ProGIC: Progressive and Lightweight Generative Image Compression with Residual Vector Quantization

TL;DR

ProGIC introduces a compact, progressive generative image compression method using residual vector quantization, achieving high perceptual quality, significant bitrate savings, and fast encoding/decoding suitable for practical deployment.

Contribution

The paper presents a novel lightweight, progressive GIC framework with residual vector quantization, enabling flexible transmission and efficient deployment on various devices.

Findings

Achieves up to 57.57% bitrate savings on DISTS.

Over 10x faster encoding and decoding on GPUs.

Provides progressive transmission with comparable quality.

Abstract

Recent advances in generative image compression (GIC) have delivered remarkable improvements in perceptual quality. However, many GICs rely on large-scale and rigid models, which severely constrain their utility for flexible transmission and practical deployment in low-bitrate scenarios. To address these issues, we propose Progressive Generative Image Compression (ProGIC), a compact codec built on residual vector quantization (RVQ). In RVQ, a sequence of vector quantizers encodes the residuals stage by stage, each with its own codebook. The resulting codewords sum to a coarse-to-fine reconstruction and a progressive bitstream, enabling previews from partial data. We pair this with a lightweight backbone based on depthwise-separable convolutions and small attention blocks, enabling practical deployment on both GPUs and CPU-only devices. Experimental results show that ProGIC attains comparable compression performance compared with previous methods. It achieves bitrate savings of up to 57.57% on DISTS and 58.83% on LPIPS compared to MS-ILLM on the Kodak dataset. Beyond perceptual quality, ProGIC enables progressive transmission for flexibility, and also delivers over 10 times faster encoding and decoding compared with MS-ILLM on GPUs for efficiency.