TU Darmstadt / ULB / TUbiblio

Massively Parallel Implementation of a Multi-View Stereo Algorithm

Beljan, Mate :
Massively Parallel Implementation of a Multi-View Stereo Algorithm.
TU Darmstadt
[Bachelor Thesis] , (2008)

Abstract

This thesis deals with the implementation of a multi-view stereo algorithm in a massively parallel way. The nature of multi-view stereo algorithms favors a parallel approach, as these algorithms operate on multiple images and in these images on multiple pixels. This work focuses on the matching optimization, of a region growing technique, in the multiview stereo approach by Goesele et al. which is the most computationally intensive step of the algorithm. Starting from a sparse 3D point cloud reconstructed using structure-from-motion methods, a nonlinear optimization is employed using a photoconsistency measure. Building upon an existing CPU implementation, an implementation for the matching optimization is written for NVIDIA graphics hardware which is essentially a massively parallel computing device. To access the hardware we make use of the Compute Unified Device Architecture (CUDA), a technology for General-Purpose computation on GPUs (GPGPU). Subsequently the GPU implementation is compared with the CPU version with regard to reconstruction quality and processing time. Furthermore different optimizations for exploiting the GPU architecture (and therefore reducing the processing time) are discussed and evaluated.

Item Type: Bachelor Thesis
Erschienen: 2008
Creators: Beljan, Mate
Title: Massively Parallel Implementation of a Multi-View Stereo Algorithm
Language: English
Abstract:

This thesis deals with the implementation of a multi-view stereo algorithm in a massively parallel way. The nature of multi-view stereo algorithms favors a parallel approach, as these algorithms operate on multiple images and in these images on multiple pixels. This work focuses on the matching optimization, of a region growing technique, in the multiview stereo approach by Goesele et al. which is the most computationally intensive step of the algorithm. Starting from a sparse 3D point cloud reconstructed using structure-from-motion methods, a nonlinear optimization is employed using a photoconsistency measure. Building upon an existing CPU implementation, an implementation for the matching optimization is written for NVIDIA graphics hardware which is essentially a massively parallel computing device. To access the hardware we make use of the Compute Unified Device Architecture (CUDA), a technology for General-Purpose computation on GPUs (GPGPU). Subsequently the GPU implementation is compared with the CPU version with regard to reconstruction quality and processing time. Furthermore different optimizations for exploiting the GPU architecture (and therefore reducing the processing time) are discussed and evaluated.

Uncontrolled Keywords: Forschungsgruppe Capturing Reality (CARE), 3D Scene reconstruction, Community photo collections, General Purpose Computation on Graphics Processing Unit (GPGPU), Compute Unified Device Architecture (CUDA), Multi-view stereo, Parallel processing, Parallelization
Divisions: UNSPECIFIED
Department of Computer Science
Department of Computer Science > Interactive Graphics Systems
Date Deposited: 16 Apr 2018 09:03
Additional Information:

36 p.

Export:

Optionen (nur für Redakteure)

View Item View Item