Diffusion Models for Polarimetric Reconstruction of Circumstellar Environments in Correlated Speckle Noise
Abstract
High-contrast polarimetric imaging of circumstellar disks is severely limited by stellar leakage and speckle noise. Building upon the RHAPSODIE framework for polarimetric inverse problems, diffusion-based methods have shown promising results by replacing classical regularization with learned priors. However, RHAPSODIE's white noise assumption fails to capture the spatial correlation structure of atmospheric and instrumental speckles. We extend this framework to handle speckle fluctuations, which are modeled as a Gaussian field with a stationary covariance matrix with a Gaussian spectral density. To efficiently apply the dense precision matrix that includes correlated speckle and heteroskedastic photon noise, we develop a dedicated preconditioner for conjugate gradient iterations. The diffusion prior is trained exclusively on disk-only data. While limitations remain at very low SNRs, our method significantly outperforms Tikhonov regularization and competes favorably with a Plug-and-Play baseline in realistic regimes, successfully recovering morphological features masked by correlated noise.
Related
- The Molecular Circumstellar Environments of Red Supergiants in Galactic Open Clusters
- Learning cosmic web environments with diffusion models
- Learning cosmic web environments with diffusion models
- Circumbinary discs as the origin of circumstellar material around interacting H-poor supernovae and fast blue optical transients
- Radio emission from radio-bright fast blue optical transients powered by trans-relativistic shocks in confined circumstellar material
BibTeX
@misc{villegas2026diffusion,
title = {{Diffusion Models for Polarimetric Reconstruction of Circumstellar Environments in Correlated Speckle Noise}},
author = {Quentin Villegas and Laurence Denneulin and Simon Prunet and André Ferrari and Éric Thiébaut and Maud Langlois},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.38229},
eprint = {2609.38229},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.38229},
}
RIS
TY - UNPB
TI - Diffusion Models for Polarimetric Reconstruction of Circumstellar Environments in Correlated Speckle Noise
AU - Quentin Villegas
AU - Laurence Denneulin
AU - Simon Prunet
AU - André Ferrari
AU - Éric Thiébaut
AU - Maud Langlois
PY - 2026
DA - 2026/09/28
PB - arXiv
DO - 10.48550/arxiv.2609.38229
UR - https://arxiv.org/abs/2609.38229
AB - High-contrast polarimetric imaging of circumstellar disks is severely limited by stellar leakage and speckle noise. Building upon the RHAPSODIE framework for polarimetric inverse problems, diffusion-based methods have shown promising results by replacing classical regularization with learned priors. However, RHAPSODIE's white noise assumption fails to capture the spatial correlation structure of atmospheric and instrumental speckles. We extend this framework to handle speckle fluctuations, which are modeled as a Gaussian field with a stationary covariance matrix with a Gaussian spectral density. To efficiently apply the dense precision matrix that includes correlated speckle and heteroskedastic photon noise, we develop a dedicated preconditioner for conjugate gradient iterations. The diffusion prior is trained exclusively on disk-only data. While limitations remain at very low SNRs, our method significantly outperforms Tikhonov regularization and competes favorably with a Plug-and-Play baseline in realistic regimes, successfully recovering morphological features masked by correlated noise.
ER -