Efficient and Parallel Separable Dictionary Learning

TL;DR

This paper introduces a highly parallelizable algorithm for learning separable dictionaries that efficiently represent 2D signals like images, achieving competitive sparsity with reduced computational costs.

Contribution

The paper presents a novel parallelizable dictionary learning algorithm for 2D signals that outperforms existing methods in efficiency while maintaining high-quality sparse representations.

Findings

Achieves sparse representations comparable to state-of-the-art methods

Reduces computational cost of dictionary learning

Effective in image denoising and hyperspectral data representation

Abstract

Separable, or Kronecker product, dictionaries provide natural decompositions for 2D signals, such as images. In this paper, we describe a highly parallelizable algorithm that learns such dictionaries which reaches sparse representations competitive with the previous state of the art dictionary learning algorithms from the literature but at a lower computational cost. We highlight the performance of the proposed method to sparsely represent image and hyperspectral data, and for image denoising.