Stochastic Solutions for Linear Inverse Problems using the Prior Implicit in a DenoiserDownload PDF

21 May 2021, 20:51 (edited 14 Jan 2022)NeurIPS 2021 PosterReaders: Everyone
  • Keywords: inverse problems, image priors, unsupervised learning, semi-supervised learning, deblurring, superresolution, compressive sensing
  • TL;DR: We've described a framework for using the prior implicitly embedded in a denoiser to solve other linear inverse problems by sampling.
  • Abstract: Deep neural networks have provided state-of-the-art solutions for problems such as image denoising, which implicitly rely on a prior probability model of natural images. Two recent lines of work – Denoising Score Matching and Plug-and-Play – propose methodologies for drawing samples from this implicit prior and using it to solve inverse problems, respectively. Here, we develop a parsimonious and robust generalization of these ideas. We rely on a classic statistical result that shows 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 to derive 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 denoising. A generalization of this algorithm to constrained sampling provides a method for using the implicit prior to solve any deterministic linear inverse problem, with no additional training, thus extending the power of supervised learning for denoising to a much broader set of problems. The algorithm relies on minimal assumptions and exhibits robust convergence over a wide range of parameter choices. To demonstrate the generality of our method, we use it to obtain state-of-the-art levels of unsupervised performance for deblurring, super-resolution, and compressive sensing.
  • Supplementary Material: pdf
  • Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.
  • Code: https://github.com/LabForComputationalVision/universal_inverse_problem
13 Replies

Loading