GAMES Webinar 2020 – 144期(可视化专题) | Chaoli Wang(University of Notre Dame)
GAMES Webinar 2020-144期】(可视化专题)
报告嘉宾：Chaoli Wang(University of Notre Dame)
报告题目：Representation Learning and Super-Resolution Generation for Scientific Visualization
As an indispensable branch of visualization, scientific visualization has the most extended history of research and development dated back to the early 1990s. Over the past decade, while we have witnessed the tremendous growth in information visualization and visual analytics, scientific visualization still plays a central role in many areas of natural and physical sciences and engineering of national interest and priority. In these areas, domain scientists run large-scale scientific simulations and produce three-dimensional scalar and vector field data that are often time-varying and multivariate, which demand effective and efficient solutions for analysis and visualization.
In this talk, I will mainly focus on two of our recent works that explore the use of deep learning solutions for representation learning (FlowNet) and super-resolution generation (TSR-TVD). FlowNet designs an autoencoder for learning latent feature representations from flow lines or flow surfaces for clustering and selection. TSR-TVD employs a recurrent generative network to generate temporal super-resolution of time-varying volumetric data. I will also briefly discuss our ongoing works that improve and expand representation learning and super-resolution generation. In the end, I will point out the tremendous opportunities for visualization researchers in the wake of the rise of artificial intelligence and machine learning.
Chaoli Wang is an associate professor of computer science and engineering at the University of Notre Dame. He holds a Ph.D. degree in computer and information science from The Ohio State University. His main research interests are data visualization and visual analytics, specifically on the topics of time-varying multivariate data visualization, flow visualization, graph visualization, information-theoretic algorithms, graph-based techniques, and deep learning solutions for big data analytics. He has published more than 90 peer-reviewed journal and conference papers, including more than 20 IEEE VIS and IEEE TVCG papers. He is a recipient of the U.S. National Science Foundation CAREER Award and multiple best paper awards. He has served numerous rounds on the program committees of IEEE VIS, EuroVis, and IEEE PacificVis, and served as a co-chair for IEEE VIS Tutorials and Workshops, a paper co-chair for ChinaVis, IEEE LDAV, and IEEE PacificVis. He is an associate editor of IEEE Transactions on Visualization and Computer Graphics (IEEE TVCG).
毕重科， 天津大学智能与计算学部副教授。2012年于东京大学获理学博士学位，2012年至2016年在日本理化学研究所担任研究员，2016年加入天津大学。主要研究方向为科学可视化，高性能计算，现任可视化与可视分析专业委员会常务委员。在日本理化学研究所期间，负责当时世界排名第一的超级计算机K Computer的原位可视化系统的研发工作；现在主要研究方向为国内超级计算机上的大规模数值仿真的可视分析。详情请见个人主页http://www.bichongke.com/
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；