A highly parallel design of image surface layout recovering on GPGPU

Guan Ru Li; Bo-Cheng Lai

doi:10.1109/VLSI-DAT.2012.6212628

A highly parallel design of image surface layout recovering on GPGPU

Guan Ru Li^*, Bo-Cheng Lai

^*Corresponding author for this work

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

Abstract

Surface layout recovering helps computers understand the intricate information in an image by assigning local segments to different geometric classes. It greatly reduces the complexity of the following-up image processing and is widely used in various computer vision applications. However, the algorithm walks through every image pixel and imposes intensive computation requirement. Through comprehensive analysis on the execution behavior, this paper identifies significant parallelism inherent in the algorithm. With careful concerns on both multi-threaded software and parallel hardware, the optimized parallel design on a modern GPGPU has reached an average of 10.7X performance enhancement.

Original language	English
Title of host publication	2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers
DOIs	https://doi.org/10.1109/VLSI-DAT.2012.6212628
State	Published - 25 Jul 2012
Event	2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Hsinchu, Taiwan Duration: 23 Apr 2012 → 25 Apr 2012

Publication series

Name	2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers

Conference

Conference	2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012
Country/Territory	Taiwan
City	Hsinchu
Period	23/04/12 → 25/04/12

Access to Document

10.1109/VLSI-DAT.2012.6212628

Cite this

Li, G. R., & Lai, B-C. (2012). A highly parallel design of image surface layout recovering on GPGPU. In 2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers [6212628] (2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers). https://doi.org/10.1109/VLSI-DAT.2012.6212628

@inproceedings{d0073c19502c40f89e4ac727d5420f4a,

title = "A highly parallel design of image surface layout recovering on GPGPU",

abstract = "Surface layout recovering helps computers understand the intricate information in an image by assigning local segments to different geometric classes. It greatly reduces the complexity of the following-up image processing and is widely used in various computer vision applications. However, the algorithm walks through every image pixel and imposes intensive computation requirement. Through comprehensive analysis on the execution behavior, this paper identifies significant parallelism inherent in the algorithm. With careful concerns on both multi-threaded software and parallel hardware, the optimized parallel design on a modern GPGPU has reached an average of 10.7X performance enhancement.",

author = "Li, {Guan Ru} and Bo-Cheng Lai",

year = "2012",

month = jul,

day = "25",

doi = "10.1109/VLSI-DAT.2012.6212628",

language = "English",

isbn = "9781457720819",

series = "2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers",

booktitle = "2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers",

note = "null ; Conference date: 23-04-2012 Through 25-04-2012",

}

Li, GR & Lai, B-C 2012, A highly parallel design of image surface layout recovering on GPGPU. in 2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers., 6212628, 2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers, 2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012, Hsinchu, Taiwan, 23/04/12. https://doi.org/10.1109/VLSI-DAT.2012.6212628

A highly parallel design of image surface layout recovering on GPGPU. / Li, Guan Ru; Lai, Bo-Cheng.

2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers. 2012. 6212628 (2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers).

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

TY - GEN

T1 - A highly parallel design of image surface layout recovering on GPGPU

AU - Li, Guan Ru

AU - Lai, Bo-Cheng

PY - 2012/7/25

Y1 - 2012/7/25

N2 - Surface layout recovering helps computers understand the intricate information in an image by assigning local segments to different geometric classes. It greatly reduces the complexity of the following-up image processing and is widely used in various computer vision applications. However, the algorithm walks through every image pixel and imposes intensive computation requirement. Through comprehensive analysis on the execution behavior, this paper identifies significant parallelism inherent in the algorithm. With careful concerns on both multi-threaded software and parallel hardware, the optimized parallel design on a modern GPGPU has reached an average of 10.7X performance enhancement.

AB - Surface layout recovering helps computers understand the intricate information in an image by assigning local segments to different geometric classes. It greatly reduces the complexity of the following-up image processing and is widely used in various computer vision applications. However, the algorithm walks through every image pixel and imposes intensive computation requirement. Through comprehensive analysis on the execution behavior, this paper identifies significant parallelism inherent in the algorithm. With careful concerns on both multi-threaded software and parallel hardware, the optimized parallel design on a modern GPGPU has reached an average of 10.7X performance enhancement.

UR - http://www.scopus.com/inward/record.url?scp=84864032897&partnerID=8YFLogxK

U2 - 10.1109/VLSI-DAT.2012.6212628

DO - 10.1109/VLSI-DAT.2012.6212628

M3 - Conference contribution

AN - SCOPUS:84864032897

SN - 9781457720819

T3 - 2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers

BT - 2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers

Y2 - 23 April 2012 through 25 April 2012

ER -

Li GR, Lai B-C. A highly parallel design of image surface layout recovering on GPGPU. In 2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers. 2012. 6212628. (2012 International Symposium on VLSI Design, Automation and Test, VLSI-DAT 2012 - Proceedings of Technical Papers). doi: 10.1109/VLSI-DAT.2012.6212628

A highly parallel design of image surface layout recovering on GPGPU

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this