Sparse Multi-layer Image Approximation: Facial Image Compression

TL;DR

This paper introduces a multi-layer image representation scheme that improves compression efficiency for facial images, outperforming JPEG2000 at high compression ratios by leveraging multi-layer dictionary learning.

Contribution

It presents a novel multi-layer image approximation method based on information theory and dictionary learning, specifically enhancing facial image compression performance.

Findings

Significant PSNR improvement over JPEG2000 at high compression ratios.

Effective multi-layer representation reduces information loss.

Demonstrated advantages for facial image compression.

Abstract

We propose a scheme for multi-layer representation of images. The problem is first treated from an information-theoretic viewpoint where we analyze the behavior of different sources of information under a multi-layer data compression framework and compare it with a single-stage (shallow) structure. We then consider the image data as the source of information and link the proposed representation scheme to the problem of multi-layer dictionary learning for visual data. For the current work we focus on the problem of image compression for a special class of images where we report a considerable performance boost in terms of PSNR at high compression ratios in comparison with the JPEG2000 codec.