18 May, 2021 (Tue)Time:
10:00 amWebinar Link:
SpeakerMiss Tianjiao Zeng
Department of Electrical and Electronic Engineering
The University of Hong Kong
Unlike traditional cameras, lensless imagers don’t directly record a real-world scene on the sensor but mapping to specific patterns and restore the image using numerical reconstruction methods. Such imaging modality has been achieved increasing attention because of its ultra-miniature design. However, the oversimplified approximation on the forward imaging model is imperfect and results in model mismatch error in the reconstruction. To this end, we introduce a deep learning solution which learns a data-driven branch in parallel to unrolled model-based layers for artifact correction and information compensation. Experimental results on real lensless datasets have demonstrated the robustness of the proposed method.
Biography of the speaker:
Tianjiao Zeng received the B.S. degree in electrical engineering from the University of Electronic Science and Technology of China, and the M.S. degree in electrical and computer engineering from Rutgers University. She is currently pursuing a Ph.D. degree from the Department of Electrical and Electronic Engineering, University of Hong Kong. Her research interests include inverse problems, computational imaging, and machine learning.
All are welcome!
OrganizerProf. E.Y.M. Lam
Most seminars are open to the general public, free of charge, unless otherwise stated. Registration is not required. Arrangement for car parking facilities on campus please contact us for details.
For enquiries, please contact:
Department of Electrical and Electronic Engineering,
Room 601, Chow Yei Ching Building,
Pokfulam Road, Hong Kong
Tel: (852) 3917 7093