (from-to)435
Number of pages1
JournalScience
Volume319
Issue number5862
DOIs
Publication statusPublished - 25 Jan 2008

Cite this

100% Accuracy in Automatic Face Recognition. / Jenkins, R.; Burton, A. M.

In: Science, Vol. 319, No. 5862, 25.01.2008, p. 435.

Research output: Contribution to journalArticle

Jenkins, R & Burton, AM 2008, '100% Accuracy in Automatic Face Recognition', Science, vol. 319, no. 5862, pp. 435. https://doi.org/10.1126/science.1149656
Jenkins, R. ; Burton, A. M. / 100% Accuracy in Automatic Face Recognition. In: Science. 2008 ; Vol. 319, No. 5862. pp. 435.
@article{652d8e0d3df84a96bee85bdd05281e4c,
title = "100{\%} Accuracy in Automatic Face Recognition",
abstract = "Accurate face recognition is critical for many security applications. Current automatic face-recognition systems are defeated by natural changes in lighting and pose, which often affect face images more profoundly than changes in identity. The only system that can reliably cope with such variability is a human observer who is familiar with the faces concerned. We modeled human familiarity by using image averaging to derive stable face representations from naturally varying photographs. This simple procedure increased the accuracy of an industry standard face-recognition algorithm from 54{\%} to 100{\%}, bringing the robust performance of a familiar human to an automated system.",
author = "R. Jenkins and Burton, {A. M.}",
year = "2008",
month = "1",
day = "25",
doi = "10.1126/science.1149656",
language = "English",
volume = "319",
pages = "435",
journal = "Science",
issn = "0036-8075",
publisher = "AMER ASSOC ADVANCEMENT SCIENCE",
number = "5862",

}

TY - JOUR

T1 - 100% Accuracy in Automatic Face Recognition

AU - Jenkins, R.

AU - Burton, A. M.

PY - 2008/1/25

Y1 - 2008/1/25

N2 - Accurate face recognition is critical for many security applications. Current automatic face-recognition systems are defeated by natural changes in lighting and pose, which often affect face images more profoundly than changes in identity. The only system that can reliably cope with such variability is a human observer who is familiar with the faces concerned. We modeled human familiarity by using image averaging to derive stable face representations from naturally varying photographs. This simple procedure increased the accuracy of an industry standard face-recognition algorithm from 54% to 100%, bringing the robust performance of a familiar human to an automated system.

AB - Accurate face recognition is critical for many security applications. Current automatic face-recognition systems are defeated by natural changes in lighting and pose, which often affect face images more profoundly than changes in identity. The only system that can reliably cope with such variability is a human observer who is familiar with the faces concerned. We modeled human familiarity by using image averaging to derive stable face representations from naturally varying photographs. This simple procedure increased the accuracy of an industry standard face-recognition algorithm from 54% to 100%, bringing the robust performance of a familiar human to an automated system.

U2 - 10.1126/science.1149656

DO - 10.1126/science.1149656

M3 - Article

VL - 319

SP - 435

JO - Science

JF - Science

SN - 0036-8075

IS - 5862

ER -