PSI3D: Plug-and-Play 3D Stochastic Inference with Slice-wise Latent Diffusion Prior
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909972462305280 |
|---|---|
| author | Guo, Wenhan Yu, Jinglun Wang, Yaning Kang, Jin U. Sun, Yu |
| author_facet | Guo, Wenhan Yu, Jinglun Wang, Yaning Kang, Jin U. Sun, Yu |
| contents | Diffusion models are highly expressive image priors for Bayesian inverse problems. However, most diffusion models cannot operate on large-scale, high-dimensional data due to high training and inference costs. In this work, we introduce a Plug-and-play algorithm for 3D stochastic inference with latent diffusion prior (PSI3D) to address massive ($1024\times 1024\times 128$) volumes. Specifically, we formulate a Markov chain Monte Carlo approach to reconstruct each two-dimensional (2D) slice by sampling from a 2D latent diffusion model. To enhance inter-slice consistency, we also incorporate total variation (TV) regularization stochastically along the concatenation axis. We evaluate our performance on optical coherence tomography (OCT) super-resolution. Our method significantly improves reconstruction quality for large-scale scientific imaging compared to traditional and learning-based baselines, while providing robust and credible reconstructions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_18367 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PSI3D: Plug-and-Play 3D Stochastic Inference with Slice-wise Latent Diffusion Prior Guo, Wenhan Yu, Jinglun Wang, Yaning Kang, Jin U. Sun, Yu Image and Video Processing Machine Learning Diffusion models are highly expressive image priors for Bayesian inverse problems. However, most diffusion models cannot operate on large-scale, high-dimensional data due to high training and inference costs. In this work, we introduce a Plug-and-play algorithm for 3D stochastic inference with latent diffusion prior (PSI3D) to address massive ($1024\times 1024\times 128$) volumes. Specifically, we formulate a Markov chain Monte Carlo approach to reconstruct each two-dimensional (2D) slice by sampling from a 2D latent diffusion model. To enhance inter-slice consistency, we also incorporate total variation (TV) regularization stochastically along the concatenation axis. We evaluate our performance on optical coherence tomography (OCT) super-resolution. Our method significantly improves reconstruction quality for large-scale scientific imaging compared to traditional and learning-based baselines, while providing robust and credible reconstructions. |
| title | PSI3D: Plug-and-Play 3D Stochastic Inference with Slice-wise Latent Diffusion Prior |
| topic | Image and Video Processing Machine Learning |
| url | https://arxiv.org/abs/2512.18367 |