Abstract
In many computational imaging applications, only sparse measurements on a source are possible, hindering reconstruction. We present a probabilistic, self-supervised method, the physics-informed variational autoencoder (P-VAE), that jointly reconstructs many sources, each with sparse measurements.
© 2023 The Author(s)
PDF Article | Presentation VideoMore Like This
Jiapeng Liu, Muralidhar M. Balaji, Christopher A. Metzler, M. Salman Asif, and Prasanna Rangarajan
CTh5A.2 Computational Optical Sensing and Imaging (COSI) 2021
Yolanda Hu, Andrew Olsen, Jan Funke, Srinivas Turaga, and Vidya Ganapati
CF1D.8 Computational Optical Sensing and Imaging (COSI) 2022
Teemu Sahlström and Tanja Tarvainen
1263108 European Conference on Biomedical Optics (ECBO) 2023