Self-supervised Deep Image Restoration via Adaptive Stochastic Gradient Langevin Dynamics

Weixi Wang, Ji Li, Hui Ji

Abstract:

While supervised deep learning has been a prominent tool for solving many image restoration problems,
there is an increasing interest on studying self-supervised or un-supervised methods to address the challenges and costs of collecting truth images. Based on the neuralization of a Bayesian estimator of the problem, this paper presents a self-supervised deep learning approach to general image restoration problems. The key ingredient of the neuralized estimator is an adaptive stochastic gradient Langevin dynamics algorithm for efficiently sampling the posterior distribution of network weights.
The proposed method is applied on two image restoration problems: compressed sensing and phase retrieval. The experiments on these applications showed that the proposed method not only outperformed existing non-learning and unsupervised solutions in terms of image restoration quality, but also is more computationally efficient.