Solving Inverse Problems by Joint Posterior Maximization with Autoencoding Prior

by   Mario González, et al.

In this work we address the problem of solving ill-posed inverse problems in imaging where the prior is a variational autoencoder (VAE). Specifically we consider the decoupled case where the prior is trained once and can be reused for many different log-concave degradation models without retraining. Whereas previous MAP-based approaches to this problem lead to highly non-convex optimization algorithms, our approach computes the joint (space-latent) MAP that naturally leads to alternate optimization algorithms and to the use of a stochastic encoder to accelerate computations. The resulting technique (JPMAP) performs Joint Posterior Maximization using an Autoencoding Prior. We show theoretical and experimental evidence that the proposed objective function is quite close to bi-convex. Indeed it satisfies a weak bi-convexity property which is sufficient to guarantee that our optimization scheme converges to a stationary point. We also highlight the importance of correctly training the VAE using a denoising criterion, in order to ensure that the encoder generalizes well to out-of-distribution images, without affecting the quality of the generative model. This simple modification is key to providing robustness to the whole procedure. Finally we show how our joint MAP methodology relates to more common MAP approaches, and we propose a continuation scheme that makes use of our JPMAP algorithm to provide more robust MAP estimates. Experimental results also show the higher quality of the solutions obtained by our JPMAP approach with respect to other non-convex MAP approaches which more often get stuck in spurious local optima.


page 15

page 17

page 18

page 19

page 20


Solving Inverse Problems by Joint Posterior Maximization with a VAE Prior

In this paper we address the problem of solving ill-posed inverse proble...

Inverse problem regularization with hierarchical variational autoencoders

In this paper, we propose to regularize ill-posed inverse problems using...

Learning Gradually Non-convex Image Priors Using Score Matching

In this paper, we propose a unified framework of denoising score-based m...

Scalable Bayesian uncertainty quantification in imaging inverse problems via convex optimization

We propose a Bayesian uncertainty quantification method for large-scale ...

Adaptive Quantile Sparse Image (AQuaSI) Prior for Inverse Imaging Problems

Inverse problems play a central role for many classical computer vision ...

Meta-learning for Multi-variable Non-convex Optimization Problems: Iterating Non-optimums Makes Optimum Possible

In this paper, we aim to address the problem of solving a non-convex opt...

Inverse Scale Space Iterations for Non-Convex Variational Problems: The Continuous and Discrete Case

Non-linear filtering approaches allow to obtain decompositions of images...

Please sign up or login with your details

Forgot password? Click here to reset