Phase Retrieval Using Conditional Generative Adversarial Networks

TL;DR

This paper introduces a conditional GAN approach for phase retrieval that improves robustness, detail, and noise resilience over existing methods, demonstrating superior performance on Fourier and compressive problems.

Contribution

The paper presents a novel conditional GAN framework that incorporates measurement knowledge, enhancing phase retrieval accuracy and robustness compared to prior neural and projection-based methods.

Findings

Outperforms traditional projection-based methods.

Provides more detailed phase retrieval results.

Robust to noise, suitable for real-world applications.

Abstract

In this paper, we propose the application of conditional generative adversarial networks to solve various phase retrieval problems. We show that including knowledge of the measurement process at training time leads to an optimization at test time that is more robust to initialization than existing approaches involving generative models. In addition, conditioning the generator network on the measurements enables us to achieve much more detailed results. We empirically demonstrate that these advantages provide meaningful solutions to the Fourier and the compressive phase retrieval problem and that our method outperforms well-established projection-based methods as well as existing methods that are based on neural networks. Like other deep learning methods, our approach is very robust to noise and can therefore be very useful for real-world applications.