三维散乱点云快速曲面重建算法

提出了一种基于Delaunay三角剖分的三维散乱点云快速曲面重建算法。算法首先计算点云的Delaunay三角剖分, 从Delaunay四面体提取初始三角网格, 根据Voronoi体元的特征构造优先队列并生成种子三角网格, 然后通过区域生长的方式进行流形提取。实验结果表明, 该算法可以高效、稳定地重构具有复杂拓扑结构、非封闭曲面甚至是非均匀采样的点云数据。与传统的基于Delaunay的方法比较, 该算法仅需要进行一次Delaunay三角剖分, 无须极点的计算, 因此算法的重构速度快。

Fast surface reconstruction algorithm for 3D scattered points

This paper presented a fast surface reconstruction algorithm from scattered points based on Delaunay. It began with computing the Delaunay triangulation of the sampling point sets, extracted the primary triangles from Delaunay tetrahedrons, constructed priority queue based on the feature of Voronoi cell, selected seed triangles from the priority queue, and then extracted manifold surface by region-growing. Experimental results show that this method is robust and effective on handling surfaces with complex topology, boundaries, and even non-uniform sample points. Compared with the traditional approach based on Delaunay, the proposed algorithm requires only one-pass Delaunay computation and needs no poles computation, so it is more efficient.