Uncover Treasures in DCT: Advancing JPEG Quality Enhancement by Exploiting Latent Correlations

TL;DR

This paper introduces AJQE, a novel DCT-domain JPEG quality enhancement method that exploits latent correlations in DCT coefficients, leading to better image quality and efficiency compared to pixel-domain approaches.

Contribution

The paper presents a new DCT-domain enhancement technique that leverages correlations in JPEG DCT coefficients, enabling the adaptation of pixel-domain models with improved performance and lower computational costs.

Findings

Achieves 0.35 dB higher PSNR than pixel-domain models.

Increases enhancement throughput by 60.5%.

Effectively exploits DCT coefficient correlations for quality improvement.

Abstract

Joint Photographic Experts Group (JPEG) achieves data compression by quantizing Discrete Cosine Transform (DCT) coefficients, which inevitably introduces compression artifacts. Most existing JPEG quality enhancement methods operate in the pixel domain, suffering from the high computational costs of decoding. Consequently, direct enhancement of JPEG images in the DCT domain has gained increasing attention. However, current DCT-domain methods often exhibit limited performance. To address this challenge, we identify two critical types of correlations within the DCT coefficients of JPEG images. Building on this insight, we propose an Advanced DCT-domain JPEG Quality Enhancement (AJQE) method that fully exploits these correlations. The AJQE method enables the adaptation of numerous well-established pixel-domain models to the DCT domain, achieving superior performance with reduced computational complexity. Compared to the pixel-domain counterparts, the DCT-domain models derived by our method demonstrate a 0.35 dB improvement in PSNR and a 60.5% increase in enhancement throughput on average.