In search of perceptually salient groupings

Y Z Song, B Xiao, Peter Hall, Liang Wang

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Finding meaningful groupings of image primitives has been a long-standing problem in computer vision. This paper studies how salient groupings can be produced using established theories in the field of visual perception alone. The major contribution is a novel definition of the Gestalt principle of Pragnanz, based upon Koffka's definition that image descriptions should be both stable and simple. Our method is global in the sense that it operates over all primitives in an image at once. It works regardless of the type of image primitives and is generally independent of image properties such as intensity, color, and texture. A novel experiment is designed to quantitatively evaluate the groupings outputs by our method, which takes human disagreement into account and is generic to outputs of any grouper. We also demonstrate the value of our method in an image segmentation application and quantitatively show that segmentations deliver promising results when benchmarked using the Berkeley Segmentation Dataset (BSDS).
Original languageEnglish
Pages (from-to)935-947
Number of pages13
JournalIEEE Transactions on Image Processing
Volume20
Issue number4
Early online date17 Oct 2010
DOIs
Publication statusPublished - Apr 2011

Fingerprint

Image segmentation
Computer vision
Textures
Color
Experiments

Keywords

  • gestalt
  • pragnanz
  • perceptual grouping
  • image segmentation

Cite this

In search of perceptually salient groupings. / Song, Y Z; Xiao, B; Hall, Peter; Wang, Liang.

In: IEEE Transactions on Image Processing, Vol. 20, No. 4, 04.2011, p. 935-947.

Research output: Contribution to journalArticle

Song, Y Z ; Xiao, B ; Hall, Peter ; Wang, Liang. / In search of perceptually salient groupings. In: IEEE Transactions on Image Processing. 2011 ; Vol. 20, No. 4. pp. 935-947.
@article{6e32f30e37ea4df397f7e7645eaa7de1,
title = "In search of perceptually salient groupings",
abstract = "Finding meaningful groupings of image primitives has been a long-standing problem in computer vision. This paper studies how salient groupings can be produced using established theories in the field of visual perception alone. The major contribution is a novel definition of the Gestalt principle of Pragnanz, based upon Koffka's definition that image descriptions should be both stable and simple. Our method is global in the sense that it operates over all primitives in an image at once. It works regardless of the type of image primitives and is generally independent of image properties such as intensity, color, and texture. A novel experiment is designed to quantitatively evaluate the groupings outputs by our method, which takes human disagreement into account and is generic to outputs of any grouper. We also demonstrate the value of our method in an image segmentation application and quantitatively show that segmentations deliver promising results when benchmarked using the Berkeley Segmentation Dataset (BSDS).",
keywords = "gestalt, pragnanz, perceptual grouping, image segmentation",
author = "Song, {Y Z} and B Xiao and Peter Hall and Liang Wang",
year = "2011",
month = "4",
doi = "10.1109/tip.2010.2087766",
language = "English",
volume = "20",
pages = "935--947",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "IEEE",
number = "4",

}

TY - JOUR

T1 - In search of perceptually salient groupings

AU - Song, Y Z

AU - Xiao, B

AU - Hall, Peter

AU - Wang, Liang

PY - 2011/4

Y1 - 2011/4

N2 - Finding meaningful groupings of image primitives has been a long-standing problem in computer vision. This paper studies how salient groupings can be produced using established theories in the field of visual perception alone. The major contribution is a novel definition of the Gestalt principle of Pragnanz, based upon Koffka's definition that image descriptions should be both stable and simple. Our method is global in the sense that it operates over all primitives in an image at once. It works regardless of the type of image primitives and is generally independent of image properties such as intensity, color, and texture. A novel experiment is designed to quantitatively evaluate the groupings outputs by our method, which takes human disagreement into account and is generic to outputs of any grouper. We also demonstrate the value of our method in an image segmentation application and quantitatively show that segmentations deliver promising results when benchmarked using the Berkeley Segmentation Dataset (BSDS).

AB - Finding meaningful groupings of image primitives has been a long-standing problem in computer vision. This paper studies how salient groupings can be produced using established theories in the field of visual perception alone. The major contribution is a novel definition of the Gestalt principle of Pragnanz, based upon Koffka's definition that image descriptions should be both stable and simple. Our method is global in the sense that it operates over all primitives in an image at once. It works regardless of the type of image primitives and is generally independent of image properties such as intensity, color, and texture. A novel experiment is designed to quantitatively evaluate the groupings outputs by our method, which takes human disagreement into account and is generic to outputs of any grouper. We also demonstrate the value of our method in an image segmentation application and quantitatively show that segmentations deliver promising results when benchmarked using the Berkeley Segmentation Dataset (BSDS).

KW - gestalt

KW - pragnanz

KW - perceptual grouping

KW - image segmentation

UR - http://www.scopus.com/inward/record.url?scp=79952906625&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1109/tip.2010.2087766

U2 - 10.1109/tip.2010.2087766

DO - 10.1109/tip.2010.2087766

M3 - Article

VL - 20

SP - 935

EP - 947

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 4

ER -