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

By Zahra Kadkhodaie, Eero P. Simoncelli. 2021

193 citations recorded by the source corpus.

Subfields: Generative Models, Numerical Methods & Scientific Computing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Generative Models

In Numerical Methods & Scientific Computing

Citations in the corpus

Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding (2022, methodology): “Related model designs include the work of [70, 32, 33].”
Video Diffusion Models (2022, neutral): “[25, 24].”
Diffusion models as plug-and-play priors (2022, methodology): “A generalization of these methods in [18] shows how any pretrained denoising network can be used as the prior for solving linear inverse problems.”
Deblurring via Stochastic Refinement (2021, methodology): “One way to circumvent the regression to the mean phenomenon is to avoid point estimations and directly learn to generate samples from the posterior distribution [31–33 , 51].”
Denoising Diffusion Models for Plug-and-Play Image Restoration (2023, unknown): “Laumont et al .”
Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise (2022, methodology): “Recently, diffusion models have been applied to them [Song et al., 2021b] for the problems of deblurring, denoising, super-resolution, and compressive sensing [Whang et al., 2021, Kawar et al., 2021, Saharia et al., 2021, Kadkhodaie and Simoncelli, 2021].”
Diffusion Posterior Sampling for General Noisy Inverse Problems (2022, methodology): “Furthermore, while diffusion models were applied to various inverse problems including inpainting (Kadkhodaie & Simoncelli, 2021; Song et al., 2021b; Chung et al., 2022b; Kawar et al., 2022; Chung et al., 2022a), super-resolution (Kadkhodaie & Simoncelli, 2021; Choi et al., 2021; Chung et al.,…”
Denoising Diffusion Restoration Models (2022, background): “, degradation model is known during training) [42, 41, 13, 10, 54] and unsupervised settings [22, 26, 25, 21, 46, 47, 9].”
Diffusion Models in Vision: A Survey (2022, methodology): “[125] is to adapt the algorithm to linear inverse problems, such as deblurring and super-resolution.”
Improving Diffusion Models for Inverse Problems using Manifold Constraints (2022, methodology): “To address this, recent conditional diffusion models [22, 41, 8, 9] utilize the unconditional score function ∇ x log p t ( x ) but rely on a projection-based measurement constraint to impose the conditions.”