Computational modeling of face recognition based on psychophysical experiments

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Recent results from psychophysical studies are discussed which clearly show that face processing is not only holistic. Humans do encode face parts (component information) in addition to information about the spatial interrelationship of facial features (global configural information). Based on these findings we propose a computational architecture of face recognition, which implements a component and configural route for encoding and recognizing faces. Modeling results showed a striking similarity between human psychophysical data and the computational model. In addition, we could show that our framework is able to achieve good recognition performance even under large view rotations. Thus, our study is an example of how an interdisciplinary approach can provide a deeper understanding of cognitive processes and lead to further insights in human psychophysics as well as computer vision.

Original languageEnglish
Pages (from-to)207-215
Number of pages9
JournalSwiss Journal of Psychology
Volume63
Issue number3
DOIs
Publication statusPublished - 2004 Sep 1
Externally publishedYes

Fingerprint

Psychophysics
Facial Recognition

Keywords

  • Component and configural processing
  • Computational modeling
  • Face recognition

ASJC Scopus subject areas

  • Psychology(all)

Cite this

Computational modeling of face recognition based on psychophysical experiments. / Schwaninger, Adrian; Wallraven, Christian; Bulthoff, Heinrich.

In: Swiss Journal of Psychology, Vol. 63, No. 3, 01.09.2004, p. 207-215.

Research output: Contribution to journalArticle

@article{4e0f2e8390c44c13bb956e8e37ba2fae,
title = "Computational modeling of face recognition based on psychophysical experiments",
abstract = "Recent results from psychophysical studies are discussed which clearly show that face processing is not only holistic. Humans do encode face parts (component information) in addition to information about the spatial interrelationship of facial features (global configural information). Based on these findings we propose a computational architecture of face recognition, which implements a component and configural route for encoding and recognizing faces. Modeling results showed a striking similarity between human psychophysical data and the computational model. In addition, we could show that our framework is able to achieve good recognition performance even under large view rotations. Thus, our study is an example of how an interdisciplinary approach can provide a deeper understanding of cognitive processes and lead to further insights in human psychophysics as well as computer vision.",
keywords = "Component and configural processing, Computational modeling, Face recognition",
author = "Adrian Schwaninger and Christian Wallraven and Heinrich Bulthoff",
year = "2004",
month = "9",
day = "1",
doi = "10.1024/1421-0185.63.3.207",
language = "English",
volume = "63",
pages = "207--215",
journal = "Swiss Journal of Psychology",
issn = "1421-0185",
publisher = "Verlag Hans Huber",
number = "3",

}

TY - JOUR

T1 - Computational modeling of face recognition based on psychophysical experiments

AU - Schwaninger, Adrian

AU - Wallraven, Christian

AU - Bulthoff, Heinrich

PY - 2004/9/1

Y1 - 2004/9/1

N2 - Recent results from psychophysical studies are discussed which clearly show that face processing is not only holistic. Humans do encode face parts (component information) in addition to information about the spatial interrelationship of facial features (global configural information). Based on these findings we propose a computational architecture of face recognition, which implements a component and configural route for encoding and recognizing faces. Modeling results showed a striking similarity between human psychophysical data and the computational model. In addition, we could show that our framework is able to achieve good recognition performance even under large view rotations. Thus, our study is an example of how an interdisciplinary approach can provide a deeper understanding of cognitive processes and lead to further insights in human psychophysics as well as computer vision.

AB - Recent results from psychophysical studies are discussed which clearly show that face processing is not only holistic. Humans do encode face parts (component information) in addition to information about the spatial interrelationship of facial features (global configural information). Based on these findings we propose a computational architecture of face recognition, which implements a component and configural route for encoding and recognizing faces. Modeling results showed a striking similarity between human psychophysical data and the computational model. In addition, we could show that our framework is able to achieve good recognition performance even under large view rotations. Thus, our study is an example of how an interdisciplinary approach can provide a deeper understanding of cognitive processes and lead to further insights in human psychophysics as well as computer vision.

KW - Component and configural processing

KW - Computational modeling

KW - Face recognition

UR - http://www.scopus.com/inward/record.url?scp=8744231038&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=8744231038&partnerID=8YFLogxK

U2 - 10.1024/1421-0185.63.3.207

DO - 10.1024/1421-0185.63.3.207

M3 - Article

AN - SCOPUS:8744231038

VL - 63

SP - 207

EP - 215

JO - Swiss Journal of Psychology

JF - Swiss Journal of Psychology

SN - 1421-0185

IS - 3

ER -