Affine alignment of compound objects: A direct approach

Csaba Domokos, Z. Kato

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

3 Citations (Scopus)

Abstract

A direct approach for parametric estimation of 2D affine deformations between compound shapes is proposed. It provides the result as a least-square solution of a linear system of equations. The basic idea is to fit Gaussian densities over the objects yielding covariant functions, which preserves the effect of the unknown transformation. Based on these functions, linear equations are constructed by integrating nonlinear functions over appropriate domains. The main advantages are: linear complexity, easy implementation, works without any time consuming optimization or established correspondences. Comparative tests show that it outperforms state-of-the-art methods both in terms of precision, robustness and complexity.

Original languageEnglish
Title of host publicationProceedings - International Conference on Image Processing, ICIP
PublisherIEEE Computer Society
Pages169-172
Number of pages4
ISBN (Print)9781424456543
DOIs
Publication statusPublished - 2009
Event2009 IEEE International Conference on Image Processing, ICIP 2009 - Cairo, Egypt
Duration: Nov 7 2009Nov 10 2009

Other

Other2009 IEEE International Conference on Image Processing, ICIP 2009
CountryEgypt
CityCairo
Period11/7/0911/10/09

Fingerprint

Linear equations
Linear systems

Keywords

  • Affine registration
  • Compound shape
  • Gaussian distribution
  • Mahalanobis distance

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Domokos, C., & Kato, Z. (2009). Affine alignment of compound objects: A direct approach. In Proceedings - International Conference on Image Processing, ICIP (pp. 169-172). [5414195] IEEE Computer Society. https://doi.org/10.1109/ICIP.2009.5414195

Affine alignment of compound objects : A direct approach. / Domokos, Csaba; Kato, Z.

Proceedings - International Conference on Image Processing, ICIP. IEEE Computer Society, 2009. p. 169-172 5414195.

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

Domokos, C & Kato, Z 2009, Affine alignment of compound objects: A direct approach. in Proceedings - International Conference on Image Processing, ICIP., 5414195, IEEE Computer Society, pp. 169-172, 2009 IEEE International Conference on Image Processing, ICIP 2009, Cairo, Egypt, 11/7/09. https://doi.org/10.1109/ICIP.2009.5414195
Domokos C, Kato Z. Affine alignment of compound objects: A direct approach. In Proceedings - International Conference on Image Processing, ICIP. IEEE Computer Society. 2009. p. 169-172. 5414195 https://doi.org/10.1109/ICIP.2009.5414195
Domokos, Csaba ; Kato, Z. / Affine alignment of compound objects : A direct approach. Proceedings - International Conference on Image Processing, ICIP. IEEE Computer Society, 2009. pp. 169-172
@inproceedings{10653b43c6364a88926e0ce070fe20b2,
title = "Affine alignment of compound objects: A direct approach",
abstract = "A direct approach for parametric estimation of 2D affine deformations between compound shapes is proposed. It provides the result as a least-square solution of a linear system of equations. The basic idea is to fit Gaussian densities over the objects yielding covariant functions, which preserves the effect of the unknown transformation. Based on these functions, linear equations are constructed by integrating nonlinear functions over appropriate domains. The main advantages are: linear complexity, easy implementation, works without any time consuming optimization or established correspondences. Comparative tests show that it outperforms state-of-the-art methods both in terms of precision, robustness and complexity.",
keywords = "Affine registration, Compound shape, Gaussian distribution, Mahalanobis distance",
author = "Csaba Domokos and Z. Kato",
year = "2009",
doi = "10.1109/ICIP.2009.5414195",
language = "English",
isbn = "9781424456543",
pages = "169--172",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Affine alignment of compound objects

T2 - A direct approach

AU - Domokos, Csaba

AU - Kato, Z.

PY - 2009

Y1 - 2009

N2 - A direct approach for parametric estimation of 2D affine deformations between compound shapes is proposed. It provides the result as a least-square solution of a linear system of equations. The basic idea is to fit Gaussian densities over the objects yielding covariant functions, which preserves the effect of the unknown transformation. Based on these functions, linear equations are constructed by integrating nonlinear functions over appropriate domains. The main advantages are: linear complexity, easy implementation, works without any time consuming optimization or established correspondences. Comparative tests show that it outperforms state-of-the-art methods both in terms of precision, robustness and complexity.

AB - A direct approach for parametric estimation of 2D affine deformations between compound shapes is proposed. It provides the result as a least-square solution of a linear system of equations. The basic idea is to fit Gaussian densities over the objects yielding covariant functions, which preserves the effect of the unknown transformation. Based on these functions, linear equations are constructed by integrating nonlinear functions over appropriate domains. The main advantages are: linear complexity, easy implementation, works without any time consuming optimization or established correspondences. Comparative tests show that it outperforms state-of-the-art methods both in terms of precision, robustness and complexity.

KW - Affine registration

KW - Compound shape

KW - Gaussian distribution

KW - Mahalanobis distance

UR - http://www.scopus.com/inward/record.url?scp=77951939917&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951939917&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2009.5414195

DO - 10.1109/ICIP.2009.5414195

M3 - Conference contribution

AN - SCOPUS:77951939917

SN - 9781424456543

SP - 169

EP - 172

BT - Proceedings - International Conference on Image Processing, ICIP

PB - IEEE Computer Society

ER -