Face averages enhance user recognition for smartphone security

David James Robertson; Robin Stewart Samuel Kramer; A. Mike Burton

doi:10.1371/journal.pone.0119460

Face averages enhance user recognition for smartphone security

David James Robertson, Robin Stewart Samuel Kramer, A. Mike Burton

Psychology

Research output: Contribution to journal › Article › peer-review

Abstract

Our recognition of familiar faces is excellent, and generalises across viewing conditions. However, unfamiliar face recognition is much poorer. For this reason, automatic face recog- nition systems might benefit from incorporating the advantages of familiarity. Here we put this to the test using the face verification system available on a popular smartphone (the Samsung Galaxy). In two experiments we tested the recognition performance of the smart- phone when it was encoded with an individual’s ‘face-average’– a representation derived from theories of human face perception. This technique significantly improved performance for both unconstrained celebrity images (Experiment 1) and for real faces (Experiment 2): users could unlock their phones more reliably when the device stored an average of the user’s face than when they stored a single image. This advantage was consistent across a wide variety of everyday viewing conditions. Furthermore, the benefit did not reduce the re- jection of imposter faces. This benefit is brought about solely by consideration of suitable representations for automatic face recognition, and we argue that this is just as important as development of matching algorithms themselves. Wepropose that this representation could significantly improve recognition rates in everyday settings.
Introduction

Original language	English
Article number	e0119460
Number of pages	11
Journal	PLoS ONE
Volume	10
Issue number	3
DOIs	https://doi.org/10.1371/journal.pone.0119460
Publication status	Published - 25 Mar 2015

Bibliographical note

© [2015, Robertson et al. This content is made available by the publisher under a Creative Commons Attribution Licence. This means that a user may copy, distribute and display the resource providing that they give credit. Users must adhere to the terms of the licence.

Access to Document

10.1371/journal.pone.0119460

journal.pone.0119460
© [2015, Robertson et al. This content is made available by the publisher under a Creative Commons Attribution Licence. This means that a user may copy, distribute and display the resource providing that they give credit. Users must adhere to the terms of the licence.
Final published version, 460 KB

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0119460

Cite this

@article{ccd6f4c773b14d1ba170bbaeb4ee5892,

title = "Face averages enhance user recognition for smartphone security",

abstract = "Our recognition of familiar faces is excellent, and generalises across viewing conditions. However, unfamiliar face recognition is much poorer. For this reason, automatic face recog- nition systems might benefit from incorporating the advantages of familiarity. Here we put this to the test using the face verification system available on a popular smartphone (the Samsung Galaxy). In two experiments we tested the recognition performance of the smart- phone when it was encoded with an individual{\textquoteright}s {\textquoteleft}face-average{\textquoteright}– a representation derived from theories of human face perception. This technique significantly improved performance for both unconstrained celebrity images (Experiment 1) and for real faces (Experiment 2): users could unlock their phones more reliably when the device stored an average of the user{\textquoteright}s face than when they stored a single image. This advantage was consistent across a wide variety of everyday viewing conditions. Furthermore, the benefit did not reduce the re- jection of imposter faces. This benefit is brought about solely by consideration of suitable representations for automatic face recognition, and we argue that this is just as important as development of matching algorithms themselves. Wepropose that this representation could significantly improve recognition rates in everyday settings.Introduction",

author = "Robertson, {David James} and Kramer, {Robin Stewart Samuel} and Burton, {A. Mike}",

note = "{\textcopyright} [2015, Robertson et al. This content is made available by the publisher under a Creative Commons Attribution Licence. This means that a user may copy, distribute and display the resource providing that they give credit. Users must adhere to the terms of the licence.",

year = "2015",

month = mar,

day = "25",

doi = "10.1371/journal.pone.0119460",

language = "English",

volume = "10",

journal = "PLoS ONE",

issn = "1932-6203",

publisher = "Public Library of Science",

number = "3",

}

TY - JOUR

T1 - Face averages enhance user recognition for smartphone security

AU - Robertson, David James

AU - Kramer, Robin Stewart Samuel

AU - Burton, A. Mike

N1 - © [2015, Robertson et al. This content is made available by the publisher under a Creative Commons Attribution Licence. This means that a user may copy, distribute and display the resource providing that they give credit. Users must adhere to the terms of the licence.

PY - 2015/3/25

Y1 - 2015/3/25

N2 - Our recognition of familiar faces is excellent, and generalises across viewing conditions. However, unfamiliar face recognition is much poorer. For this reason, automatic face recog- nition systems might benefit from incorporating the advantages of familiarity. Here we put this to the test using the face verification system available on a popular smartphone (the Samsung Galaxy). In two experiments we tested the recognition performance of the smart- phone when it was encoded with an individual’s ‘face-average’– a representation derived from theories of human face perception. This technique significantly improved performance for both unconstrained celebrity images (Experiment 1) and for real faces (Experiment 2): users could unlock their phones more reliably when the device stored an average of the user’s face than when they stored a single image. This advantage was consistent across a wide variety of everyday viewing conditions. Furthermore, the benefit did not reduce the re- jection of imposter faces. This benefit is brought about solely by consideration of suitable representations for automatic face recognition, and we argue that this is just as important as development of matching algorithms themselves. Wepropose that this representation could significantly improve recognition rates in everyday settings.Introduction

AB - Our recognition of familiar faces is excellent, and generalises across viewing conditions. However, unfamiliar face recognition is much poorer. For this reason, automatic face recog- nition systems might benefit from incorporating the advantages of familiarity. Here we put this to the test using the face verification system available on a popular smartphone (the Samsung Galaxy). In two experiments we tested the recognition performance of the smart- phone when it was encoded with an individual’s ‘face-average’– a representation derived from theories of human face perception. This technique significantly improved performance for both unconstrained celebrity images (Experiment 1) and for real faces (Experiment 2): users could unlock their phones more reliably when the device stored an average of the user’s face than when they stored a single image. This advantage was consistent across a wide variety of everyday viewing conditions. Furthermore, the benefit did not reduce the re- jection of imposter faces. This benefit is brought about solely by consideration of suitable representations for automatic face recognition, and we argue that this is just as important as development of matching algorithms themselves. Wepropose that this representation could significantly improve recognition rates in everyday settings.Introduction

U2 - 10.1371/journal.pone.0119460

DO - 10.1371/journal.pone.0119460

M3 - Article

SN - 1932-6203

VL - 10

JO - PLoS ONE

JF - PLoS ONE

IS - 3

M1 - e0119460

ER -