100% Accuracy in Automatic Face Recognition

by: R Jenkins, AM Burton

Science, Vol. 319, No. 5862. (25 January 2008), 435.

View FullText article

DOI, HighWire, Pubmed, Hubmed

Reviews [Write a review of this article]

There are no reviews of this article

Find related articles from these CiteULike users

cmunson, shivakmr, ariels, VisionLab (group), yuhc, evansd66

Find related articles with these CiteULike tags

algorithm, averaging, face, face_recognition, faces, recognition, sciencemag, wow

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. 10.1126/science.1149656

@article{citeulike:2289775, 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. 10.1126/science.1149656}, author = {Jenkins, R. and Burton, A. M. }, citeulike-article-id = {2289775}, doi = {http://dx.doi.org/10.1126/science.1149656}, journal = {Science}, keywords = {algorithm}, month = {January}, number = {5862}, pages = {435+}, posted-at = {2008-01-29 07:36:02}, priority = {2}, title = {100\% Accuracy in Automatic Face Recognition}, url = {http://dx.doi.org/10.1126/science.1149656}, volume = {319}, year = {2008} }

RIS record

TY - JOUR ID - citeulike:2289775 L3 - citeulike-article-id:2289775 TI - 100% Accuracy in Automatic Face Recognition JF - Science VL - 319 IS - 5862 SP - 435 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. 10.1126/science.1149656 KW - algorithm AU - Jenkins, R AU - Burton, AM PY - 2008/01/25/ UR - http://dx.doi.org/10.1126/science.1149656 ER -

RIS BibTeX