A saliency-based hierarchy for local symmetries

Mark Jenkinson, Michael Brady

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Shape representation is fundamental to many areas of computer vision. Local symmetries provide a tool on which to base flexible representations of complex shapes, however, 'multiple participation' symmetries such as SLS generate many more symmetries, than required. This clutters the representation with mostly irrelevant symmetries, hiding perceptually salient information. In this paper, a salience hierarchy is proposed that allows compact, perceptually pleasing representations to be extracted from the highly redundant full SLS. Both individual and competitive saliency is used to create the hierarchy which partitions the set of symmetries into subsets, each of which is an independent shape representation.

LanguageEnglish
Pages85-101
Number of pages17
JournalImage and Vision Computing
Volume20
Issue number2
DOIs
Publication statusPublished - 1 Feb 2002
Externally publishedYes

Keywords

  • Local symmetries
  • Saliency
  • Shape representation

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Jenkinson, Mark ; Brady, Michael. / A saliency-based hierarchy for local symmetries. In: Image and Vision Computing. 2002 ; Vol. 20, No. 2. pp. 85-101.
@article{c3cf8530bffb44818e3d05ec240f12c0,
title = "A saliency-based hierarchy for local symmetries",
abstract = "Shape representation is fundamental to many areas of computer vision. Local symmetries provide a tool on which to base flexible representations of complex shapes, however, 'multiple participation' symmetries such as SLS generate many more symmetries, than required. This clutters the representation with mostly irrelevant symmetries, hiding perceptually salient information. In this paper, a salience hierarchy is proposed that allows compact, perceptually pleasing representations to be extracted from the highly redundant full SLS. Both individual and competitive saliency is used to create the hierarchy which partitions the set of symmetries into subsets, each of which is an independent shape representation.",
keywords = "Local symmetries, Saliency, Shape representation",
author = "Mark Jenkinson and Michael Brady",
year = "2002",
month = "2",
day = "1",
doi = "10.1016/S0262-8856(01)00080-4",
language = "English",
volume = "20",
pages = "85--101",
journal = "Image and Vision Computing",
issn = "0262-8856",
number = "2",

}

A saliency-based hierarchy for local symmetries. / Jenkinson, Mark; Brady, Michael.

In: Image and Vision Computing, Vol. 20, No. 2, 01.02.2002, p. 85-101.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A saliency-based hierarchy for local symmetries

AU - Jenkinson, Mark

AU - Brady, Michael

PY - 2002/2/1

Y1 - 2002/2/1

N2 - Shape representation is fundamental to many areas of computer vision. Local symmetries provide a tool on which to base flexible representations of complex shapes, however, 'multiple participation' symmetries such as SLS generate many more symmetries, than required. This clutters the representation with mostly irrelevant symmetries, hiding perceptually salient information. In this paper, a salience hierarchy is proposed that allows compact, perceptually pleasing representations to be extracted from the highly redundant full SLS. Both individual and competitive saliency is used to create the hierarchy which partitions the set of symmetries into subsets, each of which is an independent shape representation.

AB - Shape representation is fundamental to many areas of computer vision. Local symmetries provide a tool on which to base flexible representations of complex shapes, however, 'multiple participation' symmetries such as SLS generate many more symmetries, than required. This clutters the representation with mostly irrelevant symmetries, hiding perceptually salient information. In this paper, a salience hierarchy is proposed that allows compact, perceptually pleasing representations to be extracted from the highly redundant full SLS. Both individual and competitive saliency is used to create the hierarchy which partitions the set of symmetries into subsets, each of which is an independent shape representation.

KW - Local symmetries

KW - Saliency

KW - Shape representation

UR - http://www.scopus.com/inward/record.url?scp=0036468045&partnerID=8YFLogxK

U2 - 10.1016/S0262-8856(01)00080-4

DO - 10.1016/S0262-8856(01)00080-4

M3 - Article

VL - 20

SP - 85

EP - 101

JO - Image and Vision Computing

T2 - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

IS - 2

ER -