Sequential Image Recovery Using Joint Hierarchical Bayesian Learning

Recovering temporal image sequences (videos) based on indirect, noisy, or incomplete data is an essential yet challenging task. We specifically consider the case where each set missing vital information, which prevents accurate recovery of individual images. Although some recent (variational) methods have demonstrated high-resolution jointly recovering sequential images, there remain robustness issues due to parameter tuning and restrictions type Here, we present a method hierarchical Bayesian learning for joint images that incorporates prior intra- inter-image information. Our restores information in by "borrowing" it from other As result, \emph{all} reconstructions yield improved accuracy. can be used various acquisitions allows uncertainty quantification. Some preliminary results indicate its potential use deblurring magnetic resonance imaging.