DOLPH: Diffusion Models for Phase Retrieval
This work addresses phase retrieval for imaging applications, presenting an incremental improvement by combining existing diffusion models with a specific data-fidelity approach.
The paper tackles the ill-posed problem of phase retrieval by proposing DOLPH, a deep model-based architecture that integrates a diffusion model as an image prior with a nonconvex data-fidelity term, resulting in robust noise handling and the ability to generate multiple candidate solutions from measurements.
Phase retrieval refers to the problem of recovering an image from the magnitudes of its complex-valued linear measurements. Since the problem is ill-posed, the recovery requires prior knowledge on the unknown image. We present DOLPH as a new deep model-based architecture for phase retrieval that integrates an image prior specified using a diffusion model with a nonconvex data-fidelity term for phase retrieval. Diffusion models are a recent class of deep generative models that are relatively easy to train due to their implementation as image denoisers. DOLPH reconstructs high-quality solutions by alternating data-consistency updates with the sampling step of a diffusion model. Our numerical results show the robustness of DOLPH to noise and its ability to generate several candidate solutions given a set of measurements.