Shape from texture without boundaries

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We describe a shape from texture method that constructs a maximum a posteriori estimate of surface coefficients using only the deformation of individual texture elements. Our method does not need to use either the boundary of the observed surface or any assumption about the overall distribution of elements. The method assumes that texture elements are of a limited number of types of fixed shape. We show that, with this assumption and assuming generic view and texture, each texture element yields the surface gradient unique up to a two-fold ambiguity. Furthermore, texture elements that are not from one of the types can be identified and ignored. An EM-like procedure yields a surface reconstruction from the data. The method is defined for othographic views — an extension to perspective views appears to be complex, but possible. Examples of reconstructions for synthetic images of surfaces are provided, and compared with ground truth. We also provide examples of reconstructions for images of real scenes. We show that our method for recovering local texture imaging transformations can be used to retexture objects in images of real scenes.

Original languageEnglish (US)
Title of host publicationComputer Vision - ECCV 2002 - 7th European Conference on Computer Vision, Proceedings
EditorsMads Nielsen, Peter Johansen, Anders Heyden, Gunnar Sparr
PublisherSpringer-Verlag
Pages225-239
Number of pages15
ISBN (Print)3540437460, 9783540437468
StatePublished - Jan 1 2002
Externally publishedYes
Event7th European Conference on Computer Vision, ECCV 2002 - Copenhagen, Denmark
Duration: May 28 2002May 31 2002

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2352
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other7th European Conference on Computer Vision, ECCV 2002
CountryDenmark
CityCopenhagen
Period5/28/025/31/02

Fingerprint

Texture
Textures
A Posteriori Estimates
Surface Reconstruction
Surface reconstruction
Maximum a Posteriori
Fold
Imaging
Gradient
Imaging techniques
Coefficient

Keywords

  • Computer vision
  • Shape from texture
  • Surface fitting
  • Texture

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Forsyth, D. A. (2002). Shape from texture without boundaries. In M. Nielsen, P. Johansen, A. Heyden, & G. Sparr (Eds.), Computer Vision - ECCV 2002 - 7th European Conference on Computer Vision, Proceedings (pp. 225-239). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2352). Springer-Verlag.

Shape from texture without boundaries. / Forsyth, D. A.

Computer Vision - ECCV 2002 - 7th European Conference on Computer Vision, Proceedings. ed. / Mads Nielsen; Peter Johansen; Anders Heyden; Gunnar Sparr. Springer-Verlag, 2002. p. 225-239 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2352).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Forsyth, DA 2002, Shape from texture without boundaries. in M Nielsen, P Johansen, A Heyden & G Sparr (eds), Computer Vision - ECCV 2002 - 7th European Conference on Computer Vision, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2352, Springer-Verlag, pp. 225-239, 7th European Conference on Computer Vision, ECCV 2002, Copenhagen, Denmark, 5/28/02.
Forsyth DA. Shape from texture without boundaries. In Nielsen M, Johansen P, Heyden A, Sparr G, editors, Computer Vision - ECCV 2002 - 7th European Conference on Computer Vision, Proceedings. Springer-Verlag. 2002. p. 225-239. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Forsyth, D. A. / Shape from texture without boundaries. Computer Vision - ECCV 2002 - 7th European Conference on Computer Vision, Proceedings. editor / Mads Nielsen ; Peter Johansen ; Anders Heyden ; Gunnar Sparr. Springer-Verlag, 2002. pp. 225-239 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{28032aecc9fa423db2b37242f5fb040b,
title = "Shape from texture without boundaries",
abstract = "We describe a shape from texture method that constructs a maximum a posteriori estimate of surface coefficients using only the deformation of individual texture elements. Our method does not need to use either the boundary of the observed surface or any assumption about the overall distribution of elements. The method assumes that texture elements are of a limited number of types of fixed shape. We show that, with this assumption and assuming generic view and texture, each texture element yields the surface gradient unique up to a two-fold ambiguity. Furthermore, texture elements that are not from one of the types can be identified and ignored. An EM-like procedure yields a surface reconstruction from the data. The method is defined for othographic views — an extension to perspective views appears to be complex, but possible. Examples of reconstructions for synthetic images of surfaces are provided, and compared with ground truth. We also provide examples of reconstructions for images of real scenes. We show that our method for recovering local texture imaging transformations can be used to retexture objects in images of real scenes.",
keywords = "Computer vision, Shape from texture, Surface fitting, Texture",
author = "Forsyth, {D. A.}",
year = "2002",
month = "1",
day = "1",
language = "English (US)",
isbn = "3540437460",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "225--239",
editor = "Mads Nielsen and Peter Johansen and Anders Heyden and Gunnar Sparr",
booktitle = "Computer Vision - ECCV 2002 - 7th European Conference on Computer Vision, Proceedings",

}

TY - GEN

T1 - Shape from texture without boundaries

AU - Forsyth, D. A.

PY - 2002/1/1

Y1 - 2002/1/1

N2 - We describe a shape from texture method that constructs a maximum a posteriori estimate of surface coefficients using only the deformation of individual texture elements. Our method does not need to use either the boundary of the observed surface or any assumption about the overall distribution of elements. The method assumes that texture elements are of a limited number of types of fixed shape. We show that, with this assumption and assuming generic view and texture, each texture element yields the surface gradient unique up to a two-fold ambiguity. Furthermore, texture elements that are not from one of the types can be identified and ignored. An EM-like procedure yields a surface reconstruction from the data. The method is defined for othographic views — an extension to perspective views appears to be complex, but possible. Examples of reconstructions for synthetic images of surfaces are provided, and compared with ground truth. We also provide examples of reconstructions for images of real scenes. We show that our method for recovering local texture imaging transformations can be used to retexture objects in images of real scenes.

AB - We describe a shape from texture method that constructs a maximum a posteriori estimate of surface coefficients using only the deformation of individual texture elements. Our method does not need to use either the boundary of the observed surface or any assumption about the overall distribution of elements. The method assumes that texture elements are of a limited number of types of fixed shape. We show that, with this assumption and assuming generic view and texture, each texture element yields the surface gradient unique up to a two-fold ambiguity. Furthermore, texture elements that are not from one of the types can be identified and ignored. An EM-like procedure yields a surface reconstruction from the data. The method is defined for othographic views — an extension to perspective views appears to be complex, but possible. Examples of reconstructions for synthetic images of surfaces are provided, and compared with ground truth. We also provide examples of reconstructions for images of real scenes. We show that our method for recovering local texture imaging transformations can be used to retexture objects in images of real scenes.

KW - Computer vision

KW - Shape from texture

KW - Surface fitting

KW - Texture

UR - http://www.scopus.com/inward/record.url?scp=84949988466&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949988466&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84949988466

SN - 3540437460

SN - 9783540437468

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

SP - 225

EP - 239

BT - Computer Vision - ECCV 2002 - 7th European Conference on Computer Vision, Proceedings

A2 - Nielsen, Mads

A2 - Johansen, Peter

A2 - Heyden, Anders

A2 - Sparr, Gunnar

PB - Springer-Verlag

ER -