Generalized Expectation Maximization Framework for Blind Image Super Resolution
This addresses the issue of error accumulation in blind super-resolution for image processing applications, but it appears incremental as it builds on existing Bayesian and learning methods.
The paper tackles the problem of error accumulation in blind single image super-resolution by proposing an end-to-end learning framework called SREMN, which integrates learning techniques into a generalized expectation-maximization algorithm and shows superiority in experiments compared to existing work.
Learning-based methods for blind single image super resolution (SISR) conduct the restoration by a learned mapping between high-resolution (HR) images and their low-resolution (LR) counterparts degraded with arbitrary blur kernels. However, these methods mostly require an independent step to estimate the blur kernel, leading to error accumulation between steps. We propose an end-to-end learning framework for the blind SISR problem, which enables image restoration within a unified Bayesian framework with either full- or semi-supervision. The proposed method, namely SREMN, integrates learning techniques into the generalized expectation-maximization (GEM) algorithm and infers HR images from the maximum likelihood estimation (MLE). Extensive experiments show the superiority of the proposed method with comparison to existing work and novelty in semi-supervised learning.