Parallel RANSAC for Point Cloud Registration

Parallel RANSAC for Point Cloud Registration

In this paper, a project and implementation of the parallel RANSAC algorithm in CUDA architecture for point cloud registration are presented. At the beginning, a serial state of the art method with several heuristic improvements from the literature compared to basic RANSAC is introduced. Subsequentl...

Full description

Bibliographic Details
Main Author: Koguciuk Daniel
Format: Article
Language:English
Published: Sciendo 2017-09-01
Series:Foundations of Computing and Decision Sciences
Subjects:
rgbd
point clouds
registration
cuda
Online Access:https://doi.org/10.1515/fcds-2017-0010
id doaj-ed5ab0a6861e43eeb9a8e2c667cff0ba
record_format Article
spelling doaj-ed5ab0a6861e43eeb9a8e2c667cff0ba2021-09-05T20:44:57ZengSciendoFoundations of Computing and Decision Sciences2300-34052017-09-0142320321710.1515/fcds-2017-0010fcds-2017-0010Parallel RANSAC for Point Cloud RegistrationKoguciuk Daniel0Faculty of Mechatronics, Warsaw University of Technology, Warsaw, PolandIn this paper, a project and implementation of the parallel RANSAC algorithm in CUDA architecture for point cloud registration are presented. At the beginning, a serial state of the art method with several heuristic improvements from the literature compared to basic RANSAC is introduced. Subsequently, its algorithmic parallelization and CUDA implementation details are discussed. The comparative test has proven a significant program execution acceleration. The result is finding of the local coordinate system of the object in the scene in the near real-time conditions. The source code is shared on the Internet as a part of the Heuros system.https://doi.org/10.1515/fcds-2017-0010rgbdpoint cloudsregistrationcuda
collection DOAJ
language English
format Article
sources DOAJ
author Koguciuk Daniel
spellingShingle Koguciuk Daniel
Parallel RANSAC for Point Cloud Registration
Foundations of Computing and Decision Sciences
rgbd
point clouds
registration
cuda
author_facet Koguciuk Daniel
author_sort Koguciuk Daniel
title Parallel RANSAC for Point Cloud Registration
title_short Parallel RANSAC for Point Cloud Registration
title_full Parallel RANSAC for Point Cloud Registration
title_fullStr Parallel RANSAC for Point Cloud Registration
title_full_unstemmed Parallel RANSAC for Point Cloud Registration
title_sort parallel ransac for point cloud registration
publisher Sciendo
series Foundations of Computing and Decision Sciences
issn 2300-3405
publishDate 2017-09-01
description In this paper, a project and implementation of the parallel RANSAC algorithm in CUDA architecture for point cloud registration are presented. At the beginning, a serial state of the art method with several heuristic improvements from the literature compared to basic RANSAC is introduced. Subsequently, its algorithmic parallelization and CUDA implementation details are discussed. The comparative test has proven a significant program execution acceleration. The result is finding of the local coordinate system of the object in the scene in the near real-time conditions. The source code is shared on the Internet as a part of the Heuros system.
topic rgbd
point clouds
registration
cuda
url https://doi.org/10.1515/fcds-2017-0010
work_keys_str_mv AT koguciukdaniel parallelransacforpointcloudregistration
_version_ 1717784806312378368

Similar Items