arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2307.07115cs.GR
收藏

内蕴网格简化

Intrinsic Mesh Simplification

Randy Shoemaker, Sam Sartor, Pieter Peers

AI总结:

提出一种内蕴三角剖分简化方法,通过曲率筛选、局部边翻转和重心坐标跟踪移除顶点,并在 Thingi10k 上验证其鲁棒性及对 PDE 求解的影响。

摘要:

本文提出了一种新颖的简化方法,用于从内蕴三角剖分中移除与位于近可展(即高斯曲率有限)曲面和一般曲面上的外蕴顶点相对应的顶点。我们以贪心方式处理所有绝对高斯曲率低于用户所选阈值的内蕴顶点。对于每个顶点,我们反复执行局部内蕴边翻转,直到该顶点达到所需价(内部顶点为三,边界顶点为二),从而可以在内蕴三角剖分中局部移除该顶点及其关联边。每个被移除顶点的内蕴位置都通过(内蕴)重心坐标进行跟踪,这些坐标会被更新以反映内蕴三角剖分的变化。我们在 Thingi10k 数据集上证明了该方法的鲁棒性和有效性,并分析了曲率阈值对偏微分方程(PDE)解的影响。

英文摘要:

This paper presents a novel simplification method for removing vertices from an intrinsic triangulation corresponding to extrinsic vertices lying on near-developable (i.e., with limited Gaussian curvature) and general surfaces. We greedily process all intrinsic vertices with an absolute Gaussian curvature below a user selected threshold. For each vertex, we repeatedly perform local intrinsic edge flips until the vertex reaches the desired valence (three for internal vertices or two for boundary vertices) such that removal of the vertex and incident edges can be locally performed in the intrinsic triangulation. Each removed vertex's intrinsic location is tracked via (intrinsic) barycentric coordinates that are updated to reflect changes in the intrinsic triangulation. We demonstrate the robustness and effectiveness of our method on the Thingi10k dataset and analyze the effect of the curvature threshold on the solutions of PDEs.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑