Sparse Coding on Cascaded Residuals

TL;DR

This paper introduces a multi-resolution cascade framework for dictionary learning and sparse coding that improves image representation efficiency and accuracy across various tasks like coding, denoising, and inpainting.

Contribution

It presents a novel two-pass multi-resolution cascade approach that enables collaborative reconstructions and adaptive dictionary atoms for enhanced image processing.

Findings

Outperforms state-of-the-art in image coding, denoising, inpainting, and artifact removal.

Uses fewer coefficients for more accurate representations.

Achieves computational efficiency by encoding at coarser resolutions and sparse residuals.

Abstract

This paper seeks to combine dictionary learning and hierarchical image representation in a principled way. To make dictionary atoms capturing additional information from extended receptive fields and attain improved descriptive capacity, we present a two-pass multi-resolution cascade framework for dictionary learning and sparse coding. The cascade allows collaborative reconstructions at different resolutions using the same dimensional dictionary atoms. Our jointly learned dictionary comprises atoms that adapt to the information available at the coarsest layer where the support of atoms reaches their maximum range and the residual images where the supplementary details progressively refine the reconstruction objective. The residual at a layer is computed by the difference between the aggregated reconstructions of the previous layers and the downsampled original image at that layer. Our method generates more flexible and accurate representations using much less number of coefficients. Its computational efficiency stems from encoding at the coarsest resolution, which is minuscule, and encoding the residuals, which are relatively much sparse. Our extensive experiments on multiple datasets demonstrate that this new method is powerful in image coding, denoising, inpainting and artifact removal tasks outperforming the state-of-the-art techniques.