Pseudoinverse-Guided Diffusion Models for Inverse Problems

By Jiaming Song, Arash Vahdat, Morteza Mardani, Jan Kautz. International Conference on Learning Representations, 2023

583 citations recorded by the source corpus.

How researchers see this paper

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

FreeDoM: Training-Free Energy-Guided Conditional Diffusion Model (2023, background): “[27, 11, 14] attempt to use the cross-attention control to realize the conditional generation; [5, 6, 7, 8, 19, 24, 39, 41, 44, 43] directly modify the intermediate results to achieve zero-shot image restoration; [25] realizes image translation by adjusting the initial noisy images.”
Practical and Asymptotically Exact Conditional Sampling in Diffusion Models (2023, unknown): “Outside of the context of SMC and the twisting technique, the use of denoising estimate x̂θ(x, t) to form an approximation p̃θ(x | x, y) to pθ(x | x, y) is called reconstruction guidance [5, 16, 27].”
DDFM: Denoising Diffusion Model for Multi-Modality Image Fusion (2023, methodology): “In Π GDM [47], the authors employ few approximations to make the log-likelihood term tractable and hence make it able to solve inverse problems with even non-differentiable measurements.”
Exploiting Diffusion Prior for Real-World Image Super-Resolution (2023, unknown): “In this study, we present StableSR , an approach that preserves pre-trained diffusion priors without making explicit assumptions about the degradations.”