Title: Robust Parameter Estimation in Computer Vision
Speaker: Huu Le - Chalmers University of Technology, Sweden
Date/Time: Wednesday, Jan 08 2020 - 03:00 pm (GMT + 7)
About the Speaker:
Dr Huu Le is currently a Postdoctoral Research Fellow at the Department of Electrical Engineering, Chalmers University of Technology, Sweden. He obtained his Ph.D. in Computer Science from the School of Computer Science, University of Adelaide, Australia. His research focuses on developing new algorithms for robust geometric estimation in Computer Vision. Before joining Chalmers, he was a Postdoctoral Researcher at Queensland University of Technology, working on compression algorithms for visual place recognition.
Robust model fitting plays a crucial role in a wide variety of computer vision applications, especially 3D vision tasks such as structure from motion (SfM), rigid/non-rigid point cloud alignment. It is usually required to handle data contaminated by a large fraction of outliers. However, solving robust estimation for large-scale data remains a challenging task due to the highly non-convex nature of the underlying problem.
In this talk, I would like to give a quick overview of robust parameter estimation in computer vision and review existing methods that are commonly employed to address this problem. I will then present recent algorithmic developments that our group has developed during the past few years and future research directions to tackle this challenging problem.