Finding tree structures by grouping symmetries

Hiroshi Ishikawa, Davi Geiger, Richard Cole

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

6 Citations (Scopus)

Abstract

The representation of objects in images as tree structures is of great interest to vision, as they can represent articulated objects such as people as well as other structured objects like arteries in human bodies, roads, circuit board patterns, etc. Tree structures are often related to the symmetry axis representation of shapes, which captures their local symmetries. Algorithms have been introduced to detect (i) open contours in images in quadratic time (ii) closed contours in images in cubic time, and (in) tree structures from contours in quadratic time. The algorithms are based on dynamic programming and Single Source Shortest Path algorithms. However, in this paper, we show that the problem of finding tree structures in images in a principled manner is a much harder problem. We argue that the optimization problem of finding tree structures in images is essentially equivalent to a variant of the Steiner Tree problem, which is NP-hard. Nevertheless, an approximate polynomial-time algorithm for this problem exists: we apply a fast implementation of the Goemans-Williamson approximate algorithm to the problem of finding a tree representation after an image is transformed by a local symmetry mapping. Examples of extracting tree structures from images illustrate the idea and applicability of the approximate method.

Original languageEnglish
Title of host publicationProceedings of the IEEE International Conference on Computer Vision
Pages1132-1139
Number of pages8
VolumeII
DOIs
Publication statusPublished - 2005
Externally publishedYes
EventProceedings - 10th IEEE International Conference on Computer Vision, ICCV 2005 - Beijing
Duration: 2005 Oct 172005 Oct 20

Other

OtherProceedings - 10th IEEE International Conference on Computer Vision, ICCV 2005
CityBeijing
Period05/10/1705/10/20

Fingerprint

Dynamic programming
Polynomials
Networks (circuits)

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Ishikawa, H., Geiger, D., & Cole, R. (2005). Finding tree structures by grouping symmetries. In Proceedings of the IEEE International Conference on Computer Vision (Vol. II, pp. 1132-1139). [1544848] https://doi.org/10.1109/ICCV.2005.100

Finding tree structures by grouping symmetries. / Ishikawa, Hiroshi; Geiger, Davi; Cole, Richard.

Proceedings of the IEEE International Conference on Computer Vision. Vol. II 2005. p. 1132-1139 1544848.

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

Ishikawa, H, Geiger, D & Cole, R 2005, Finding tree structures by grouping symmetries. in Proceedings of the IEEE International Conference on Computer Vision. vol. II, 1544848, pp. 1132-1139, Proceedings - 10th IEEE International Conference on Computer Vision, ICCV 2005, Beijing, 05/10/17. https://doi.org/10.1109/ICCV.2005.100
Ishikawa H, Geiger D, Cole R. Finding tree structures by grouping symmetries. In Proceedings of the IEEE International Conference on Computer Vision. Vol. II. 2005. p. 1132-1139. 1544848 https://doi.org/10.1109/ICCV.2005.100
Ishikawa, Hiroshi ; Geiger, Davi ; Cole, Richard. / Finding tree structures by grouping symmetries. Proceedings of the IEEE International Conference on Computer Vision. Vol. II 2005. pp. 1132-1139
@inproceedings{3350361f7c8e438692349439ae196b06,
title = "Finding tree structures by grouping symmetries",
abstract = "The representation of objects in images as tree structures is of great interest to vision, as they can represent articulated objects such as people as well as other structured objects like arteries in human bodies, roads, circuit board patterns, etc. Tree structures are often related to the symmetry axis representation of shapes, which captures their local symmetries. Algorithms have been introduced to detect (i) open contours in images in quadratic time (ii) closed contours in images in cubic time, and (in) tree structures from contours in quadratic time. The algorithms are based on dynamic programming and Single Source Shortest Path algorithms. However, in this paper, we show that the problem of finding tree structures in images in a principled manner is a much harder problem. We argue that the optimization problem of finding tree structures in images is essentially equivalent to a variant of the Steiner Tree problem, which is NP-hard. Nevertheless, an approximate polynomial-time algorithm for this problem exists: we apply a fast implementation of the Goemans-Williamson approximate algorithm to the problem of finding a tree representation after an image is transformed by a local symmetry mapping. Examples of extracting tree structures from images illustrate the idea and applicability of the approximate method.",
author = "Hiroshi Ishikawa and Davi Geiger and Richard Cole",
year = "2005",
doi = "10.1109/ICCV.2005.100",
language = "English",
isbn = "076952334X",
volume = "II",
pages = "1132--1139",
booktitle = "Proceedings of the IEEE International Conference on Computer Vision",

}

TY - GEN

T1 - Finding tree structures by grouping symmetries

AU - Ishikawa, Hiroshi

AU - Geiger, Davi

AU - Cole, Richard

PY - 2005

Y1 - 2005

N2 - The representation of objects in images as tree structures is of great interest to vision, as they can represent articulated objects such as people as well as other structured objects like arteries in human bodies, roads, circuit board patterns, etc. Tree structures are often related to the symmetry axis representation of shapes, which captures their local symmetries. Algorithms have been introduced to detect (i) open contours in images in quadratic time (ii) closed contours in images in cubic time, and (in) tree structures from contours in quadratic time. The algorithms are based on dynamic programming and Single Source Shortest Path algorithms. However, in this paper, we show that the problem of finding tree structures in images in a principled manner is a much harder problem. We argue that the optimization problem of finding tree structures in images is essentially equivalent to a variant of the Steiner Tree problem, which is NP-hard. Nevertheless, an approximate polynomial-time algorithm for this problem exists: we apply a fast implementation of the Goemans-Williamson approximate algorithm to the problem of finding a tree representation after an image is transformed by a local symmetry mapping. Examples of extracting tree structures from images illustrate the idea and applicability of the approximate method.

AB - The representation of objects in images as tree structures is of great interest to vision, as they can represent articulated objects such as people as well as other structured objects like arteries in human bodies, roads, circuit board patterns, etc. Tree structures are often related to the symmetry axis representation of shapes, which captures their local symmetries. Algorithms have been introduced to detect (i) open contours in images in quadratic time (ii) closed contours in images in cubic time, and (in) tree structures from contours in quadratic time. The algorithms are based on dynamic programming and Single Source Shortest Path algorithms. However, in this paper, we show that the problem of finding tree structures in images in a principled manner is a much harder problem. We argue that the optimization problem of finding tree structures in images is essentially equivalent to a variant of the Steiner Tree problem, which is NP-hard. Nevertheless, an approximate polynomial-time algorithm for this problem exists: we apply a fast implementation of the Goemans-Williamson approximate algorithm to the problem of finding a tree representation after an image is transformed by a local symmetry mapping. Examples of extracting tree structures from images illustrate the idea and applicability of the approximate method.

UR - http://www.scopus.com/inward/record.url?scp=33745929058&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745929058&partnerID=8YFLogxK

U2 - 10.1109/ICCV.2005.100

DO - 10.1109/ICCV.2005.100

M3 - Conference contribution

SN - 076952334X

SN - 9780769523347

VL - II

SP - 1132

EP - 1139

BT - Proceedings of the IEEE International Conference on Computer Vision

ER -