Solving Linear Inverse Problems Using the Prior Implicit in a Denoiser

Abstract

Prior probability models are a fundamental component of many image processing problems, but density estimation is notoriously difficult for high-dimensional signals such as photographic images. Deep neural networks have provided state-of-the-art solutions for problems such as denoising, which implicitly rely on a prior probability model of natural images. Here, we develop a robust and general methodology for making use of this implicit prior. We rely on a statistical result due to Miyasawa (1961), who showed that the least-squares solution for removing additive Gaussian noise can be written directly in terms of the gradient of the log of the noisy signal density. We use this fact to develop a stochastic coarse-to-fine gradient ascent procedure for drawing high-probability samples from the implicit prior embedded within a CNN trained to perform blind (i.e., with unknown noise level) least-squares denoising. A generalization of this algorithm to constrained sampling provides a method for using the implicit prior to solve any linear inverse problem, with no additional training. We demonstrate this general form of transfer learning in multiple applications, using the same algorithm to produce state-of-the-art levels of unsupervised performance for deblurring, super-resolution, inpainting, and compressive sensing.

19 pages, 12 figures. Changes: more detailed description of relationships to previous literature, including empirical comparisons for super-resolution, debarring, and compressive sensing

Cited by in corpus (8)

• CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation
• Adversarial score matching and improved sampling for image generation
• Gotta Go Fast When Generating Data with Score-Based Models
• SNIPS: Solving Noisy Inverse Problems Stochastically
• Score-Based Generative Classifiers
• Posterior Sampling for Image Restoration using Explicit Patch Priors
• D2C: Diffusion-Denoising Models for Few-shot Conditional Generation
• Dual Training of Energy-Based Models with Overparametrized Shallow Neural Networks