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.

Efficient Single Image Super-Resolution via Graph Embedding

Efficient Single Image Super-Resolution Via Graph-Constrained Least Squares Regression.

Image Super-Resolution with Sparse Neighbor Embedding

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

Hyperspectral Imagery Spatial Super-Resolution Using Generative Adversarial Network.

Learning Structral Coherence Via Generative Adversarial Network for Single Image Super-Resolution

Remote Sensing Single-Image Resolution Improvement Using A Deep Gradient-Aware Network with Image-Specific Enhancement

Image Super-resolution Based on Sparse Representation With Joint Constraints

Efficient Image Super-Resolution Integration

Single image super-resolution via learned representative features and sparse manifold embedding

Single Image Super-Resolution Based on Nonlocal Similarity and Sparse Representation

Single Image Super Resolution Based on Multiscale Local Similarity and Neighbor Embedding.

Single Image Super-Resolution Via Locally Regularized Anchored Neighborhood Regression and Nonlocal Means.

Image Super-Resolution via Hierarchical and Collaborative Sparse Representation

Fast Image Super-Resolution Based on Limit Gradient Embedding Cascaded Forest

Efficient Real-world Image Super-Resolution Via Adaptive Directional Gradient Convolution

SGCRSR: Sequential Gradient Constrained Regression for Single Image Super-Resolution

Learning Local Gaussian Process Regression for Image Super-Resolution.

SRGAT: Single Image Super-Resolution with Graph Attention Network

Image Super Resolution Using Gaussian Process Regression with Patch Clustering

Fast Single Image Super-Resolution Using Sparse Gaussian Process Regression