Residual of resonant SVD as salient feature

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

4 Citations (Scopus)

Abstract

Computer vision approaches to saliency are based, among others, on uniqueness [1], local complexity [2], distinctiveness [3,4], spectral variation [5], and irregularity [6]. Saliency can also be viewed as the information in the data relative to a representation or model [7]. When a representation is built, a residual error is often minimised. The residual can be used to obtain saliency maps for solving challenging tasks of image and video processing. We introduce the notion of the resonant SVD and demonstrate that the SVD residual at the resonant spacing is selective to defects in spatially periodic surface textures and events in time-periodic videos. Examples with real-world images and videos are shown and discussed.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages143-153
Number of pages11
Volume5337 LNCS
DOIs
Publication statusPublished - 2009
EventInternational Conference on Computer Vision and Graphics, ICCVG 2008 - Warsaw, Poland
Duration: Nov 10 2008Nov 12 2008

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5337 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

OtherInternational Conference on Computer Vision and Graphics, ICCVG 2008
CountryPoland
CityWarsaw
Period11/10/0811/12/08

Fingerprint

Singular value decomposition
Saliency
Computer vision
Saliency Map
Surface Texture
Video Processing
Textures
Irregularity
Computer Vision
Defects
Spacing
Image Processing
Uniqueness
Processing
Demonstrate
Model

Keywords

  • Defects
  • Periodicity
  • Saliency
  • SVD
  • Texture
  • Video processing

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Chetverikov, D. (2009). Residual of resonant SVD as salient feature. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5337 LNCS, pp. 143-153). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5337 LNCS). https://doi.org/10.1007/978-3-642-02345-3_15

Residual of resonant SVD as salient feature. / Chetverikov, D.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5337 LNCS 2009. p. 143-153 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5337 LNCS).

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

Chetverikov, D 2009, Residual of resonant SVD as salient feature. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 5337 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 5337 LNCS, pp. 143-153, International Conference on Computer Vision and Graphics, ICCVG 2008, Warsaw, Poland, 11/10/08. https://doi.org/10.1007/978-3-642-02345-3_15
Chetverikov D. Residual of resonant SVD as salient feature. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5337 LNCS. 2009. p. 143-153. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-02345-3_15
Chetverikov, D. / Residual of resonant SVD as salient feature. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5337 LNCS 2009. pp. 143-153 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{1cc7f6661c6a45ca8e26284684940e20,
title = "Residual of resonant SVD as salient feature",
abstract = "Computer vision approaches to saliency are based, among others, on uniqueness [1], local complexity [2], distinctiveness [3,4], spectral variation [5], and irregularity [6]. Saliency can also be viewed as the information in the data relative to a representation or model [7]. When a representation is built, a residual error is often minimised. The residual can be used to obtain saliency maps for solving challenging tasks of image and video processing. We introduce the notion of the resonant SVD and demonstrate that the SVD residual at the resonant spacing is selective to defects in spatially periodic surface textures and events in time-periodic videos. Examples with real-world images and videos are shown and discussed.",
keywords = "Defects, Periodicity, Saliency, SVD, Texture, Video processing",
author = "D. Chetverikov",
year = "2009",
doi = "10.1007/978-3-642-02345-3_15",
language = "English",
isbn = "3642023444",
volume = "5337 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "143--153",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Residual of resonant SVD as salient feature

AU - Chetverikov, D.

PY - 2009

Y1 - 2009

N2 - Computer vision approaches to saliency are based, among others, on uniqueness [1], local complexity [2], distinctiveness [3,4], spectral variation [5], and irregularity [6]. Saliency can also be viewed as the information in the data relative to a representation or model [7]. When a representation is built, a residual error is often minimised. The residual can be used to obtain saliency maps for solving challenging tasks of image and video processing. We introduce the notion of the resonant SVD and demonstrate that the SVD residual at the resonant spacing is selective to defects in spatially periodic surface textures and events in time-periodic videos. Examples with real-world images and videos are shown and discussed.

AB - Computer vision approaches to saliency are based, among others, on uniqueness [1], local complexity [2], distinctiveness [3,4], spectral variation [5], and irregularity [6]. Saliency can also be viewed as the information in the data relative to a representation or model [7]. When a representation is built, a residual error is often minimised. The residual can be used to obtain saliency maps for solving challenging tasks of image and video processing. We introduce the notion of the resonant SVD and demonstrate that the SVD residual at the resonant spacing is selective to defects in spatially periodic surface textures and events in time-periodic videos. Examples with real-world images and videos are shown and discussed.

KW - Defects

KW - Periodicity

KW - Saliency

KW - SVD

KW - Texture

KW - Video processing

UR - http://www.scopus.com/inward/record.url?scp=68749102612&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=68749102612&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-02345-3_15

DO - 10.1007/978-3-642-02345-3_15

M3 - Conference contribution

AN - SCOPUS:68749102612

SN - 3642023444

SN - 9783642023446

VL - 5337 LNCS

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

SP - 143

EP - 153

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

ER -