Stochastic Deep Restoration Priors for Imaging Inverse Problems
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914963360055296 |
|---|---|
| author | Hu, Yuyang Peng, Albert Gan, Weijie Milanfar, Peyman Delbracio, Mauricio Kamilov, Ulugbek S. |
| author_facet | Hu, Yuyang Peng, Albert Gan, Weijie Milanfar, Peyman Delbracio, Mauricio Kamilov, Ulugbek S. |
| contents | Deep neural networks trained as image denoisers are widely used as priors for solving imaging inverse problems. While Gaussian denoising is thought sufficient for learning image priors, we show that priors from deep models pre-trained as more general restoration operators can perform better. We introduce Stochastic deep Restoration Priors (ShaRP), a novel method that leverages an ensemble of such restoration models to regularize inverse problems. ShaRP improves upon methods using Gaussian denoiser priors by better handling structured artifacts and enabling self-supervised training even without fully sampled data. We prove ShaRP minimizes an objective function involving a regularizer derived from the score functions of minimum mean square error (MMSE) restoration operators, and theoretically analyze its convergence. Empirically, ShaRP achieves state-of-the-art performance on tasks such as magnetic resonance imaging reconstruction and single-image super-resolution, surpassing both denoiser-and diffusion-model-based methods without requiring retraining. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_02057 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Stochastic Deep Restoration Priors for Imaging Inverse Problems Hu, Yuyang Peng, Albert Gan, Weijie Milanfar, Peyman Delbracio, Mauricio Kamilov, Ulugbek S. Image and Video Processing Deep neural networks trained as image denoisers are widely used as priors for solving imaging inverse problems. While Gaussian denoising is thought sufficient for learning image priors, we show that priors from deep models pre-trained as more general restoration operators can perform better. We introduce Stochastic deep Restoration Priors (ShaRP), a novel method that leverages an ensemble of such restoration models to regularize inverse problems. ShaRP improves upon methods using Gaussian denoiser priors by better handling structured artifacts and enabling self-supervised training even without fully sampled data. We prove ShaRP minimizes an objective function involving a regularizer derived from the score functions of minimum mean square error (MMSE) restoration operators, and theoretically analyze its convergence. Empirically, ShaRP achieves state-of-the-art performance on tasks such as magnetic resonance imaging reconstruction and single-image super-resolution, surpassing both denoiser-and diffusion-model-based methods without requiring retraining. |
| title | Stochastic Deep Restoration Priors for Imaging Inverse Problems |
| topic | Image and Video Processing |
| url | https://arxiv.org/abs/2410.02057 |