An improved GrabCut using a saliency map

Kwang Shik Kim; Yeo Jin Yoon; Mun Cheon Kang; Jee Young Sun; Sung-Jea Ko

doi:10.1109/GCCE.2014.7031272

An improved GrabCut using a saliency map

Kwang Shik Kim, Yeo Jin Yoon, Mun Cheon Kang, Jee Young Sun, Sung-Jea Ko

School of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Citations (Scopus)

Abstract

The GrabCut, which uses the graph-cut iteratively, is popularly used as an interactive image segmentation method since it can produce the globally optimal result. However, since the initialization of the GrabCut is roughly performed by the manual interaction, the accuracy of the segmentation result is not guaranteed when the user defines an inaccurate guide. To solve this problem, in this paper, we present an improved GrabCut method which uses a visual saliency of the target region for the effective initialization. Experimental results demonstrate that the proposed method provides more accurate segmentation results compared with the conventional method.

Original language	English
Title of host publication	2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	317-318
Number of pages	2
ISBN (Print)	9781479951451
DOIs	https://doi.org/10.1109/GCCE.2014.7031272
Publication status	Published - 2014 Feb 3
Event	2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014 - Tokyo, Japan Duration: 2014 Oct 7 → 2014 Oct 10

Other

Other	2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014
Country/Territory	Japan
City	Tokyo
Period	14/10/7 → 14/10/10

Keywords

GrabCut
image segmentation
visual saliency

ASJC Scopus subject areas

Electrical and Electronic Engineering

Access to Document

10.1109/GCCE.2014.7031272

Cite this

Kim, KS, Yoon, YJ, Kang, MC, Sun, JY & Ko, S-J 2014, An improved GrabCut using a saliency map. in 2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014., 7031272, Institute of Electrical and Electronics Engineers Inc., pp. 317-318, 2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014, Tokyo, Japan, 14/10/7. https://doi.org/10.1109/GCCE.2014.7031272

@inproceedings{5e91ec65bb2e48f8a485f222def93621,

title = "An improved GrabCut using a saliency map",

abstract = "The GrabCut, which uses the graph-cut iteratively, is popularly used as an interactive image segmentation method since it can produce the globally optimal result. However, since the initialization of the GrabCut is roughly performed by the manual interaction, the accuracy of the segmentation result is not guaranteed when the user defines an inaccurate guide. To solve this problem, in this paper, we present an improved GrabCut method which uses a visual saliency of the target region for the effective initialization. Experimental results demonstrate that the proposed method provides more accurate segmentation results compared with the conventional method.",

keywords = "GrabCut, image segmentation, visual saliency",

author = "Kim, {Kwang Shik} and Yoon, {Yeo Jin} and Kang, {Mun Cheon} and Sun, {Jee Young} and Sung-Jea Ko",

year = "2014",

month = feb,

day = "3",

doi = "10.1109/GCCE.2014.7031272",

language = "English",

isbn = "9781479951451",

pages = "317--318",

booktitle = "2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014 ; Conference date: 07-10-2014 Through 10-10-2014",

}

TY - GEN

T1 - An improved GrabCut using a saliency map

AU - Kim, Kwang Shik

AU - Yoon, Yeo Jin

AU - Kang, Mun Cheon

AU - Sun, Jee Young

AU - Ko, Sung-Jea

PY - 2014/2/3

Y1 - 2014/2/3

N2 - The GrabCut, which uses the graph-cut iteratively, is popularly used as an interactive image segmentation method since it can produce the globally optimal result. However, since the initialization of the GrabCut is roughly performed by the manual interaction, the accuracy of the segmentation result is not guaranteed when the user defines an inaccurate guide. To solve this problem, in this paper, we present an improved GrabCut method which uses a visual saliency of the target region for the effective initialization. Experimental results demonstrate that the proposed method provides more accurate segmentation results compared with the conventional method.

AB - The GrabCut, which uses the graph-cut iteratively, is popularly used as an interactive image segmentation method since it can produce the globally optimal result. However, since the initialization of the GrabCut is roughly performed by the manual interaction, the accuracy of the segmentation result is not guaranteed when the user defines an inaccurate guide. To solve this problem, in this paper, we present an improved GrabCut method which uses a visual saliency of the target region for the effective initialization. Experimental results demonstrate that the proposed method provides more accurate segmentation results compared with the conventional method.

KW - GrabCut

KW - image segmentation

KW - visual saliency

UR - http://www.scopus.com/inward/record.url?scp=84946686392&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84946686392&partnerID=8YFLogxK

U2 - 10.1109/GCCE.2014.7031272

DO - 10.1109/GCCE.2014.7031272

M3 - Conference contribution

AN - SCOPUS:84946686392

SN - 9781479951451

SP - 317

EP - 318

BT - 2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2014 IEEE 3rd Global Conference on Consumer Electronics, GCCE 2014

Y2 - 7 October 2014 through 10 October 2014

ER -

An improved GrabCut using a saliency map

Abstract

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this