Designing A Composite Dictionary Adaptively From Joint Examples

TL;DR

This paper introduces a composite dictionary framework combining external and internal image examples, adaptively weighted to improve image restoration tasks like denoising and super-resolution.

Contribution

It proposes a novel adaptive composite dictionary design that leverages both external and internal examples for enhanced image restoration.

Findings

Significant improvements in image denoising and super-resolution.

Effective combination of external and internal examples.

Adaptive weighting enhances model performance.

Abstract

We study the complementary behaviors of external and internal examples in image restoration, and are motivated to formulate a composite dictionary design framework. The composite dictionary consists of the global part learned from external examples, and the sample-specific part learned from internal examples. The dictionary atoms in both parts are further adaptively weighted to emphasize their model statistics. Experiments demonstrate that the joint utilization of external and internal examples leads to substantial improvements, with successful applications in image denoising and super resolution.