A fast method for inferring high-quality simply-connected superpixels

Oren Freifeld, Yixin Li, John W. Fisher

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Scopus citations

Abstract

Superpixel segmentation is a key step in many image processing and vision tasks. Our recently-proposed connectivity-constrained probabilistic model [1] yields high-quality super-pixels. Seemingly, however, connectivity constraints preclude parallelized inference. As such, the implementation from [1] is serial. The contributions of this work are as follows. First, we demonstrate that effective parallelization is possible via a fast GPU implementation that scales gracefully with both the number of pixels and number of superpixels. Second, we show that the superpixels are improved by replacing the fixed and restricted spatial covariances from [1] with a flexible Bayesian prior. Quantitative evaluation on public benchmarks shows the proposed method outperforms the state-of-the-art. We make our implementation publicly available.

Original languageEnglish
Title of host publication2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings
PublisherIEEE Computer Society
Pages2184-2188
Number of pages5
ISBN (Electronic)9781479983391
DOIs
StatePublished - 9 Dec 2015
Externally publishedYes
EventIEEE International Conference on Image Processing, ICIP 2015 - Quebec City, Canada
Duration: 27 Sep 201530 Sep 2015

Publication series

NameProceedings - International Conference on Image Processing, ICIP
Volume2015-December
ISSN (Print)1522-4880

Conference

ConferenceIEEE International Conference on Image Processing, ICIP 2015
Country/TerritoryCanada
CityQuebec City
Period27/09/1530/09/15

Keywords

  • Superpixel
  • connectivity
  • inference
  • parallelization
  • probabilistic model
  • speedup

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Fingerprint

Dive into the research topics of 'A fast method for inferring high-quality simply-connected superpixels'. Together they form a unique fingerprint.

Cite this