Implicit surface reconstruction of 3D objects: Preconditioned iterative methods for the CSRBF‐type linear systems

Abstract

AbstractWe consider a surface reconstruction problem of a 3D object. The surface can be reconstructed as an implicit function, which is determined by solving a linear system derived from the input scattered data. Here, we use a CSRBF (Compactly‐Supported Radial Basis Functions)‐based method which leads to a linear system with a sparse matrix. In this paper, we describe preconditioned iterative methods which are efficient especially for the linear system with sparse and large matrix. Moreover, an appropriate combination of preconditioning strategies and iterative methods for this type of system is shown as a result of numerical experiments. © 2005 Wiley Periodicals, Inc. Electron Comm Jpn Pt 3, 89(4): 11–20, 2006; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjc.20197