A Computational Model for Object-Based Visual Saliency: Spreading Attention Along Gestalt Cues

Jin Gang Yu, Gui Song Xia, Changxin Gao, Ashok K Samal

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

The past few years have witnessed impressive progress on the research of salient object detection. Nevertheless, existing approaches still cannot perform satisfactorily in the case of complex scenes, particularly when the salient objects have non-uniform appearance or complicated shapes, and the background is complexly structured. One important reason for such limitations may be that these approaches commonly ignore the factor of perceptual grouping in saliency modeling. To address this issue, this paper presents a novel computational model for object-based visual saliency, which explicitly takes into consideration the connections between attention and perceptual grouping, and incorporates Gestalt grouping cues into saliency computation. Inspired by the sensory enhancement theory, we suggest a paradigm for object-based saliency modeling, that is, object-based saliency stems from spreading attention along Gestalt grouping cues. Computationally, three typical Gestalt cues, including proximity, similarity, and closure, are respectively extracted from the given image, which are then integrated by constructing a unified Gestalt graph. A new algorithm named personalized power iteration clustering is developed to effectively fulfill the spreading of attention information across the Gestalt graph. Intensive experiments have been carried out to demonstrate the superior performance of the proposed model in comparison to the state-of-the-art.

Original languageEnglish (US)
Article number7347445
Pages (from-to)273-286
Number of pages14
JournalIEEE Transactions on Multimedia
Volume18
Issue number2
DOIs
StatePublished - Feb 1 2016

Fingerprint

Experiments
Object detection

Keywords

  • Gestalt principles
  • object-based visual saliency
  • perceptual grouping
  • personalized power iteration clustering (PPIC)
  • salient object detection
  • sensory enhancement theory

ASJC Scopus subject areas

  • Signal Processing
  • Media Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

A Computational Model for Object-Based Visual Saliency : Spreading Attention Along Gestalt Cues. / Yu, Jin Gang; Xia, Gui Song; Gao, Changxin; Samal, Ashok K.

In: IEEE Transactions on Multimedia, Vol. 18, No. 2, 7347445, 01.02.2016, p. 273-286.

Research output: Contribution to journalArticle

@article{2510e40702c444498f3ff8bd269c84e5,
title = "A Computational Model for Object-Based Visual Saliency: Spreading Attention Along Gestalt Cues",
abstract = "The past few years have witnessed impressive progress on the research of salient object detection. Nevertheless, existing approaches still cannot perform satisfactorily in the case of complex scenes, particularly when the salient objects have non-uniform appearance or complicated shapes, and the background is complexly structured. One important reason for such limitations may be that these approaches commonly ignore the factor of perceptual grouping in saliency modeling. To address this issue, this paper presents a novel computational model for object-based visual saliency, which explicitly takes into consideration the connections between attention and perceptual grouping, and incorporates Gestalt grouping cues into saliency computation. Inspired by the sensory enhancement theory, we suggest a paradigm for object-based saliency modeling, that is, object-based saliency stems from spreading attention along Gestalt grouping cues. Computationally, three typical Gestalt cues, including proximity, similarity, and closure, are respectively extracted from the given image, which are then integrated by constructing a unified Gestalt graph. A new algorithm named personalized power iteration clustering is developed to effectively fulfill the spreading of attention information across the Gestalt graph. Intensive experiments have been carried out to demonstrate the superior performance of the proposed model in comparison to the state-of-the-art.",
keywords = "Gestalt principles, object-based visual saliency, perceptual grouping, personalized power iteration clustering (PPIC), salient object detection, sensory enhancement theory",
author = "Yu, {Jin Gang} and Xia, {Gui Song} and Changxin Gao and Samal, {Ashok K}",
year = "2016",
month = "2",
day = "1",
doi = "10.1109/TMM.2015.2505908",
language = "English (US)",
volume = "18",
pages = "273--286",
journal = "IEEE Transactions on Multimedia",
issn = "1520-9210",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - A Computational Model for Object-Based Visual Saliency

T2 - Spreading Attention Along Gestalt Cues

AU - Yu, Jin Gang

AU - Xia, Gui Song

AU - Gao, Changxin

AU - Samal, Ashok K

PY - 2016/2/1

Y1 - 2016/2/1

N2 - The past few years have witnessed impressive progress on the research of salient object detection. Nevertheless, existing approaches still cannot perform satisfactorily in the case of complex scenes, particularly when the salient objects have non-uniform appearance or complicated shapes, and the background is complexly structured. One important reason for such limitations may be that these approaches commonly ignore the factor of perceptual grouping in saliency modeling. To address this issue, this paper presents a novel computational model for object-based visual saliency, which explicitly takes into consideration the connections between attention and perceptual grouping, and incorporates Gestalt grouping cues into saliency computation. Inspired by the sensory enhancement theory, we suggest a paradigm for object-based saliency modeling, that is, object-based saliency stems from spreading attention along Gestalt grouping cues. Computationally, three typical Gestalt cues, including proximity, similarity, and closure, are respectively extracted from the given image, which are then integrated by constructing a unified Gestalt graph. A new algorithm named personalized power iteration clustering is developed to effectively fulfill the spreading of attention information across the Gestalt graph. Intensive experiments have been carried out to demonstrate the superior performance of the proposed model in comparison to the state-of-the-art.

AB - The past few years have witnessed impressive progress on the research of salient object detection. Nevertheless, existing approaches still cannot perform satisfactorily in the case of complex scenes, particularly when the salient objects have non-uniform appearance or complicated shapes, and the background is complexly structured. One important reason for such limitations may be that these approaches commonly ignore the factor of perceptual grouping in saliency modeling. To address this issue, this paper presents a novel computational model for object-based visual saliency, which explicitly takes into consideration the connections between attention and perceptual grouping, and incorporates Gestalt grouping cues into saliency computation. Inspired by the sensory enhancement theory, we suggest a paradigm for object-based saliency modeling, that is, object-based saliency stems from spreading attention along Gestalt grouping cues. Computationally, three typical Gestalt cues, including proximity, similarity, and closure, are respectively extracted from the given image, which are then integrated by constructing a unified Gestalt graph. A new algorithm named personalized power iteration clustering is developed to effectively fulfill the spreading of attention information across the Gestalt graph. Intensive experiments have been carried out to demonstrate the superior performance of the proposed model in comparison to the state-of-the-art.

KW - Gestalt principles

KW - object-based visual saliency

KW - perceptual grouping

KW - personalized power iteration clustering (PPIC)

KW - salient object detection

KW - sensory enhancement theory

UR - http://www.scopus.com/inward/record.url?scp=84962254984&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962254984&partnerID=8YFLogxK

U2 - 10.1109/TMM.2015.2505908

DO - 10.1109/TMM.2015.2505908

M3 - Article

AN - SCOPUS:84962254984

VL - 18

SP - 273

EP - 286

JO - IEEE Transactions on Multimedia

JF - IEEE Transactions on Multimedia

SN - 1520-9210

IS - 2

M1 - 7347445

ER -