Multi-domain, higher order level set scheme for 3D image segmentation on the GPU

Ojaswa Sharma; Qin Zhang; François Anton; Chandrajit Bajaj

doi:10.1109/CVPR.2010.5539902

Multi-domain, higher order level set scheme for 3D image segmentation on the GPU

Ojaswa Sharma, Qin Zhang, François Anton, Chandrajit Bajaj

University of Texas at Dallas

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

Abstract

Level set method based segmentation provides an efficient tool for topological and geometrical shape handling. Conventional level set surfaces are only $C^0$ continuous since the level set evolution involves linear interpolation to compute derivatives. Bajaj et al. present a higher order method to evaluate level set surfaces that are $C^2$ continuous, but are slow due to high computational burden. In this paper, we provide a higher order GPU based solver for fast and efficient segmentation of large volumetric images. We also extend the higher order method to multi-domain segmentation. Our streaming solver is efficient in memory usage.

Original language	English
Title of host publication	Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition
Publisher	IEEE
Publication date	2010
Pages	1063-6919
ISBN (Print)	978-1-4244-6984-0
DOIs	https://doi.org/10.1109/CVPR.2010.5539902
Publication status	Published - 2010
Event	2010 IEEE Conference on Computer Vision and Pattern Recognition - San Francisco, United States Duration: 13 Jun 2010 → 18 Jun 2010 Conference number: 23 https://ieeexplore.ieee.org/xpl/conhome/5521876/proceeding

Conference

Conference	2010 IEEE Conference on Computer Vision and Pattern Recognition
Number	23
Country/Territory	United States
City	San Francisco
Period	13/06/2010 → 18/06/2010
Internet address	https://ieeexplore.ieee.org/xpl/conhome/5521876/proceeding

Keywords

Volume segmentation
level set method
CUDA
GPU computing
Higher order segmentation

Access to Document

10.1109/CVPR.2010.5539902

OpenUrl availability

Full text

Cite this

@inproceedings{0859813786f74afc85ddbe62de9d5ee1,

title = "Multi-domain, higher order level set scheme for 3D image segmentation on the GPU",

abstract = "Level set method based segmentation provides an efficient tool for topological and geometrical shape handling. Conventional level set surfaces are only $C^0$ continuous since the level set evolution involves linear interpolation to compute derivatives. Bajaj et al. present a higher order method to evaluate level set surfaces that are $C^2$ continuous, but are slow due to high computational burden. In this paper, we provide a higher order GPU based solver for fast and efficient segmentation of large volumetric images. We also extend the higher order method to multi-domain segmentation. Our streaming solver is efficient in memory usage.",

keywords = "Volume segmentation, level set method, CUDA, GPU computing, Higher order segmentation",

author = "Ojaswa Sharma and Qin Zhang and Fran{\c c}ois Anton and Chandrajit Bajaj",

year = "2010",

doi = "10.1109/CVPR.2010.5539902",

language = "English",

isbn = "978-1-4244-6984-0",

pages = "1063--6919",

booktitle = "Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition",

publisher = "IEEE",

address = "United States",

note = "2010 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2010 ; Conference date: 13-06-2010 Through 18-06-2010",

url = "https://ieeexplore.ieee.org/xpl/conhome/5521876/proceeding",

}

Sharma, O, Zhang, Q, Anton, F & Bajaj, C 2010, Multi-domain, higher order level set scheme for 3D image segmentation on the GPU. in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE, pp. 1063-6919, 2010 IEEE Conference on Computer Vision and Pattern Recognition, San Francisco, California, United States, 13/06/2010. https://doi.org/10.1109/CVPR.2010.5539902

TY - GEN

T1 - Multi-domain, higher order level set scheme for 3D image segmentation on the GPU

AU - Sharma, Ojaswa

AU - Zhang, Qin

AU - Anton, François

AU - Bajaj, Chandrajit

N1 - Conference code: 23

PY - 2010

Y1 - 2010

N2 - Level set method based segmentation provides an efficient tool for topological and geometrical shape handling. Conventional level set surfaces are only $C^0$ continuous since the level set evolution involves linear interpolation to compute derivatives. Bajaj et al. present a higher order method to evaluate level set surfaces that are $C^2$ continuous, but are slow due to high computational burden. In this paper, we provide a higher order GPU based solver for fast and efficient segmentation of large volumetric images. We also extend the higher order method to multi-domain segmentation. Our streaming solver is efficient in memory usage.

AB - Level set method based segmentation provides an efficient tool for topological and geometrical shape handling. Conventional level set surfaces are only $C^0$ continuous since the level set evolution involves linear interpolation to compute derivatives. Bajaj et al. present a higher order method to evaluate level set surfaces that are $C^2$ continuous, but are slow due to high computational burden. In this paper, we provide a higher order GPU based solver for fast and efficient segmentation of large volumetric images. We also extend the higher order method to multi-domain segmentation. Our streaming solver is efficient in memory usage.

KW - Volume segmentation

KW - level set method

KW - CUDA

KW - GPU computing

KW - Higher order segmentation

U2 - 10.1109/CVPR.2010.5539902

DO - 10.1109/CVPR.2010.5539902

M3 - Article in proceedings

SN - 978-1-4244-6984-0

SP - 1063

EP - 6919

BT - Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

PB - IEEE

T2 - 2010 IEEE Conference on Computer Vision and Pattern Recognition

Y2 - 13 June 2010 through 18 June 2010

ER -

Multi-domain, higher order level set scheme for 3D image segmentation on the GPU

Abstract

Conference

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this