Abstract
A two-step deep learning approach is proposed for computational imaging. This approach is robust to image model mismatches since physics priors are not needed and mitigates over-parameterization by training the network in 2 steps.
© 2021 The Author(s)
PDF Article | Presentation VideoMore Like This
George Barbastathis, Alexandre Goy, Kwabena Arthur, Mo Deng, and Shuai Li
ITh4B.1 Integrated Photonics Research, Silicon and Nanophotonics (IPR) 2018
Xin Yuan
IW4A.1 Imaging Systems and Applications (IS) 2021
Zhen Guo, Abraham Levitan, George Barbastathis, and Riccardo Comin
CTh7A.6 Computational Optical Sensing and Imaging (COSI) 2021