How Good Are Deep Generative Models for Solving Inverse Problems?

Shichong Peng, Alireza Moazeni, Ke Li

Keywords: deep generative models, inverse problem, diffusion, GAN, IMLE, model uncertainty, output validity

Abstract: Deep generative models, such as diffusion models, GANs, and IMLE, have shown impressive capability in tackling inverse problems. However, the validity of model-generated solutions w.r.t. the forward process and the reliability of associated uncertainty estimates remain understudied. This study evaluates recent diffusion-based, GAN-based, and IMLE-based methods on three inverse problems, i.e., 16x super-resolution, colourization, and image decompression. We assess the validity of these models' outputs as solutions to the inverse problems and conduct a thorough analysis of the reliability of the models' estimates of uncertainty over the solution. Overall, we find that the IMLE-based CHIMLE method outperforms other methods in terms of producing valid solutions and reliable uncertainty estimates.

Submission Number: 11