Quantum Implicit Neural Compression

TL;DR

This paper introduces quantum INR (quINR), leveraging quantum neural networks to enhance signal compression, especially for high-frequency details, outperforming traditional codecs with up to 1.2dB gain.

Contribution

The paper proposes quantum INR (quINR), a novel approach that uses quantum neural networks to significantly improve data compression efficiency.

Findings

quINR outperforms traditional codecs in rate-distortion performance

Achieves up to 1.2dB gain in image compression

Enhances high-frequency detail preservation in signals

Abstract

Signal compression based on implicit neural representation (INR) is an emerging technique to represent multimedia signals with a small number of bits. While INR-based signal compression achieves high-quality reconstruction for relatively low-resolution signals, the accuracy of high-frequency details is significantly degraded with a small model. To improve the compression efficiency of INR, we introduce quantum INR (quINR), which leverages the exponentially rich expressivity of quantum neural networks for data compression. Evaluations using some benchmark datasets show that the proposed quINR-based compression could improve rate-distortion performance in image compression compared with traditional codecs and classic INR-based coding methods, up to 1.2dB gain.