{"ID":2895217,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.09608","arxiv_id":"2507.09608","title":"prNet: Data-Driven Phase Retrieval via Stochastic Refinement","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","short_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 t...","url_abs":"https://arxiv.org/abs/2507.09608","url_pdf":"https://arxiv.org/pdf/2507.09608v2","authors":"[\"Mehmet Onurcan Kaya\",\"Figen S. Oktem\"]","published":"2025-07-13T12:25:06Z","proceeding":"eess.IV","tasks":"[\"eess.IV\",\"cs.CV\"]","methods":"[]","has_code":false,"code_links":[{"ID":612168,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2895217,"paper_url":"https://arxiv.org/abs/2507.09608","paper_title":"prNet: Data-Driven Phase Retrieval via Stochastic Refinement","repo_url":"https://github.com/METU-SPACE-Lab/prNet-for-Phase-Retrieval","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}