Finding semantic structures in image hierarchies using Laplacian graph energy

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

Research output: Chapter or section in a book/report/conference proceedingChapter in a published conference proceeding

29 Citations (SciVal)
228 Downloads (Pure)

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.
Original 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
Publication statusPublished - Sept 2010
Event11th European Conference on Computer Vision, ECCV 2010, September 5, 2010 - September 11, 2010 - Heraklion, Crete, Greece
Duration: 1 Sept 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
Country/TerritoryGreece
CityHeraklion, Crete
Period1/09/10 → …

Fingerprint

Dive into the research topics of 'Finding semantic structures in image hierarchies using Laplacian graph energy'. Together they form a unique fingerprint.

Cite this