Salient object detection employing a local tree-structured low-rank representation and foreground consistency

Qiang Zhang, Zhen Huo, Yi Liu, Yunhui Pan, Caifeng Shan, Jungong Han (Corresponding author)

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

2 Citaties (Scopus)

Uittreksel

We propose a local tree-structured low-rank representation (TS-LRR) model to detect salient objects under the complicated background with diverse local regions, which is problematic for most low-rank matrix recovery (LRMR) based salient object detection methods. We first impose a local tree-structured low-rank constraint on the representation coefficients matrix to capture the complicated background. Specifically, a primitive background dictionary is constructed for TS-LRR to promote its background representation ability, and thus enlarge the gap between the salient objects and the background. We then impose a group-sparsity constraint on the sparse error matrix with the intention to ensure the saliency consistency among patches with similar features. At last, a foreground consistency is introduced to identically highlight the distinctive regions within the salient object. Experimental results on three public benchmark datasets demonstrate the effectiveness and superiority of the proposed model over the state-of-the-art methods.

Originele taal-2Engels
Pagina's (van-tot)119-134
Aantal pagina's16
TijdschriftPattern Recognition
Volume92
DOI's
StatusGepubliceerd - 1 aug 2019

Vingerafdruk

Glossaries
Recovery
Object detection

Citeer dit

Zhang, Qiang ; Huo, Zhen ; Liu, Yi ; Pan, Yunhui ; Shan, Caifeng ; Han, Jungong. / Salient object detection employing a local tree-structured low-rank representation and foreground consistency. In: Pattern Recognition. 2019 ; Vol. 92. blz. 119-134.
@article{f3108ac868384308b11021d099c9c163,
title = "Salient object detection employing a local tree-structured low-rank representation and foreground consistency",
abstract = "We propose a local tree-structured low-rank representation (TS-LRR) model to detect salient objects under the complicated background with diverse local regions, which is problematic for most low-rank matrix recovery (LRMR) based salient object detection methods. We first impose a local tree-structured low-rank constraint on the representation coefficients matrix to capture the complicated background. Specifically, a primitive background dictionary is constructed for TS-LRR to promote its background representation ability, and thus enlarge the gap between the salient objects and the background. We then impose a group-sparsity constraint on the sparse error matrix with the intention to ensure the saliency consistency among patches with similar features. At last, a foreground consistency is introduced to identically highlight the distinctive regions within the salient object. Experimental results on three public benchmark datasets demonstrate the effectiveness and superiority of the proposed model over the state-of-the-art methods.",
keywords = "Background dictionary, Foreground consistency, Salient object detection, Structured low-rank representation",
author = "Qiang Zhang and Zhen Huo and Yi Liu and Yunhui Pan and Caifeng Shan and Jungong Han",
year = "2019",
month = "8",
day = "1",
doi = "10.1016/j.patcog.2019.03.023",
language = "English",
volume = "92",
pages = "119--134",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier",

}

Salient object detection employing a local tree-structured low-rank representation and foreground consistency. / Zhang, Qiang; Huo, Zhen; Liu, Yi; Pan, Yunhui; Shan, Caifeng; Han, Jungong (Corresponding author).

In: Pattern Recognition, Vol. 92, 01.08.2019, blz. 119-134.

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademicpeer review

TY - JOUR

T1 - Salient object detection employing a local tree-structured low-rank representation and foreground consistency

AU - Zhang, Qiang

AU - Huo, Zhen

AU - Liu, Yi

AU - Pan, Yunhui

AU - Shan, Caifeng

AU - Han, Jungong

PY - 2019/8/1

Y1 - 2019/8/1

N2 - We propose a local tree-structured low-rank representation (TS-LRR) model to detect salient objects under the complicated background with diverse local regions, which is problematic for most low-rank matrix recovery (LRMR) based salient object detection methods. We first impose a local tree-structured low-rank constraint on the representation coefficients matrix to capture the complicated background. Specifically, a primitive background dictionary is constructed for TS-LRR to promote its background representation ability, and thus enlarge the gap between the salient objects and the background. We then impose a group-sparsity constraint on the sparse error matrix with the intention to ensure the saliency consistency among patches with similar features. At last, a foreground consistency is introduced to identically highlight the distinctive regions within the salient object. Experimental results on three public benchmark datasets demonstrate the effectiveness and superiority of the proposed model over the state-of-the-art methods.

AB - We propose a local tree-structured low-rank representation (TS-LRR) model to detect salient objects under the complicated background with diverse local regions, which is problematic for most low-rank matrix recovery (LRMR) based salient object detection methods. We first impose a local tree-structured low-rank constraint on the representation coefficients matrix to capture the complicated background. Specifically, a primitive background dictionary is constructed for TS-LRR to promote its background representation ability, and thus enlarge the gap between the salient objects and the background. We then impose a group-sparsity constraint on the sparse error matrix with the intention to ensure the saliency consistency among patches with similar features. At last, a foreground consistency is introduced to identically highlight the distinctive regions within the salient object. Experimental results on three public benchmark datasets demonstrate the effectiveness and superiority of the proposed model over the state-of-the-art methods.

KW - Background dictionary

KW - Foreground consistency

KW - Salient object detection

KW - Structured low-rank representation

UR - http://www.scopus.com/inward/record.url?scp=85063527113&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2019.03.023

DO - 10.1016/j.patcog.2019.03.023

M3 - Article

AN - SCOPUS:85063527113

VL - 92

SP - 119

EP - 134

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

ER -