Finding semantic structures in image hierarchies using Laplacian graph energy

Yi-Zhe Song, Pablo Arbelaez, Peter Hall, Chuan Li, Anupriya Balikai

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

  • 25 Citations

Abstract

Many segmentation algorithms describe images in terms of a hierarchy of regions. Although such hierarchies can produce state of the art segmentations and have many applications, they often contain more data than is required for an efficient description. This paper shows Laplacian graph energy is a generic measure that can be used to identify semantic structures within hierarchies, independently of the algorithm that produces them. Quantitative experimental validation using hierarchies from two state of art algorithms show we can reduce the number of levels and regions in a hierarchy by an order of magnitude with little or no loss in performance when compared against human produced ground truth. We provide a tracking application that illustrates the value of reduced hierarchies.
LanguageEnglish
Title of host publicationComputer Vision, ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV
EditorsK Daniilidis, P Maragos, N Paragios
Place of PublicationBerlin
PublisherSpringer
Pages694-707
Number of pages14
ISBN (Electronic)9783642155611
ISBN (Print)9783642155604
DOIs
StatusPublished - Sep 2010
Event11th European Conference on Computer Vision, ECCV 2010, September 5, 2010 - September 11, 2010 - Heraklion, Crete, Greece
Duration: 1 Sep 2010 → …

Publication series

NameLecture Notes in Computer Science
Volume6314
ISSN (Print)0302-9743

Conference

Conference11th European Conference on Computer Vision, ECCV 2010, September 5, 2010 - September 11, 2010
CountryGreece
CityHeraklion, Crete
Period1/09/10 → …

Fingerprint

Semantics

Cite this

Song, Y-Z., Arbelaez, P., Hall, P., Li, C., & Balikai, A. (2010). Finding semantic structures in image hierarchies using Laplacian graph energy. In K. Daniilidis, P. Maragos, & N. Paragios (Eds.), Computer Vision, ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV (pp. 694-707). (Lecture Notes in Computer Science; Vol. 6314). Berlin: Springer. DOI: 10.1007/978-3-642-15561-1_50

Finding semantic structures in image hierarchies using Laplacian graph energy. / Song, Yi-Zhe; Arbelaez, Pablo; Hall, Peter; Li, Chuan; Balikai, Anupriya.

Computer Vision, ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV. ed. / K Daniilidis; P Maragos; N Paragios. Berlin : Springer, 2010. p. 694-707 (Lecture Notes in Computer Science; Vol. 6314).

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

Song, Y-Z, Arbelaez, P, Hall, P, Li, C & Balikai, A 2010, Finding semantic structures in image hierarchies using Laplacian graph energy. in K Daniilidis, P Maragos & N Paragios (eds), Computer Vision, ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV. Lecture Notes in Computer Science, vol. 6314, Springer, Berlin, pp. 694-707, 11th European Conference on Computer Vision, ECCV 2010, September 5, 2010 - September 11, 2010, Heraklion, Crete, Greece, 1/09/10. DOI: 10.1007/978-3-642-15561-1_50
Song Y-Z, Arbelaez P, Hall P, Li C, Balikai A. Finding semantic structures in image hierarchies using Laplacian graph energy. In Daniilidis K, Maragos P, Paragios N, editors, Computer Vision, ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV. Berlin: Springer. 2010. p. 694-707. (Lecture Notes in Computer Science). Available from, DOI: 10.1007/978-3-642-15561-1_50
Song, Yi-Zhe ; Arbelaez, Pablo ; Hall, Peter ; Li, Chuan ; Balikai, Anupriya. / Finding semantic structures in image hierarchies using Laplacian graph energy. Computer Vision, ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV. editor / K Daniilidis ; P Maragos ; N Paragios. Berlin : Springer, 2010. pp. 694-707 (Lecture Notes in Computer Science).
@inproceedings{93a434fa7d8f4964bbcb3ccebc600039,
title = "Finding semantic structures in image hierarchies using Laplacian graph energy",
abstract = "Many segmentation algorithms describe images in terms of a hierarchy of regions. Although such hierarchies can produce state of the art segmentations and have many applications, they often contain more data than is required for an efficient description. This paper shows Laplacian graph energy is a generic measure that can be used to identify semantic structures within hierarchies, independently of the algorithm that produces them. Quantitative experimental validation using hierarchies from two state of art algorithms show we can reduce the number of levels and regions in a hierarchy by an order of magnitude with little or no loss in performance when compared against human produced ground truth. We provide a tracking application that illustrates the value of reduced hierarchies.",
author = "Yi-Zhe Song and Pablo Arbelaez and Peter Hall and Chuan Li and Anupriya Balikai",
year = "2010",
month = "9",
doi = "10.1007/978-3-642-15561-1_50",
language = "English",
isbn = "9783642155604",
series = "Lecture Notes in Computer Science",
publisher = "Springer",
pages = "694--707",
editor = "K Daniilidis and P Maragos and N Paragios",
booktitle = "Computer Vision, ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV",

}

TY - GEN

T1 - Finding semantic structures in image hierarchies using Laplacian graph energy

AU - Song,Yi-Zhe

AU - Arbelaez,Pablo

AU - Hall,Peter

AU - Li,Chuan

AU - Balikai,Anupriya

PY - 2010/9

Y1 - 2010/9

N2 - Many segmentation algorithms describe images in terms of a hierarchy of regions. Although such hierarchies can produce state of the art segmentations and have many applications, they often contain more data than is required for an efficient description. This paper shows Laplacian graph energy is a generic measure that can be used to identify semantic structures within hierarchies, independently of the algorithm that produces them. Quantitative experimental validation using hierarchies from two state of art algorithms show we can reduce the number of levels and regions in a hierarchy by an order of magnitude with little or no loss in performance when compared against human produced ground truth. We provide a tracking application that illustrates the value of reduced hierarchies.

AB - Many segmentation algorithms describe images in terms of a hierarchy of regions. Although such hierarchies can produce state of the art segmentations and have many applications, they often contain more data than is required for an efficient description. This paper shows Laplacian graph energy is a generic measure that can be used to identify semantic structures within hierarchies, independently of the algorithm that produces them. Quantitative experimental validation using hierarchies from two state of art algorithms show we can reduce the number of levels and regions in a hierarchy by an order of magnitude with little or no loss in performance when compared against human produced ground truth. We provide a tracking application that illustrates the value of reduced hierarchies.

UR - http://www.scopus.com/inward/record.url?scp=78149337071&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1007/978-3-642-15561-1_50

U2 - 10.1007/978-3-642-15561-1_50

DO - 10.1007/978-3-642-15561-1_50

M3 - Conference contribution

SN - 9783642155604

T3 - Lecture Notes in Computer Science

SP - 694

EP - 707

BT - Computer Vision, ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV

PB - Springer

CY - Berlin

ER -