Multilevel Stochastic Plug-and-Play for Sparse-View CT Reconstruction
Abstract
Sparse-view computed tomography (SVCT) reduces radiation exposure and acquisition time, but the limited number of projection views makes the reconstruction problem severely ill-posed and leads to streak artifacts when analytical methods are used. Plug-and-Play (PnP) methods combine data fidelity with learned image priors, while stochastic PnP methods use re-noising to reduce the mismatch between denoiser training inputs and reconstruction iterates. However, these methods often require many iterations, which limits their practical efficiency. In this work, we propose multilevel (ML) Stochastic Plug-and-Play (ML-SPnP), a ML stochastic PnP method for SVCT reconstruction formulated on wavelet multiresolution analysis (MRA) approximation spaces. The method combines coarse-to-fine initialization with a finite sequence of ML corrections using resolution-specific denoisers. Under suitable assumptions, the fine-level iterations following the finite ML phase inherit the convergence guarantees of the underlying stochastic PnP algorithm. Experiments on two CT datasets with 20-50 projection views show that ML-SPnP achieves reconstruction quality comparable to the evaluated stochastic PnP methods, while generally requiring less reconstruction time. These results support ML acceleration as an effective strategy for reducing the computational cost of stochastic PnP reconstruction. The code will be made publicly available at https://github.com/antoinedepaepe/ml-spnp.
Related
- A Contrast-Source Inversion Scheme Based on Stochastic Optimization and Plug-and-Play Regularization
- Fully Fiber-Integrated 3D-Printed Probes for Plug-and-Play Endoscopic Optical Coherence Tomography
- A plug-and-play platform for self-aligning 3D-printed structures on optical fibers
- Low-Rank Tensor Manifolds for Turbulent Field Reconstruction from Sparse Data
- Physical-Field Reconstruction from Sparse Observations: When Are Diffusion Models Preferable to Deterministic Regression?
BibTeX
@misc{paepe2026multilevel,
title = {{Multilevel Stochastic Plug-and-Play for Sparse-View CT Reconstruction}},
author = {Antoine De Paepe and Alexandre Bousse and Dimitris Visvikis},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2606.24567},
eprint = {2606.24567},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2606.24567},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Multilevel Stochastic Plug-and-Play for Sparse-View CT Reconstruction
AU - Antoine De Paepe
AU - Alexandre Bousse
AU - Dimitris Visvikis
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2606.24567
UR - https://arxiv.org/abs/2606.24567
AB - Sparse-view computed tomography (SVCT) reduces radiation exposure and acquisition time, but the limited number of projection views makes the reconstruction problem severely ill-posed and leads to streak artifacts when analytical methods are used. Plug-and-Play (PnP) methods combine data fidelity with learned image priors, while stochastic PnP methods use re-noising to reduce the mismatch between denoiser training inputs and reconstruction iterates. However, these methods often require many iterations, which limits their practical efficiency. In this work, we propose multilevel (ML) Stochastic Plug-and-Play (ML-SPnP), a ML stochastic PnP method for SVCT reconstruction formulated on wavelet multiresolution analysis (MRA) approximation spaces. The method combines coarse-to-fine initialization with a finite sequence of ML corrections using resolution-specific denoisers. Under suitable assumptions, the fine-level iterations following the finite ML phase inherit the convergence guarantees of the underlying stochastic PnP algorithm. Experiments on two CT datasets with 20-50 projection views show that ML-SPnP achieves reconstruction quality comparable to the evaluated stochastic PnP methods, while generally requiring less reconstruction time. These results support ML acceleration as an effective strategy for reducing the computational cost of stochastic PnP reconstruction. The code will be made publicly available at https://github.com/antoinedepaepe/ml-spnp.
ER -