Point-cut: Interactive image segmentation using point supervision

Changjae Oh; Bumsub Ham; Kwanghoon Sohn

doi:10.1007/978-3-319-54181-5_15

Point-cut: Interactive image segmentation using point supervision

Changjae Oh, Bumsub Ham, Kwanghoon Sohn

Department of Electrical and Electronic Engineering

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

2 Citations (Scopus)

Abstract

Interactive image segmentation is a fundamental task in many applications in graphics, image processing, and computational photography. Many leading methods formulate elaborated energy functionals, achieving high performance with reflecting human’s intention. However, they show limitations in practical usage since user interaction is labor intensive to obtain segments efficiently. We present an interactive segmentation method to handle this problem. Our approach, called point cut, requires minimal point supervision only. To this end, we use off-the-shelf object proposal methods that generate object candidates with high recall. With the single point supervision, foreground appearance can be estimated with high accuracy, and then integrated into a graph cut optimization to generate binary segments. Intensive experiments show that our approach outperforms existing methods for interactive object segmentation both qualitatively and quantitatively.

Original language	English
Title of host publication	Computer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers
Editors	Yoichi Sato, Ko Nishino, Vincent Lepetit, Shang-Hong Lai
Publisher	Springer Verlag
Pages	229-244
Number of pages	16
ISBN (Print)	9783319541808
DOIs	https://doi.org/10.1007/978-3-319-54181-5_15
Publication status	Published - 2017
Event	13th Asian Conference on Computer Vision, ACCV 2016 - Taipei, Taiwan, Province of China Duration: 2016 Nov 20 → 2016 Nov 24

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	10111 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	13th Asian Conference on Computer Vision, ACCV 2016
Country	Taiwan, Province of China
City	Taipei
Period	16/11/20 → 16/11/24

Bibliographical note

Funding Information:
This work was supported by Institute for Information and communications Technology Promotion (IITP) grant funded by the Korea government (MSIP) (No. R0115-15-1007, High quality 2d-to-multiview contents generation from large-scale RGB+D database).

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-319-54181-5_15

Fingerprint Dive into the research topics of 'Point-cut: Interactive image segmentation using point supervision'. Together they form a unique fingerprint.

Cite this

Oh, C., Ham, B., & Sohn, K. (2017). Point-cut: Interactive image segmentation using point supervision. In Y. Sato, K. Nishino, V. Lepetit, & S-H. Lai (Eds.), Computer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers (pp. 229-244). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10111 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-54181-5_15

Oh, Changjae ; Ham, Bumsub ; Sohn, Kwanghoon. / Point-cut : Interactive image segmentation using point supervision. Computer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers. editor / Yoichi Sato ; Ko Nishino ; Vincent Lepetit ; Shang-Hong Lai. Springer Verlag, 2017. pp. 229-244 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ba55df29de68413386bb655da4b666a9,

title = "Point-cut: Interactive image segmentation using point supervision",

abstract = "Interactive image segmentation is a fundamental task in many applications in graphics, image processing, and computational photography. Many leading methods formulate elaborated energy functionals, achieving high performance with reflecting human{\textquoteright}s intention. However, they show limitations in practical usage since user interaction is labor intensive to obtain segments efficiently. We present an interactive segmentation method to handle this problem. Our approach, called point cut, requires minimal point supervision only. To this end, we use off-the-shelf object proposal methods that generate object candidates with high recall. With the single point supervision, foreground appearance can be estimated with high accuracy, and then integrated into a graph cut optimization to generate binary segments. Intensive experiments show that our approach outperforms existing methods for interactive object segmentation both qualitatively and quantitatively.",

author = "Changjae Oh and Bumsub Ham and Kwanghoon Sohn",

note = "Funding Information: This work was supported by Institute for Information and communications Technology Promotion (IITP) grant funded by the Korea government (MSIP) (No. R0115-15-1007, High quality 2d-to-multiview contents generation from large-scale RGB+D database).; 13th Asian Conference on Computer Vision, ACCV 2016 ; Conference date: 20-11-2016 Through 24-11-2016",

year = "2017",

doi = "10.1007/978-3-319-54181-5_15",

language = "English",

isbn = "9783319541808",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "229--244",

editor = "Yoichi Sato and Ko Nishino and Vincent Lepetit and Shang-Hong Lai",

booktitle = "Computer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers",

address = "Germany",

}

Oh, C, Ham, B & Sohn, K 2017, Point-cut: Interactive image segmentation using point supervision. in Y Sato, K Nishino, V Lepetit & S-H Lai (eds), Computer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10111 LNCS, Springer Verlag, pp. 229-244, 13th Asian Conference on Computer Vision, ACCV 2016, Taipei, Taiwan, Province of China, 16/11/20. https://doi.org/10.1007/978-3-319-54181-5_15

Point-cut : Interactive image segmentation using point supervision. / Oh, Changjae; Ham, Bumsub ; Sohn, Kwanghoon.

Computer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers. ed. / Yoichi Sato; Ko Nishino; Vincent Lepetit; Shang-Hong Lai. Springer Verlag, 2017. p. 229-244 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10111 LNCS).

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

TY - GEN

T1 - Point-cut

T2 - 13th Asian Conference on Computer Vision, ACCV 2016

AU - Oh, Changjae

AU - Ham, Bumsub

AU - Sohn, Kwanghoon

N1 - Funding Information: This work was supported by Institute for Information and communications Technology Promotion (IITP) grant funded by the Korea government (MSIP) (No. R0115-15-1007, High quality 2d-to-multiview contents generation from large-scale RGB+D database).

PY - 2017

Y1 - 2017

N2 - Interactive image segmentation is a fundamental task in many applications in graphics, image processing, and computational photography. Many leading methods formulate elaborated energy functionals, achieving high performance with reflecting human’s intention. However, they show limitations in practical usage since user interaction is labor intensive to obtain segments efficiently. We present an interactive segmentation method to handle this problem. Our approach, called point cut, requires minimal point supervision only. To this end, we use off-the-shelf object proposal methods that generate object candidates with high recall. With the single point supervision, foreground appearance can be estimated with high accuracy, and then integrated into a graph cut optimization to generate binary segments. Intensive experiments show that our approach outperforms existing methods for interactive object segmentation both qualitatively and quantitatively.

AB - Interactive image segmentation is a fundamental task in many applications in graphics, image processing, and computational photography. Many leading methods formulate elaborated energy functionals, achieving high performance with reflecting human’s intention. However, they show limitations in practical usage since user interaction is labor intensive to obtain segments efficiently. We present an interactive segmentation method to handle this problem. Our approach, called point cut, requires minimal point supervision only. To this end, we use off-the-shelf object proposal methods that generate object candidates with high recall. With the single point supervision, foreground appearance can be estimated with high accuracy, and then integrated into a graph cut optimization to generate binary segments. Intensive experiments show that our approach outperforms existing methods for interactive object segmentation both qualitatively and quantitatively.

UR - http://www.scopus.com/inward/record.url?scp=85015979645&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015979645&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-54181-5_15

DO - 10.1007/978-3-319-54181-5_15

M3 - Conference contribution

AN - SCOPUS:85015979645

SN - 9783319541808

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 229

EP - 244

BT - Computer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers

A2 - Sato, Yoichi

A2 - Nishino, Ko

A2 - Lepetit, Vincent

A2 - Lai, Shang-Hong

PB - Springer Verlag

Y2 - 20 November 2016 through 24 November 2016

ER -

Oh C, Ham B , Sohn K. Point-cut: Interactive image segmentation using point supervision. In Sato Y, Nishino K, Lepetit V, Lai S-H, editors, Computer Vision - ACCV 2016 - 13th Asian Conference on Computer Vision, Revised Selected Papers. Springer Verlag. 2017. p. 229-244. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-54181-5_15