报告时间：2025年7月21日（星期一）14:20
报告地点：翡翠湖校区科教楼A1104室
报 告 人：Dr. Yilin Gao
工作单位：University of Waterloo
举办单位：计算机与信息学院（人工智能学院）
报告简介：
Lie Algebras and Lie Groups form the mathematical theory of infinitesimal and differentiable transformations. Though often viewed as abstract, these concepts hold practical significance in computer vision. This seminar will offer a concise and accessible introduction to applied Lie Groups and Lie Algebras in Computer Vision. We will explore how Lie Algebra underpins homography in image transformations. We will examine its role in pose optimization within Simultaneous Localization and Mapping (SLAM). The session will close with a brief explanation of how Lie Algebra is applied in the recent VGGT-SLAM paper.
报告人简介：
Dr. Yilin Gao is a Postdoctoral Fellow at the University of Waterloo. In 2025, he completed his PhD in Systems Design Engineering at the University of Waterloo with a thesis titled: “Towards Urban Digital Twins with Gaussian Splatting, Large-Language-Models, and Cloud Mapping Services”. He received his MSc degree in physics from the University of Victoria in 2020, researching particle bunch electrodynamics inside particle accelerators using Lie Algebra. His research interests include 2D/3D computer vision and large language models applied to remote sensing and Geographic Information Systems (GIS). He has published over 30 peer-reviewed publications, including over 20 in top-tier remote sensing journals.