GAMES Webinar 2020 – 136期(仿真模拟专题) | 胡渊鸣(麻省理工学院), Junbang Liang(University of Maryland, College Park)
【GAMES Webinar 2020-136期】(仿真模拟专题)
报告嘉宾1:胡渊鸣(麻省理工学院)
报告时间:2020年4月23号星期四晚上8:00-8:45(北京时间)
报告题目:DiffTaichi:面向物理模拟的可微编程 (ICLR 2020)
报告摘要:
讲者简介:
麻省理工学院三年级博士生,研究方向是计算机图形学、物理仿真、高性能编程语言与编译器、计算摄影学,在SIGGRAPH/ACM TOG/ICLR/NIPS/CVPR/ICRA发表论文十余篇。2017年本科毕业于清华大学姚班,博士阶段获得Facebook、Adobe、Snap等公司奖学金支持。他主导设计、开发了Taichi编程语言及其优化编译器。
讲者个人主页: http://taichi.graphics/me
报告嘉宾2:Junbang Liang(University of Maryland, College Park)
报告时间:2020年4月23号星期四晚上8:45-9:30(北京时间)
报告题目:Differentiable Cloth Simulation for Inverse Problems
报告摘要:
We propose a differentiable cloth simulator that can be embedded as a layer in deep neural networks. This approach provides an effective, robust framework for modeling cloth dynamics, self-collisions, and contacts. Due to the high dimensionality of the dynamical system in modeling cloth, traditional gradient computation for collision response can become impractical. To address this problem, we propose to compute the gradient directly using QR decomposition of a much smaller matrix. Experimental results indicate that our method can speed up backpropagation by two orders of magnitude. We demonstrate the presented approach on a number of inverse problems, including parameter estimation and motion control for cloth.
讲者简介:
Junbang Liang is a Ph.D. student at the University of Maryland’s Department of Computer Science, working under the supervision of Prof. Ming Lin. His research interest is physics-based cloth simulation and inverse problem for virtual reality applications.
讲者个人主页: cs.umd.edu/~liangjb
主持人简介:
刘天添,微软亚洲研究院网络图形组副研究员。2009年本科毕业于浙江大学,2012年与2018年于宾夕法尼亚大学分别获得硕士与博士学位。研究方向主要是基于物理的仿真以及高效的数值方法。详情请见个人主页:http://tiantianliu.cn/
GAMES主页的“使用教程”中有 “如何观看GAMES Webinar直播?”及“如何加入GAMES微信群?”的信息;
GAMES主页的“资源分享”有往届的直播讲座的视频及PPT等。
观看直播的链接:http://webinar.games-cn.org