【直播】【翔龙鸣凤科学论坛】High order versions of threshold dynamics

活动名称：

High order versions of threshold dynamics

活动时间：

2022年8月1日（周一）上午 10:00 - 11:00

报告嘉宾：

Prof. Selim Esedoglu, University of Michigan

主持人：

Prof. Xiaoping WANG, Presidential Chair Professor, The Chinese University of Hong Kong, Shenzhen

主办单位：

香港中文大学（深圳）理工学院

直播通道

蔻享学术直播间 识别二维码，即可观看直播。

报告人介绍

Prof. Selim Esedoglu

University of Michigan

Selim Esedoglu works on mathematical problems from image processing, computer vision, and materials science. He received his Ph.D. in mathematics from the Courant Institute in 2000. Afterwards, he was a postdoc at the Institute for Mathematics and its Applications of the University of Minnesota between 2000 and 2002, and a CAM Assistant Professor at UCLA between 2002 and 2005. Since 2005, he has been at the University of Michigan, where he is currently a professor of mathematics.

报告简介

Threshold dynamics is an appealing class of algorithms for simulating the motion of interfaces (curves and surfaces), especially under geometric motions such as mean curvature flow, including in the multiphase setting. Among the many applications it has found, a particularly challenging one is simulating the evolution of microstructure in polycrystalline materials. Much like phase field and level set methods, it represents interfaces implicitly, allowing topological changes to take place naturally. It is also unconditionally stable, allowing arbitrarily large time step sizes, constrained only by accuracy considerations. One of its chief drawbacks is low order of accuracy in time.

I will discuss my recent work to find natural high order in time (hence much more accurate) versions of threshold dynamics algorithms, while maintaining all of their original desirable properties.

---

推荐阅读

编辑：王亚琨

蔻享学术 平台介绍

蔻享学术平台，国内领先的一站式科学资源共享平台，依托国内外一流科研院所、高等院校和企业的科研力量，聚焦前沿科学，以优化科研创新环境、传播和服务科学、促进学科交叉融合为宗旨，打造优质学术资源的共享数据平台。

识别二维码，

下载 蔻享APP  查看最新资源数据。

点击阅读原文，查看更多精彩报告！