prNet: Data-Driven Phase Retrieval via Stochastic Refinement

TL;DR

prNet introduces a stochastic, Langevin dynamics-based framework for phase retrieval that balances measurement fidelity and perceptual quality, outperforming traditional methods in multiple benchmarks.

Contribution

The paper presents a novel, data-driven phase retrieval approach using Langevin sampling, incorporating learned denoising and model-based updates for improved perceptual and fidelity tradeoffs.

Findings

Achieves state-of-the-art results in phase retrieval benchmarks.

Balances distortion and perceptual quality effectively.

Provides multiple variants with increasing complexity and performance.

Abstract

Phase retrieval is an ill-posed inverse problem in which classical and deep learning-based methods struggle to jointly achieve measurement fidelity and perceptual realism. We propose a novel framework for phase retrieval that leverages Langevin dynamics to enable efficient posterior sampling, yielding reconstructions that explicitly balance distortion and perceptual quality. Unlike conventional approaches that prioritize pixel-wise accuracy, our methods navigate the perception-distortion tradeoff through a principled combination of stochastic sampling, learned denoising, and model-based updates. The framework comprises three variants of increasing complexity, integrating theoretically grounded Langevin inference, adaptive noise schedule learning, parallel reconstruction sampling, and warm-start initialization from classical solvers. Extensive experiments demonstrate that our methods achieve state-of-the-art performance across multiple benchmarks, both in terms of fidelity and perceptual quality. The source code and trained models are available at https://github.com/METU-SPACE-Lab/prNet-for-Phase-Retrieval