Abstract:This paper presents a new approach to single-image super-resolution, based on sparse signal representation. Research on image statistics suggests that image patches can be well-represented as a sparse linear combination of elements from an appropriately chosen over-complete dictionary. Inspired by this observation, we seek a sparse representation for each patch of the low-resolution input, and then use the coefficients of this representation to generate the high-resolution output. Theoretical results from compressed sensing suggest that under mild conditions, the sparse representation can be correctly recovered from the downsampled signals. By jointly training two dictionaries for the low- and high-resolution image patches, we can enforce the similarity of sparse representations between the low resolution and high resolution image patch pair with respect to their own dictionaries. Therefore, the sparse representation of a low resolution image patch can be applied with the high resolution image patch dictionary to generate a high resolution image patch. The learned dictionary pair is a more compact representation of the patch pairs, compared to previous approaches, which simply sample a large amount of image patch pairs, reducing the computational cost substantially. The effectiveness of such a sparsity prior is demonstrated for both general image super-resolution and the special case of face hallucination. In both cases, our algorithm generates high-resolution images that are competitive or even superior in quality to images produced by other similar SR methods. In addition, the local sparse modeling of our approach is naturally robust to noise, and therefore the proposed algorithm can handle super-resolution with noisy inputs in a more unified framework.

Image Super-Resolution Reconstruction Based on Adaptive Sparse Representation

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

Image super-resolution based on sparse representation and self-similarity learning

Image Super-Resolution Reconstruction Algorithm Using Over-Complete Sparse Representation

Super-Resolution Reconstruction Via Multiple Sparse Dictionary Combined With Sparse Coding

Image Super-resolution Reconstruction Based on Smoothly Approximate Over-complete Sparse Representation

Sparse Representation over Adaptive Dictionaries for Super-Resolution

Double Sparse Dictionary Learning for Image Super Resolution

Collaborative Sparse Dictionary Learning for Reconstruction of Single Image Super Resolution

Robust Single Image Super-resolution Reconstruction Model and Algorithm Based on Over-complete Dictionary

Super-resolution Reconstruction of Sparse Dictionaries Based on Integrated Feature.

A Local Adaptive Structure Sparse Representation Algorithm for Image Reconstruction

Single Image Super-Resolution Via Adaptive Sparse Representation and Low-Rank Constraint.

A Sparse Representation Based Approach for Super-Resolution Reconstruction over Geometric Dictionaries and Consistency of Gradients

Alternatively Constrained Dictionary Learning for Image Superresolution

Image Super-Resolution by Dictionary Concatenation and Sparse Representation with Approximate L0 Norm Minimization

Noisy Image Super-Resolution Reconstruction Based on Sparse Representation

Image Super-Resolution Via Nonlocal Similarity and Group Structured Sparse Representation

A Novel Image Super-Resolution Reconstruction Method Based on Sparse Representation Using Classified Dictionaries.

Super-resolution reconstruction algorithm based on multi-component dictionary and sparse representation

Image super-resolution via sparse representation.