Diffusion models for polarimetric reconstruction of circumstellar environments

2025/10/14 by Quentin Villegas, Laurence Denneulin, Villegas, Quentin +13
Physics and Astronomy · Earth and Planetary Sciences · Biochemistry, Genetics and Molecular Biology · #Planetary Science and Exploration #Geophysics and Gravity Measurements #Geomagnetism and Paleomagnetism Studies

paper · pdf · doi:10.48550/arxiv.2510.12853

Abstract

In this paper, we propose an approach combining diffusion models and inverse problems for the reconstruction of circumstellar disk images. Our method builds upon the Rhapsodie framework for polarimetric imaging, substituting its classical prior with a diffusion model trained on synthetic data. Our formulation explicitly incorporates stellar leakage while efficiently handling missing data and high level noise inherent to high-contrast polarimetric imaging. Experiments show significant improvement over conventional methods within our framework of assumptions, opening new perspectives for studying circumstellar environments.

Citations

• A Survey on Diffusion Models for Inverse Problems
• Solving Inverse Problems with Latent Diffusion Models via Hard Data Consistency
• Robust Compressed Sensing MRI with Deep Generative Priors
• Diffusion Models Beat GANs on Image Synthesis
• Score-Based Generative Modeling through Stochastic Differential Equations
• Plug-and-Play Image Restoration with Deep Denoiser Prior
• Denoising Diffusion Probabilistic Models

Related

• Astronomical high-contrast imaging of circumstellar disks: MUSTARD inverse-problem versus PCA-based methods
• Reference-less algorithm for circumstellar disks imaging
• Resolution and calibration effects in high contrast polarimetric imaging of circumstellar scattering regions
• Bayesian Deconvolution of Astronomical Images with Diffusion Models: Quantifying Prior-Driven Features in Reconstructions
• REXPACO ASDI: Joint unmixing and deconvolution of the circumstellar environment by angular and spectral differential imaging