Surface feature based mesh segmentation

Mesh segmentation has a variety of applications in product design, reverse engineering, and rapid prototyping fields. This paper presents a novel algorithm of mesh segmentation from original scanning data points, which essentially consists of three steps. Normal based initial decomposing is first performed to recognize plane features. Then we implement further segmentation based on curvature criteria and Gauss mapping, followed by the detection of quadric surface features. The segmentation refinement is finally achieved using B-spline surface fitting technology. The experimental results on many 3D models have demonstrated the effectiveness and robustness of the proposed segmentation method.

Figure optionsDownload high-quality image (53 K)Download as PowerPoint slideHighlights
► We segment meshes using feature detection and B-Spline surface fitting technologies.
► Hierarchical feature recognition makes segmentation more reliable and efficient.
► Geometrical compatibility criteria purposed are effective to mesh segmentation.
► This approach is applicable to the meshes with sharp and smooth features.