[Abstract]

Building a machine capable of understanding and generating human motion is a fundamental research topic across various fields, including computer graphics, computer vision, robotics, and medicine. It also has numerous practical applications, such as in entertainment, the Metaverse, human-robot collaboration, and gait analysis for medical diagnosis. The talk will begin by outlining the key advantages of using physics simulation for understanding and generating human motion. It will then explore several recent approaches for generating 3D motion from diverse input sources, including videos, IMUs, 3D markers, manual keyframes, and text. If time permits, the talk will conclude with a discussion of remaining challenges and potential directions for future research.

[Biography]

Jungdam Won is an assistant professor in the Department of Computer Science and Engineering at Seoul National University (SNU) in Korea. Prior to joining SNU, he worked as a research scientist at Meta AI (formerly Facebook) in the Creative AI team led by Jessica Hodgins. Before that, he was a postdoctoral researcher at the Movement Research Lab at SNU, under the supervision of Prof. Jehee Lee. He received his Ph.D. and B.S. from SNU in 2017 and 2011, respectively. As part of his academic service, he has served on the technical papers and courses committees for ACM SIGGRAPH and SIGGRAPH Asia, as well as on the technical papers committees for Eurographics, Pacific Graphics, Motion, Interaction and Games, CASA, and IEEE AIVR. His current research interests include designing controllers for virtual and physical agents, modeling interactions between multiple agents, developing user-friendly methods that bridge the gap between users and their virtual personas, and biomechanical modeling of humans for medical applications.