Plug-and-Play Posterior Sampling for Blind Inverse Problems

2025-05-28

AI Summary

This work addresses blind inverse problems, where both the clean image and the degradation operator are unknown. We propose Blind-PnPDM, a blind plug-and-play diffusion model that introduces, for the first time, a dual-diffusion-model collaborative prior: one diffusion model captures the natural distribution of clean images, while the other implicitly learns the parameter distribution of the degradation operator (e.g., blur kernels), obviating explicit prior design or separate parameter estimation. Within a plug-and-play framework, posterior sampling is reformulated as alternating Gaussian denoising steps. Evaluated on blind image deblurring, Blind-PnPDM achieves significant improvements over state-of-the-art methods, with substantial gains in PSNR and SSIM. Reconstructed images exhibit enhanced sharpness, visual naturalness, and strong generalization across diverse degradation types.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

We introduce Blind Plug-and-Play Diffusion Models (Blind-PnPDM) as a novel framework for solving blind inverse problems where both the target image and the measurement operator are unknown. Unlike conventional methods that rely on explicit priors or separate parameter estimation, our approach performs posterior sampling by recasting the problem into an alternating Gaussian denoising scheme. We leverage two diffusion models as learned priors: one to capture the distribution of the target image and another to characterize the parameters of the measurement operator. This PnP integration of diffusion models ensures flexibility and ease of adaptation. Our experiments on blind image deblurring show that Blind-PnPDM outperforms state-of-the-art methods in terms of both quantitative metrics and visual fidelity. Our results highlight the effectiveness of treating blind inverse problems as a sequence of denoising subproblems while harnessing the expressive power of diffusion-based priors.

Problem

• Solving blind inverse problems with unknown image and operator
• Alternating Gaussian denoising for posterior sampling
• Leveraging diffusion models as flexible learned priors

Innovation

• Blind-PnPDM solves blind inverse problems
• Uses two diffusion models as priors
• Alternating Gaussian denoising scheme