Computing Medial Axis Transform with Feature Preservation via Restricted Power Diagram

报告题目：Computing Medial Axis Transform with Feature Preservation via Restricted Power Diagram

报告摘要：The medial axis is a fundamental geometric structure and has been widely used in approximating, simplifying, and analyzing shapes. Many applications, especially related to CAD/CAM, often need the medial axis to be properly approximated near the sharp lines and corners. However, feature preservation is a long-lasting problem for medial axis transform (MAT) since the inherent under-sampling in the vicinity of features. We have observed that medial features play a significant role in guaranteeing many topological and geometric properties of medial axis. In this talk, I will introduce our novel framework for computing the medial axis transform of 3D shapes while preserving their medial features via restricted power diagram (RPD). Compared with existing sampling-based or voxel-based methods, our method is the first one that can preserve not only external features but also internal features of medial axes.

讲者简介：王宁娜，本科毕业于吉林大学，获计算数学学士学位，硕士毕业于卡内基梅隆大学，获计算机科学硕士学位。现博士就读于德克萨斯大学达拉斯分校，主要研究方向为图形学的几何建模与处理，发表SIGRAPH Asia 1篇、CCF-B类1篇。曾担任Booking.com (缤客)荷兰总部高级软件工程师。

时间：2022年11月4日(周五)上午10:00-11:00

腾讯会议：360554535