Stereo integration, mean field theory and psychophysics

A. L. Yuille; D. Geiger; H. Bülthoff

doi:10.1007/BFb0014852

Stereo integration, mean field theory and psychophysics

A. L. Yuille, D. Geiger, H. Bülthoff

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We describe a theoretical formulation for stereo in terms of the Markov Random Field and Bayesian approach to vision. This formulation enables us to integrate the depth information from different types of matching primitives, or from different vision modules. We treat the correspondence problem and surface interpolation as different aspects of the same problem and solve them simultaneously, unlike most previous theories. We use techniques from statistical physics to compute properties of our theory and show how it relates to previous work. These techniques also suggest novel algorithms for stereo which are argued to be preferable to standard algorithms on theoretical and experimental grounds. It can be shown (Yuille, Geiger and Bülthoff 1989) that the theory is consistent with some psychophysical experiments which investigate the relative importance of different matching primitives.

Original language	English (US)
Title of host publication	Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings
Editors	Olivier Faugeras
Publisher	Springer Verlag
Pages	73-82
Number of pages	10
ISBN (Print)	9783540525226
DOIs	https://doi.org/10.1007/BFb0014852
State	Published - 1990
Event	1st European Conference on Computer Vision, ECCV 1990 - Antibes, France Duration: Apr 23 1990 → Apr 27 1990

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	427 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	1st European Conference on Computer Vision, ECCV 1990
Country/Territory	France
City	Antibes
Period	4/23/90 → 4/27/90

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/BFb0014852

Cite this

Yuille, A. L., Geiger, D., & Bülthoff, H. (1990). Stereo integration, mean field theory and psychophysics. In O. Faugeras (Ed.), Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings (pp. 73-82). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 427 LNCS). Springer Verlag. https://doi.org/10.1007/BFb0014852

Stereo integration, mean field theory and psychophysics. / Yuille, A. L.; Geiger, D.; Bülthoff, H.
Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings. ed. / Olivier Faugeras. Springer Verlag, 1990. p. 73-82 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 427 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Yuille, AL, Geiger, D & Bülthoff, H 1990, Stereo integration, mean field theory and psychophysics. in O Faugeras (ed.), Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 427 LNCS, Springer Verlag, pp. 73-82, 1st European Conference on Computer Vision, ECCV 1990, Antibes, France, 4/23/90. https://doi.org/10.1007/BFb0014852

Yuille AL, Geiger D, Bülthoff H. Stereo integration, mean field theory and psychophysics. In Faugeras O, editor, Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings. Springer Verlag. 1990. p. 73-82. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/BFb0014852

Yuille, A. L. ; Geiger, D. ; Bülthoff, H. / Stereo integration, mean field theory and psychophysics. Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings. editor / Olivier Faugeras. Springer Verlag, 1990. pp. 73-82 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{c1acdc8fa4c94869b717193f73c9de4a,

title = "Stereo integration, mean field theory and psychophysics",

abstract = "We describe a theoretical formulation for stereo in terms of the Markov Random Field and Bayesian approach to vision. This formulation enables us to integrate the depth information from different types of matching primitives, or from different vision modules. We treat the correspondence problem and surface interpolation as different aspects of the same problem and solve them simultaneously, unlike most previous theories. We use techniques from statistical physics to compute properties of our theory and show how it relates to previous work. These techniques also suggest novel algorithms for stereo which are argued to be preferable to standard algorithms on theoretical and experimental grounds. It can be shown (Yuille, Geiger and B{\"u}lthoff 1989) that the theory is consistent with some psychophysical experiments which investigate the relative importance of different matching primitives.",

author = "Yuille, {A. L.} and D. Geiger and H. B{\"u}lthoff",

note = "Funding Information: A.L.Y. would like to acknowledge support from the Brown/Harvard/MIT Center tror Intelligent Control Systems with U.S. Army Research Office grant number DAAL03-86-K-0171. Some of these ideas were initially developed with Mike Gennert. We would like to thank Mike Gennert, Manfred Fable, Jim Clark, and Norberto Grzywacz for helpful conversations. Publisher Copyright: {\textcopyright} Springer-Verlag Berlin Heidelberg 1990.; 1st European Conference on Computer Vision, ECCV 1990 ; Conference date: 23-04-1990 Through 27-04-1990",

year = "1990",

doi = "10.1007/BFb0014852",

language = "English (US)",

isbn = "9783540525226",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "73--82",

editor = "Olivier Faugeras",

booktitle = "Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings",

}

TY - GEN

T1 - Stereo integration, mean field theory and psychophysics

AU - Yuille, A. L.

AU - Geiger, D.

AU - Bülthoff, H.

N1 - Funding Information: A.L.Y. would like to acknowledge support from the Brown/Harvard/MIT Center tror Intelligent Control Systems with U.S. Army Research Office grant number DAAL03-86-K-0171. Some of these ideas were initially developed with Mike Gennert. We would like to thank Mike Gennert, Manfred Fable, Jim Clark, and Norberto Grzywacz for helpful conversations. Publisher Copyright: © Springer-Verlag Berlin Heidelberg 1990.

PY - 1990

Y1 - 1990

N2 - We describe a theoretical formulation for stereo in terms of the Markov Random Field and Bayesian approach to vision. This formulation enables us to integrate the depth information from different types of matching primitives, or from different vision modules. We treat the correspondence problem and surface interpolation as different aspects of the same problem and solve them simultaneously, unlike most previous theories. We use techniques from statistical physics to compute properties of our theory and show how it relates to previous work. These techniques also suggest novel algorithms for stereo which are argued to be preferable to standard algorithms on theoretical and experimental grounds. It can be shown (Yuille, Geiger and Bülthoff 1989) that the theory is consistent with some psychophysical experiments which investigate the relative importance of different matching primitives.

AB - We describe a theoretical formulation for stereo in terms of the Markov Random Field and Bayesian approach to vision. This formulation enables us to integrate the depth information from different types of matching primitives, or from different vision modules. We treat the correspondence problem and surface interpolation as different aspects of the same problem and solve them simultaneously, unlike most previous theories. We use techniques from statistical physics to compute properties of our theory and show how it relates to previous work. These techniques also suggest novel algorithms for stereo which are argued to be preferable to standard algorithms on theoretical and experimental grounds. It can be shown (Yuille, Geiger and Bülthoff 1989) that the theory is consistent with some psychophysical experiments which investigate the relative importance of different matching primitives.

UR - http://www.scopus.com/inward/record.url?scp=84921842488&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84921842488&partnerID=8YFLogxK

U2 - 10.1007/BFb0014852

DO - 10.1007/BFb0014852

M3 - Conference contribution

AN - SCOPUS:84921842488

SN - 9783540525226

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 73

EP - 82

BT - Computer Vision – ECCV 1990 - 1st European Conference on Computer Vision, Proceedings

A2 - Faugeras, Olivier

PB - Springer Verlag

T2 - 1st European Conference on Computer Vision, ECCV 1990

Y2 - 23 April 1990 through 27 April 1990

ER -

Stereo integration, mean field theory and psychophysics

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this