RestoreGrad: Signal Restoration Using Conditional Denoising Diffusion Models with Jointly Learned Prior
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908398358888448 |
|---|---|
| author | Lee, Ching-Hua Yang, Chouchang Cho, Jaejin Saidutta, Yashas Malur Srinivasa, Rakshith Sharma Shen, Yilin Jin, Hongxia |
| author_facet | Lee, Ching-Hua Yang, Chouchang Cho, Jaejin Saidutta, Yashas Malur Srinivasa, Rakshith Sharma Shen, Yilin Jin, Hongxia |
| contents | Denoising diffusion probabilistic models (DDPMs) can be utilized to recover a clean signal from its degraded observation(s) by conditioning the model on the degraded signal. The degraded signals are themselves contaminated versions of the clean signals; due to this correlation, they may encompass certain useful information about the target clean data distribution. However, existing adoption of the standard Gaussian as the prior distribution in turn discards such information when shaping the prior, resulting in sub-optimal performance. In this paper, we propose to improve conditional DDPMs for signal restoration by leveraging a more informative prior that is jointly learned with the diffusion model. The proposed framework, called RestoreGrad, seamlessly integrates DDPMs into the variational autoencoder (VAE) framework, taking advantage of the correlation between the degraded and clean signals to encode a better diffusion prior. On speech and image restoration tasks, we show that RestoreGrad demonstrates faster convergence (5-10 times fewer training steps) to achieve better quality of restored signals over existing DDPM baselines and improved robustness to using fewer sampling steps in inference time (2-2.5 times fewer), advocating the advantages of leveraging jointly learned prior for efficiency improvements in the diffusion process. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_13574 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RestoreGrad: Signal Restoration Using Conditional Denoising Diffusion Models with Jointly Learned Prior Lee, Ching-Hua Yang, Chouchang Cho, Jaejin Saidutta, Yashas Malur Srinivasa, Rakshith Sharma Shen, Yilin Jin, Hongxia Image and Video Processing Machine Learning Audio and Speech Processing Denoising diffusion probabilistic models (DDPMs) can be utilized to recover a clean signal from its degraded observation(s) by conditioning the model on the degraded signal. The degraded signals are themselves contaminated versions of the clean signals; due to this correlation, they may encompass certain useful information about the target clean data distribution. However, existing adoption of the standard Gaussian as the prior distribution in turn discards such information when shaping the prior, resulting in sub-optimal performance. In this paper, we propose to improve conditional DDPMs for signal restoration by leveraging a more informative prior that is jointly learned with the diffusion model. The proposed framework, called RestoreGrad, seamlessly integrates DDPMs into the variational autoencoder (VAE) framework, taking advantage of the correlation between the degraded and clean signals to encode a better diffusion prior. On speech and image restoration tasks, we show that RestoreGrad demonstrates faster convergence (5-10 times fewer training steps) to achieve better quality of restored signals over existing DDPM baselines and improved robustness to using fewer sampling steps in inference time (2-2.5 times fewer), advocating the advantages of leveraging jointly learned prior for efficiency improvements in the diffusion process. |
| title | RestoreGrad: Signal Restoration Using Conditional Denoising Diffusion Models with Jointly Learned Prior |
| topic | Image and Video Processing Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2502.13574 |