Fast Convolutional Sparse Coding in the Dual Domain

TL;DR

This paper introduces a dual domain optimization framework for convolutional sparse coding that significantly accelerates computation and extends applicability to higher-dimensional data like videos and RGB images.

Contribution

The paper presents a novel dual domain optimization approach for CSC and extends it to higher dimensions, achieving substantial speed improvements.

Findings

Up to 20x speedup over existing CSC methods

Applicable to higher-dimensional data such as videos and RGB images

Enhanced efficiency in convolutional sparse coding tasks

Abstract

Convolutional sparse coding (CSC) is an important building block of many computer vision applications ranging from image and video compression to deep learning. We present two contributions to the state of the art in CSC. First, we significantly speed up the computation by proposing a new optimization framework that tackles the problem in the dual domain. Second, we extend the original formulation to higher dimensions in order to process a wider range of inputs, such as RGB images and videos. Our results show up to 20 times speedup compared to current state-of-the-art CSC solvers.