Learning from humans: Computational modeling of face recognition

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

In this paper, we propose a computational architecture of face recognition based on evidence from cognitive research. Several recent psychophysical experiments have shown that humans process faces by a combination of configural and component information. Using an appearance-based implementation of this architecture based on low-level features and their spatial relations, we were able to model aspects of human performance found in psychophysical studies. Furthermore, results from additional computational recognition experiments show that our framework is able to achieve excellent recognition performance even under large view rotations. Our interdisciplinary study is an example of how results from cognitive research can be used to construct recognition systems with increased performance. Finally, our modeling results also make new experimental predictions that will be tested in further psychophysical studies, thus effectively closing the loop between psychophysical experimentation and computational modeling.

Original languageEnglish
Pages (from-to)401-418
Number of pages18
JournalNetwork: Computation in Neural Systems
Volume16
Issue number4
DOIs
Publication statusPublished - 2005 Dec 1
Externally publishedYes

Fingerprint

Learning
Interdisciplinary Studies
Research
Facial Recognition
Recognition (Psychology)

Keywords

  • Configural and component information
  • Face recognition
  • Local features

ASJC Scopus subject areas

  • Neuroscience(all)

Cite this

Learning from humans : Computational modeling of face recognition. / Wallraven, Christian; Schwaninger, Adrian; Bulthoff, Heinrich.

In: Network: Computation in Neural Systems, Vol. 16, No. 4, 01.12.2005, p. 401-418.

Research output: Contribution to journalArticle

@article{4620cf2071084d619af6580dbecab020,
title = "Learning from humans: Computational modeling of face recognition",
abstract = "In this paper, we propose a computational architecture of face recognition based on evidence from cognitive research. Several recent psychophysical experiments have shown that humans process faces by a combination of configural and component information. Using an appearance-based implementation of this architecture based on low-level features and their spatial relations, we were able to model aspects of human performance found in psychophysical studies. Furthermore, results from additional computational recognition experiments show that our framework is able to achieve excellent recognition performance even under large view rotations. Our interdisciplinary study is an example of how results from cognitive research can be used to construct recognition systems with increased performance. Finally, our modeling results also make new experimental predictions that will be tested in further psychophysical studies, thus effectively closing the loop between psychophysical experimentation and computational modeling.",
keywords = "Configural and component information, Face recognition, Local features",
author = "Christian Wallraven and Adrian Schwaninger and Heinrich Bulthoff",
year = "2005",
month = "12",
day = "1",
doi = "10.1080/09548980500508844",
language = "English",
volume = "16",
pages = "401--418",
journal = "Network: Computation in Neural Systems",
issn = "0954-898X",
publisher = "Informa Healthcare",
number = "4",

}

TY - JOUR

T1 - Learning from humans

T2 - Computational modeling of face recognition

AU - Wallraven, Christian

AU - Schwaninger, Adrian

AU - Bulthoff, Heinrich

PY - 2005/12/1

Y1 - 2005/12/1

N2 - In this paper, we propose a computational architecture of face recognition based on evidence from cognitive research. Several recent psychophysical experiments have shown that humans process faces by a combination of configural and component information. Using an appearance-based implementation of this architecture based on low-level features and their spatial relations, we were able to model aspects of human performance found in psychophysical studies. Furthermore, results from additional computational recognition experiments show that our framework is able to achieve excellent recognition performance even under large view rotations. Our interdisciplinary study is an example of how results from cognitive research can be used to construct recognition systems with increased performance. Finally, our modeling results also make new experimental predictions that will be tested in further psychophysical studies, thus effectively closing the loop between psychophysical experimentation and computational modeling.

AB - In this paper, we propose a computational architecture of face recognition based on evidence from cognitive research. Several recent psychophysical experiments have shown that humans process faces by a combination of configural and component information. Using an appearance-based implementation of this architecture based on low-level features and their spatial relations, we were able to model aspects of human performance found in psychophysical studies. Furthermore, results from additional computational recognition experiments show that our framework is able to achieve excellent recognition performance even under large view rotations. Our interdisciplinary study is an example of how results from cognitive research can be used to construct recognition systems with increased performance. Finally, our modeling results also make new experimental predictions that will be tested in further psychophysical studies, thus effectively closing the loop between psychophysical experimentation and computational modeling.

KW - Configural and component information

KW - Face recognition

KW - Local features

UR - http://www.scopus.com/inward/record.url?scp=33646248423&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646248423&partnerID=8YFLogxK

U2 - 10.1080/09548980500508844

DO - 10.1080/09548980500508844

M3 - Article

C2 - 16611592

AN - SCOPUS:33646248423

VL - 16

SP - 401

EP - 418

JO - Network: Computation in Neural Systems

JF - Network: Computation in Neural Systems

SN - 0954-898X

IS - 4

ER -