SmartSplat: Feature-Smart Gaussians for Scalable Compression of Ultra-High-Resolution Images

TL;DR

SmartSplat introduces a feature-aware Gaussian splatting framework for ultra-high-resolution image compression, achieving high fidelity and scalability with limited primitives, outperforming existing methods.

Contribution

It proposes a novel, adaptive Gaussian splatting method that incorporates image-aware features and joint optimization for scalable, high-quality compression of ultra-high-resolution images.

Findings

Outperforms state-of-the-art methods at similar compression ratios

Achieves higher reconstruction quality under strong compression

Demonstrates scalability on 16K resolution datasets

Abstract

Recent advances in generative AI have accelerated the production of ultra-high-resolution visual content, posing significant challenges for efficient compression and real-time decoding on end-user devices. Inspired by 3D Gaussian Splatting, recent 2D Gaussian image models improve representation efficiency, yet existing methods struggle to balance compression ratio and reconstruction fidelity in ultra-high-resolution scenarios. To address this issue, we propose SmartSplat, a highly adaptive and feature-aware GS-based image compression framework that supports arbitrary image resolutions and compression ratios. SmartSplat leverages image-aware features such as gradients and color variances, introducing a Gradient-Color Guided Variational Sampling strategy together with an Exclusion-based Uniform Sampling scheme to improve the non-overlapping coverage of Gaussian primitives in pixel space. In addition, we propose a Scale-Adaptive Gaussian Color Sampling method to enhance color initialization across scales. Through joint optimization of spatial layout, scale, and color initialization, SmartSplat efficiently captures both local structures and global textures using a limited number of Gaussians, achieving high reconstruction quality under strong compression. Extensive experiments on DIV8K and a newly constructed 16K dataset demonstrate that SmartSplat consistently outperforms state-of-the-art methods at comparable compression ratios and exceeds their compression limits, showing strong scalability and practical applicability. The code is publicly available at https://github.com/lif314/SmartSplat.