•   When: Monday, April 05, 2021 from 11:00 AM to 12:00 PM
•   Speakers: Jinwei Ye, Assistant Professor, Division of Computer Science and Engineering, Louisiana State University
•   Location: ZOOM
•   Export to iCal

Abstract:

Transparent or mirroring objects are “invisible” as they borrow appearances from the surroundings. Recovering the three-dimensional (3D) structure of these invisible objects has been a notoriously difficult problem in computer vision for decades. This because their visual characteristics break down fundamental assumptions of classical 3D vision algorithms. In this talk, I’ll present several computational imaging solutions, that leverage novel cameras and displays, for recovering the external and internal 3D structures of invisible objects. Specifically, I’ll talk about using novel cameras and deep neural networks for recovering and visualizing the wavefront and internal flows of dynamic fluid. I’ll also showcase a polarization-encoded display for the 3D reconstruction of mirroring surfaces. I’ll conclude my talk with future trends in computational imaging techniques. 

Short Bio:

Jinwei Ye is an Assistant Professor in the Division of Computer Science and Engineering at Louisiana State University. She received her Ph.D. in Computer Science from the University of Delaware in 2014. She leads the Imaging and Vision (IV) lab at LSU. Her research interests lie in the areas of computational photography, computer vision, and computer graphics. She receives the LSU Leveraging Innovation for Technology Transfer (LIFT2) award and the NSF CISE Research Initiation Initiative (CRII) award. She serves as an area chair for CVPR 2021 and a local chair for CVPR 2022.