Variational Bayes image restoration with compressive autoencoders

arXiv:2311.17744v2 Announce Type: replace
Abstract: Regularization of inverse problems is of paramount importance in computational imaging. The ability of neural networks to learn efficient image representations has been recently exploited to design powerful data-driven regularizers. While state-of-the-art plug-and-play methods rely on an implicit regularization provided by neural denoisers, alternative Bayesian approaches consider Maximum A Posteriori (MAP) estimation in the latent space of a generative model, thus with an explicit regularization. However, state-of-the-art deep generative models require a huge amount of …

abstract art arxiv autoencoders bayes bayesian computational cs.cv data data-driven design image image restoration imaging importance learn networks neural networks regularization state stat.ml type