Abstract:Example learning-based image super-resolution (SR) is recognized as an effective way to produce a high-resolution (HR) image with the help of an external training set. The effectiveness of learning-based SR methods, however, depends highly upon the consistency between the supporting training set and low-resolution (LR) images to be handled. To reduce the adverse effect brought by incompatible high-frequency details in the training set, we propose a single image SR approach by learning multiscale self-similarities from an LR image itself. The proposed SR approach is based upon an observation that small patches in natural images tend to redundantly repeat themselves many times both within the same scale and across different scales. To synthesize the missing details, we establish the HR-LR patch pairs using the initial LR input and its down-sampled version to capture the similarities across different scales and utilize the neighbor embedding algorithm to estimate the relationship between the LR and HR image pairs. To fully exploit the similarities across various scales inside the input LR image, we accumulate the previous resultant images as training examples for the subsequent reconstruction processes and adopt a gradual magnification scheme to upscale the LR input to the desired size step by step. In addition, to preserve sharper edges and suppress aliasing artifacts, we further apply the nonlocal means method to learn the similarity within the same scale and formulate a nonlocal prior regularization term to well pose SR estimation under a reconstruction-based SR framework. Experimental results demonstrate that the proposed method can produce compelling SR recovery both quantitatively and perceptually in comparison with other state-of-the-art baselines.

Example-Based Super-Resolution with Soft Information and Decision

Multi-frame Image Super Resolution with Natural Image Prior

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

A Sparse Representation Based Super-Resolution Image Reconstruction Scheme Utilizing Dual Dictionaries

Single Image Super-Resolution with Multiscale Similarity Learning.

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

Sparse Support Regression for Image Super-Resolution

Learning-based Super-Resolution Method with a Combining of Both Global and Local Constraints

Single-Image Superresolution Based on Local Regression and Nonlocal Self-Similarity

Learning Cross-scale Correspondence and Patch-based Synthesis for Reference-based Super-Resolution.

Image Super-Resolution Based on Adaptive Joint Distribution Modeling

Image Super-Resolution via Hierarchical and Collaborative Sparse Representation

Coarse-to-Fine Learning for Single-Image Super-Resolution

Super-Resolution of Low-Quality Images Based on Compressed Sensing and Sequence Information

Learning Super-Resolution Jointly from External and Internal Examples

Image Super-resolution with Parametric Sparse Model Learning

A Multi-Frame Image Super-Resolution Method.

Single Image Super-Resolution Based on Nonlocal Sparse and Low-Rank Regularization.

Nonlocal Pixel Selection for Multisurface Fitting-Based Super-Resolution.

An Example Image Super-Resolution Algorithm Based on Modified K-Means with Hybrid Particle Swarm Optimization

Multi-scale Dictionary for Single Image Super-Resolution