Acceleration of stereo-matching on multi-core CPU and GPU

Tian, X., Cockshott, P. and Oehler, S. (2014) Acceleration of stereo-matching on multi-core CPU and GPU. In: 16th IEEE International Conference on High Performance Computing and Communications (HPCC 2014), Paris, France, 20-22 Aug 2014,

[img]
Preview
Text
96188.pdf - Published Version

710kB

Publisher's URL: http://conference.hpcc2014.studiocheik.fr/

Abstract

This paper presents an accelerated version of a dense stereo-correspondence algorithm for two different parallelism enabled architectures, multi-core CPU and GPU. The algorithm is part of the vision system developed for a binocular robot-head in the context of the CloPeMa 1 research project. This research project focuses on the conception of a new clothes folding robot with real-time and high resolution requirements for the vision system. The performance analysis shows that the parallelised stereo-matching algorithm has been significantly accelerated, maintaining 12x and 176x speed-up respectively for multi-core CPU and GPU, compared with non-SIMD singlethread CPU. To analyse the origin of the speed-up and gain deeper understanding about the choice of the optimal hardware, the algorithm was broken into key sub-tasks and the performance was tested for four different hardware architectures.

Item Type:Conference Proceedings
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Cockshott, Dr William and Oehler, Ms Susanne
Authors: Tian, X., Cockshott, P., and Oehler, S.
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
College/School:College of Science and Engineering > School of Computing Science
Copyright Holders:Copyright © 2014 The Authors
Publisher Policy:Reproduced with the permission of the authors

University Staff: Request a correction | Enlighten Editors: Update this record