PrNet: Efficient and Robust Phase Retrieval via Stochastic Refinement

Kaya M. O., ÖKTEM SEVEN S. F.

35th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2025, İstanbul, Turkey, 31 August - 03 September 2025, (Full Text) identifier

  • Publication Type: Conference Paper / Full Text
  • Doi Number: 10.1109/mlsp62443.2025.11204288
  • City: İstanbul
  • Country: Turkey
  • Keywords: computational imaging, deep learning, diffusion models, inverse problems, Phase retrieval
  • Middle East Technical University Affiliated: Yes

Abstract

Phase retrieval is a fundamental inverse problem that arises in many scientific and engineering disciplines, where the goal is to reconstruct a signal from intensity-only measurements. In this work, we develop prNet, a novel phase retrieval framework that stochastically refines initial reconstructions with learned denoising and model-based updates. Our framework combines Langevin dynamics-based posterior sampling, adaptive noise schedule learning, warm-start initialization from classical solvers, and a progressive training strategy inspired by algorithm unrolling. By considering the perception-distortion tradeoff, our method also mitigates the over-smoothing effects commonly observed in prior approaches and enables reconstructions with fine details while minimizing artifacts. Extensive experiments demonstrate that our approach achieves state-of-the-art performance in both efficiency and reconstruction quality.