Abstract:Learning based super-resolution (SR) methods, which predict the high-resolution pixel values but not directly provide an estimation of uncertainty, are typically non-probabilistic and have limited generalization ability. Gaussian processes can provide a framework for deriving regression techniques with explicit uncertainty models, but Gaussian Process Regression (GPR) has a significant drawback in being time consuming. The computational complexity of GPR is cubic in the number of training examples, which is prohibitively expensive for a large-scale training set. In this article, we have proposed learning local GPR for image SR. Two algorithms are developed to support local GPR for super resolution. A data-driven GPR based super-resolution algorithm is first developed to learn a local GPR model for every LR patch on an input oriented training dataset with moderate size. In order to further improve the running speed, a prototype based GPR algorithm is developed for super resolution. The proposed algorithm is about one-order faster than the data-driven GPR solution because it makes models for the prototypes of image patches rather than for each image patch. Thus, the local regression efforts are greatly reduced to just finding the nearest prototype for each LR image patch and applying its corresponding pre-computed projective matrix for super-resolution prediction. Our algorithms have greater robustness and usability as they provide a formularized way to automatically learn the hyper-parameters introduced for optimizing the covariance function, while most of the state-of-the-art super-resolution methods could only utilize these parameters in a cross-verification way. Moreover, our algorithms offer confidence values at the test points which benefit the pixels' post-processing. Our algorithms are evaluated on popular datasets that are widely used in the super-resolution literature, and the experimental results have demonstrated that the efficiency and effectiveness of our proposed algorithms are comparative with several state-of-the-arts super-resolution methods.

Supervised Deep Kriging for Single-Image Super-Resolution

Depth Image Super-resolution via Two-Branch Network

Epistemic-Uncertainty-Based Divide-and-Conquer Network for Single-Image Super-Resolution

Single Remote Sensing Image Super-Resolution Via a Generative Adversarial Network with Stratified Dense Sampling and Chain Training

Generalized Joint Kernel Regression and Adaptive Dictionary Learning for Single-Image Super-Resolution.

Super-Resolution Reconstruction of Depth Image Based on Kriging Interpolation

ESKN: Enhanced Selective Kernel Network for Single Image Super-Resolution

Image Super-Resolution Via Deterministic-Stochastic Synthesis and Local Statistical Rectification

SuperGS: Super-Resolution 3D Gaussian Splatting Enhanced by Variational Residual Features and Uncertainty-Augmented Learning

Kernel-attended Residual Network for Single Image Super-Resolution.

Image Super-Resolution Using Deep Convolutional Networks

Unsupervised Learning for Real-World Super-Resolution

Deep Networks for Image and Video Super-Resolution

Guided Super-Resolution as Pixel-to-Pixel Transformation

DeepSUM: Deep neural network for Super-resolution of Unregistered Multitemporal images

Deep Learning Based Single Image Super-resolution: A Survey

Single Image Super-Resolution Using Sparse Prior

Deep Learning for Single Image Super-Resolution: A Brief Review

Learning Local Gaussian Process Regression for Image Super-Resolution.

Data-Driven External Compensation Guided Deep Networks for Image Super-Resolution

Single Image Super-Resolution with Non-Local Means and Steering Kernel Regression