摘要： 提出一种新的利用点云进行曲面重构的算法，该算法基于点云的几何与形状特征，根据点云的几何分布进行分类和按照点的局部形状特征进行分类，对每类点进行局部网格重构，并进行后续处理以修补拓扑和几何错误。实验结果表明，该算法强健有效，能生成高质量的网格，并能较好地保持模型的几何与形状特征。

关键词: 点云, 曲面重构, 分类, 几何, 形状特征

Abstract: A novel mesh reconstruction algorithm using point cloud is presented. The proposed algorithm is based on geometric and shape feature of point cloud. The points are classified according to their geometric distribution, and they are classified according to the local shape feature. The mesh reconstruction is done on different type of points. A post-processing is performed to repair the topological and geometric errors. Experimental results show this algorithm is robust and effective. High quality meshes can be created while geometric and shape features are well preserved.

Key words: point cloud, mesh reconstruction, classification, geometric, shape feature

中图分类号: 

• TP311.52