Deep Ptych: Subsampled Fourier Ptychography using Generative Priors
This addresses the problem of efficient and robust image reconstruction in computational imaging for researchers and practitioners, representing an incremental improvement with a novel regularization approach.
The paper tackles the ill-posed Fourier ptychography problem by using generative models as priors, resulting in a method called Deep Ptych that achieves higher quality reconstructions and better noise robustness with far fewer samples compared to existing techniques.
This paper proposes a novel framework to regularize the highly ill-posed and non-linear Fourier ptychography problem using generative models. We demonstrate experimentally that our proposed algorithm, Deep Ptych, outperforms the existing Fourier ptychography techniques, in terms of quality of reconstruction and robustness against noise, using far fewer samples. We further modify the proposed approach to allow the generative model to explore solutions outside the range, leading to improved performance.