Primal-dual method for continuous max-flow approaches

Ke Wei; Xue Cheng Tai; Tony F. Chan; Shingyu Leung

doi:10.1201/b19241-5

Primal-dual method for continuous max-flow approaches

Ke Wei, Xue Cheng Tai, Tony F. Chan, Shingyu Leung

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

9 Scopus citations

Abstract

We review the continuous max-flow approaches for the variational image segmentation models with piecewise constant representations. The review is conducted by exploring the primal-dual relationships between the continuous min-cut and max-flow problems. In addition, we introduce the parameter free primal-dual method for solving those max-flow problems. Empirical results show that the primal-dual method is competitive to the augmented Lagrangian method.

Original language	English (US)
Title of host publication	Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015
Editors	Joao Manuel R.S. Tavares, R.M. Natal Jorge
Publisher	CRC Press/Balkema
Pages	17-24
Number of pages	8
ISBN (Print)	9781138029262
DOIs	https://doi.org/10.1201/b19241-5
State	Published - 2016
Externally published	Yes
Event	5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015 - Tenerife, Spain Duration: Oct 19 2015 → Oct 21 2015

Publication series

Name	Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015

Conference

Conference	5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015
Country/Territory	Spain
City	Tenerife
Period	10/19/15 → 10/21/15

Bibliographical note

Publisher Copyright:
© 2016 Taylor & Francis Group, London.

ASJC Scopus subject areas

Computer Vision and Pattern Recognition
Biomedical Engineering
Electrical and Electronic Engineering

Access to Document

10.1201/b19241-5

Cite this

Wei, K., Tai, X. C., Chan, T. F., & Leung, S. (2016). Primal-dual method for continuous max-flow approaches. In J. M. R. S. Tavares, & R. M. N. Jorge (Eds.), Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015 (pp. 17-24). (Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015). CRC Press/Balkema. https://doi.org/10.1201/b19241-5

Wei, Ke ; Tai, Xue Cheng ; Chan, Tony F. et al. / Primal-dual method for continuous max-flow approaches. Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015. editor / Joao Manuel R.S. Tavares ; R.M. Natal Jorge. CRC Press/Balkema, 2016. pp. 17-24 (Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015).

@inproceedings{b4e2598dce46423297edb9d840e82ea8,

title = "Primal-dual method for continuous max-flow approaches",

abstract = "We review the continuous max-flow approaches for the variational image segmentation models with piecewise constant representations. The review is conducted by exploring the primal-dual relationships between the continuous min-cut and max-flow problems. In addition, we introduce the parameter free primal-dual method for solving those max-flow problems. Empirical results show that the primal-dual method is competitive to the augmented Lagrangian method.",

author = "Ke Wei and Tai, {Xue Cheng} and Chan, {Tony F.} and Shingyu Leung",

note = "Publisher Copyright: {\textcopyright} 2016 Taylor & Francis Group, London.; 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015 ; Conference date: 19-10-2015 Through 21-10-2015",

year = "2016",

doi = "10.1201/b19241-5",

language = "English (US)",

isbn = "9781138029262",

series = "Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015",

publisher = "CRC Press/Balkema",

pages = "17--24",

editor = "Tavares, {Joao Manuel R.S.} and Jorge, {R.M. Natal}",

booktitle = "Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015",

}

Wei, K, Tai, XC, Chan, TF & Leung, S 2016, Primal-dual method for continuous max-flow approaches. in JMRS Tavares & RMN Jorge (eds), Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015. Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015, CRC Press/Balkema, pp. 17-24, 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015, Tenerife, Spain, 10/19/15. https://doi.org/10.1201/b19241-5

Primal-dual method for continuous max-flow approaches. / Wei, Ke; Tai, Xue Cheng; Chan, Tony F. et al.
Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015. ed. / Joao Manuel R.S. Tavares; R.M. Natal Jorge. CRC Press/Balkema, 2016. p. 17-24 (Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015).

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

TY - GEN

T1 - Primal-dual method for continuous max-flow approaches

AU - Wei, Ke

AU - Tai, Xue Cheng

AU - Chan, Tony F.

AU - Leung, Shingyu

PY - 2016

Y1 - 2016

N2 - We review the continuous max-flow approaches for the variational image segmentation models with piecewise constant representations. The review is conducted by exploring the primal-dual relationships between the continuous min-cut and max-flow problems. In addition, we introduce the parameter free primal-dual method for solving those max-flow problems. Empirical results show that the primal-dual method is competitive to the augmented Lagrangian method.

AB - We review the continuous max-flow approaches for the variational image segmentation models with piecewise constant representations. The review is conducted by exploring the primal-dual relationships between the continuous min-cut and max-flow problems. In addition, we introduce the parameter free primal-dual method for solving those max-flow problems. Empirical results show that the primal-dual method is competitive to the augmented Lagrangian method.

UR - http://www.scopus.com/inward/record.url?scp=84959279547&partnerID=8YFLogxK

U2 - 10.1201/b19241-5

DO - 10.1201/b19241-5

M3 - Conference contribution

AN - SCOPUS:84959279547

SN - 9781138029262

T3 - Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015

SP - 17

EP - 24

BT - Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015

A2 - Tavares, Joao Manuel R.S.

A2 - Jorge, R.M. Natal

PB - CRC Press/Balkema

T2 - 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015

Y2 - 19 October 2015 through 21 October 2015

ER -

Wei K, Tai XC, Chan TF, Leung S. Primal-dual method for continuous max-flow approaches. In Tavares JMRS, Jorge RMN, editors, Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015. CRC Press/Balkema. 2016. p. 17-24. (Computational Vision and Medical Image Processing V - Proceedings of 5th Eccomas Thematic Conference on Computational Vision and Medical Image Processing, VipIMAGE 2015). doi: 10.1201/b19241-5

Primal-dual method for continuous max-flow approaches

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this