An international conference connecting people
in CAD research, education and business
Bookmark and Share
Copyright (C) CAD Solutions, LLC. All rights reserved.
Proceedings of CAD'17, 2017, 410-414
Fast and Cross-vendor OpenCL-based Implementation for Voxelization of Triangular Mesh Models

Mohammadreza Faieghi, O. Remus Tutunea-Fatan, Roy Eagleson, Western University, London, Canada

Abstract. Voxel based modelling of geometrical objects is receiving a lot of interest over the recent years. The simple topology of voxel-based models makes them an effective alternative to triangular meshes in many applications, including simulation of material removal, 3D printing, generation of porous surfaces, etc. However, many applications continue to rely solely on triangular mesh models and therefore voxelization is often required to convert them into alternate voxel-based representations. The core of the voxelization process essentially consists of the calculation of the intersection between the triangular facets of the mesh and a 3D grid obtained through the discretization of the bounding box of the tessellated geometry. This process involves numerous iterative procedures and thereby is computationally demanding, particularly for small-sized meshes and/or high resolution voxel grids. To address this challenge, earlier studies made efforts to use the graphics pipeline – that is available in virtually all commercial graphics cards – in order to achieve fast voxelization during rendering passes.         

Keywords. Voxelization, OpenCL, general purpose graphical processing unit (GPGPU), parallel computing

DOI: 10.14733/cadconfP.2017.410-414