Representation and self-similarity of shapes

Davi Geiger, Tyng Luh Liu, Robert Kohn

Research output: Contribution to journalArticle

Abstract

Representing shapes in a compact and informative form is a significant problem for vision systems that must recognize or classify objects. We describe a compact representation model for two-dimensional (2D) shapes by investigating their self-similarities and constructing their shape axis trees (SA-trees). Our approach can be formulated as a variational one (or, equivalently, as MAP estimation of a Markov random field). We start with a 2D shape, its boundary contour, and two different parameterizations for the contour (one parameterization is oriented counterclockwise and the other clockwise). To measure its self-similarity, the two parameterizations are matched to derive the best set of one-to-one point-to-point correspondences along the contour. The cost functional used in the matching may vary and is determined by the adopted self-similarity criteria, e.g., cocircularity, distance variation, parallelism, and region homogeneity. The loci of middle points of the pairing contour points yield the shape axis and they can be grouped into a unique free tree structure, the SA-tree. By implicitly encoding the (local and global) shape information into an SA-tree, a variety of vision tasks, e.g., shape recognition, comparison, and retrieval, can be performed in a more robust and efficient way via various tree-based algorithms. A dynamic programming algorithm gives the optimal solution in O(N4), where N is the size of the contour.

Original languageEnglish (US)
Pages (from-to)86-99
Number of pages14
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume25
Issue number1
DOIs
StatePublished - Jan 2003

Fingerprint

Self-similarity
Parameterization
Trees (mathematics)
Dynamic programming
MAP Estimation
Shape Recognition
Anticlockwise
Clockwise
Vision System
Tree Structure
Pairing
Homogeneity
Random Field
Parallelism
Dynamic Programming
Locus
Retrieval
Encoding
Correspondence
Optimal Solution

Keywords

  • Dynamic programming
  • MRF
  • Self-similarity
  • Shape representation
  • Variational matching

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Representation and self-similarity of shapes. / Geiger, Davi; Liu, Tyng Luh; Kohn, Robert.

In: IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 25, No. 1, 01.2003, p. 86-99.

Research output: Contribution to journalArticle

@article{45cee157fec44da2bcec55e731de5f8a,
title = "Representation and self-similarity of shapes",
abstract = "Representing shapes in a compact and informative form is a significant problem for vision systems that must recognize or classify objects. We describe a compact representation model for two-dimensional (2D) shapes by investigating their self-similarities and constructing their shape axis trees (SA-trees). Our approach can be formulated as a variational one (or, equivalently, as MAP estimation of a Markov random field). We start with a 2D shape, its boundary contour, and two different parameterizations for the contour (one parameterization is oriented counterclockwise and the other clockwise). To measure its self-similarity, the two parameterizations are matched to derive the best set of one-to-one point-to-point correspondences along the contour. The cost functional used in the matching may vary and is determined by the adopted self-similarity criteria, e.g., cocircularity, distance variation, parallelism, and region homogeneity. The loci of middle points of the pairing contour points yield the shape axis and they can be grouped into a unique free tree structure, the SA-tree. By implicitly encoding the (local and global) shape information into an SA-tree, a variety of vision tasks, e.g., shape recognition, comparison, and retrieval, can be performed in a more robust and efficient way via various tree-based algorithms. A dynamic programming algorithm gives the optimal solution in O(N4), where N is the size of the contour.",
keywords = "Dynamic programming, MRF, Self-similarity, Shape representation, Variational matching",
author = "Davi Geiger and Liu, {Tyng Luh} and Robert Kohn",
year = "2003",
month = "1",
doi = "10.1109/TPAMI.2003.1159948",
language = "English (US)",
volume = "25",
pages = "86--99",
journal = "IEEE Transactions on Pattern Analysis and Machine Intelligence",
issn = "0162-8828",
publisher = "IEEE Computer Society",
number = "1",

}

TY - JOUR

T1 - Representation and self-similarity of shapes

AU - Geiger, Davi

AU - Liu, Tyng Luh

AU - Kohn, Robert

PY - 2003/1

Y1 - 2003/1

N2 - Representing shapes in a compact and informative form is a significant problem for vision systems that must recognize or classify objects. We describe a compact representation model for two-dimensional (2D) shapes by investigating their self-similarities and constructing their shape axis trees (SA-trees). Our approach can be formulated as a variational one (or, equivalently, as MAP estimation of a Markov random field). We start with a 2D shape, its boundary contour, and two different parameterizations for the contour (one parameterization is oriented counterclockwise and the other clockwise). To measure its self-similarity, the two parameterizations are matched to derive the best set of one-to-one point-to-point correspondences along the contour. The cost functional used in the matching may vary and is determined by the adopted self-similarity criteria, e.g., cocircularity, distance variation, parallelism, and region homogeneity. The loci of middle points of the pairing contour points yield the shape axis and they can be grouped into a unique free tree structure, the SA-tree. By implicitly encoding the (local and global) shape information into an SA-tree, a variety of vision tasks, e.g., shape recognition, comparison, and retrieval, can be performed in a more robust and efficient way via various tree-based algorithms. A dynamic programming algorithm gives the optimal solution in O(N4), where N is the size of the contour.

AB - Representing shapes in a compact and informative form is a significant problem for vision systems that must recognize or classify objects. We describe a compact representation model for two-dimensional (2D) shapes by investigating their self-similarities and constructing their shape axis trees (SA-trees). Our approach can be formulated as a variational one (or, equivalently, as MAP estimation of a Markov random field). We start with a 2D shape, its boundary contour, and two different parameterizations for the contour (one parameterization is oriented counterclockwise and the other clockwise). To measure its self-similarity, the two parameterizations are matched to derive the best set of one-to-one point-to-point correspondences along the contour. The cost functional used in the matching may vary and is determined by the adopted self-similarity criteria, e.g., cocircularity, distance variation, parallelism, and region homogeneity. The loci of middle points of the pairing contour points yield the shape axis and they can be grouped into a unique free tree structure, the SA-tree. By implicitly encoding the (local and global) shape information into an SA-tree, a variety of vision tasks, e.g., shape recognition, comparison, and retrieval, can be performed in a more robust and efficient way via various tree-based algorithms. A dynamic programming algorithm gives the optimal solution in O(N4), where N is the size of the contour.

KW - Dynamic programming

KW - MRF

KW - Self-similarity

KW - Shape representation

KW - Variational matching

UR - http://www.scopus.com/inward/record.url?scp=0037250075&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037250075&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2003.1159948

DO - 10.1109/TPAMI.2003.1159948

M3 - Article

AN - SCOPUS:0037250075

VL - 25

SP - 86

EP - 99

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

SN - 0162-8828

IS - 1

ER -