Citation Context

...ark, Macclesfield, Cheshire, SK10 4TG, UK. are exacerbated when the approach is applied to 3D images. Several previous attempts have been made to automate model building [3], [4], [5], [6], [7], [8], =-=[9]-=-. The problem of establishing dense correspondence over a set of training boundaries can be posed as that of defining a parameterization for each of the training set, leading to a dense correspondence...

Citation Context

...e shape parameterization, i.e., by using spherical harmonics [15]. Other approaches include the use of geodesic distance and surface curvature to establish correspondences as suggested by Wang et al. =-=[22]-=-. Recently, Davies et al. [23], [24] proposed a new method for establishing correspondences between shapes by optimizing the quality of the statistical model based on the minimum description length of...

Citation Context

...hors have proposed techniques to find point (landmark) correspondences, but only a few of them have indicated or investigated their applicability in the field of statistical shape models. Wang et al. =-=[2]-=- use a surface registration technique to find 3-D point correspondences based on a metric matching surface-to-surface distance, surface normals, and curvature. The authors suggest that this technique ...

Citation Context

...we show how an ‘optimal’ model can be built by automatically defining correspondences across a training set of 3D shapes. Several previous attempts have been made to build 3D statistical shape mod=-=els [3,4,10,12,13,14,21,25]-=-. The problem of establishing dense correspondence over a set of training shapes can be posed as that of defining a parameterisation for each of the training set, assuming correspondence between equiv...

Citation Context

...ently parameterized points. In this paper we compare methods introduced by Brechbühler[2], Kotcheff[16] and - [8]. A fourth method is based on manually initialized subdivision surfaces similar to Wan=-=g[24]-=-. These methods are presented in more detail in sections 2.1-2.4. Similar approaches have also been proposed e.g. Hill[11] and Meier [18]. Christensen[4], Szeliski[22] and Rueckert[20] describe concep...

Citation Context

...int (ICP) algorithm [2] is a popular method for modeling 3D objects from range data. The classical ICP algorithm rests on a rigid surface assumption. Building on recent work on nonrigid object models =-=[5; 16; 9]-=-, this paper presents an ICP algorithm capable of modeling nonrigid objects, where individual scans may be subject to local deformations. We describe an integrated mathematical framework for simultane...

Citation Context

...orrespondence has often been established using manually defined ‘landmarks’ - a time-consuming, subjective and error-prone process. Several previous attempts have been made to automate model build=-=ing [1, 2, 6, 7, 8, 9, 10, 13]-=- . The problem of establishing dense correspondence over a set of training boundaries can be posed as that of defining a parameterisation for each of the training shapes, leading to an implicit corres...

Citation Context

...ly difficult in 3D. Several approaches have been proposed to face the problem in 3D and to automate the task: 3sA method for 3D shape correspondence using local geometry and geodesics is described in =-=[12]-=- without its application to statistical shape models. It is evaluated by computing residual surface distances. In [13] the shapes are represented by their expansion into a series of elliptical harmoni...