Generalizable Holographic Reconstruction via Amplitude-Only Diffusion Priors

2025-09-16

AI Summary

In coherent imaging, phase retrieval in inline holography is severely ill-posed due to nonlinear amplitude-phase coupling. To address this, we propose an unsupervised diffusion prior method: a diffusion model is trained solely on amplitude-only images, enabling an amplitude-driven prediction-correction sampling framework that decouples amplitude likelihood optimization from phase gradient estimation—thereby eliminating reliance on ground-truth phase labels. Our approach is the first to demonstrate that a single amplitude prior suffices for reconstructing the complex optical field of intricate biological tissues. It achieves high-fidelity phase recovery in both simulations and experiments, generalizing across diverse object morphologies, imaging configurations, and lensless setups. Compared with conventional methods, it significantly improves computational efficiency and broadens applicability for nonlinear inverse problems in quantitative phase imaging.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Non-convex Optimization
• Machine Learning: Unsupervised & Self-Supervised Learning

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance

Abstract

Phase retrieval in inline holography is a fundamental yet ill-posed inverse problem due to the nonlinear coupling between amplitude and phase in coherent imaging. We present a novel off-the-shelf solution that leverages a diffusion model trained solely on object amplitude to recover both amplitude and phase from diffraction intensities. Using a predictor-corrector sampling framework with separate likelihood gradients for amplitude and phase, our method enables complex field reconstruction without requiring ground-truth phase data for training. We validate the proposed approach through extensive simulations and experiments, demonstrating robust generalization across diverse object shapes, imaging system configurations, and modalities, including lensless setups. Notably, a diffusion prior trained on simple amplitude data (e.g., polystyrene beads) successfully reconstructs complex biological tissue structures, highlighting the method's adaptability. This framework provides a cost-effective, generalizable solution for nonlinear inverse problems in computational imaging, and establishes a foundation for broader coherent imaging applications beyond holography.

Problem

• Phase retrieval in inline holography is ill-posed
• Recovering amplitude and phase without ground-truth data
• Solving nonlinear inverse problems in computational imaging

Innovation

• Amplitude-only diffusion model for phase retrieval
• Predictor-corrector sampling with separate gradients
• Generalizable reconstruction without ground-truth phase data