Shape smoothing with feature preserving weighted filters

Several techniques for arbitrary shape recovery from scanned data attempt to recognize and further regularize shape. For arbitrary shape, one can recognize several shape features which should be guided with global shape parameters. Smoothed curves and surfaces created with subdivision can visually improve recovery when overall smoothness is expected. Local features such as sharp edges are not preserved during smoothing. In this paper we show procedural approach to preserve such features while globally smoothing shape. Weighted filters are applied according to local shape variations. For flat and smooth areas, weighted mean face is dominant. Sharp features are detected with normal difference variation and dominated by nearest face normal. After suggested face normals are calculaced, vertices are moved by simplified version of nonlinear diffusion. Performance of proposed method is compared with other methods for mesh smoothing and edge creasing on "CAD" and "media" models.

[1]  Gabriel Taubin,et al.  Geometric Signal Processing on Polygonal Meshes , 2000, Eurographics.

[2]  Paolo Cignoni,et al.  Metro: Measuring Error on Simplified Surfaces , 1998, Comput. Graph. Forum.

[3]  Konrad Polthier,et al.  Anisotropic Filtering of Non‐Linear Surface Features , 2004, Comput. Graph. Forum.

[4]  Hujun Bao,et al.  A Novel Volume Constrained Smoothing Method for Meshes , 2002, Graph. Model..

[5]  David P. Dobkin,et al.  MAPS: multiresolution adaptive parameterization of surfaces , 1998, SIGGRAPH.

[6]  Martin Rumpf,et al.  Robust feature detection and local classification for surfaces based on moment analysis , 2004, IEEE Transactions on Visualization and Computer Graphics.

[7]  D. Cohen-Or,et al.  Robust moving least-squares fitting with sharp features , 2005, ACM Trans. Graph..

[8]  Joonki Paik,et al.  Triangle mesh-based edge detection and its application to surface segmentation and adaptive surface smoothing , 2002, Proceedings. International Conference on Image Processing.

[9]  Mark Meyer,et al.  Discrete Differential-Geometry Operators for Triangulated 2-Manifolds , 2002, VisMath.

[10]  Mark Meyer,et al.  Implicit fairing of irregular meshes using diffusion and curvature flow , 1999, SIGGRAPH.

[11]  Rafael C. González,et al.  Local Determination of a Moving Contrast Edge , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Yutaka Ohtake,et al.  Mesh smoothing via mean and median filtering applied to face normals , 2002, Geometric Modeling and Processing. Theory and Applications. GMP 2002. Proceedings.

[13]  Yutaka Ohtake,et al.  Mesh regularization and adaptive smoothing , 2001, Comput. Aided Des..

[14]  Yuzhong Shen,et al.  Fuzzy vector median-based surface smoothing , 2004, IEEE Transactions on Visualization and Computer Graphics.

[15]  Szymon Rusinkiewicz,et al.  Estimating curvatures and their derivatives on triangle meshes , 2004, Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004..

[16]  Gabriel Taubin,et al.  A signal processing approach to fair surface design , 1995, SIGGRAPH.

[17]  G. Taubin.  LINEAR ANISOTROPIC MESH FILTERING , 2001 .

[18]  Kuo-Young Cheng,et al.  A sharpness dependent filter for mesh smoothing , 2005, Comput. Aided Geom. Des..

[19]  Zoë J. Wood,et al.  Topological Noise Removal , 2001, Graphics Interface.