Structured Image-based Coding for Efficient Gaussian Splatting Compression

2026-01-20

AI Summary

This work addresses the challenge of deploying Gaussian Splatting models in practical multimedia systems due to their large parameter count and high storage overhead. To this end, we propose GSICO, a novel encoder that reformulates Gaussian parameters into a structured image representation and introduces a spatial coherence enhancement algorithm to strengthen local correlations, thereby enabling compatibility with standard image codecs for efficient compression. Experimental results demonstrate that our method achieves an average compression ratio of 20.2× across multiple benchmark datasets while incurring minimal degradation in PSNR, SSIM, and LPIPS metrics. The rate-distortion performance significantly outperforms existing Gaussian Splatting compression approaches.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Representation Learning for Vision

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Gaussian Splatting (GS) has recently emerged as a state-of-the-art representation for radiance fields, combining real-time rendering with high visual fidelity. However, GS models require storing millions of parameters, leading to large file sizes that impair their use in practical multimedia systems. To address this limitation, this paper introduces GS Image-based Compression (GSICO), a novel GS codec that efficiently compresses pre-trained GS models while preserving perceptual fidelity. The core contribution lies in a mapping procedure that arranges GS parameters into structured images, guided by a novel algorithm that enhances spatial coherence. These GS parameter images are then encoded using a conventional image codec. Experimental evaluations on Tanks and Temples, Deep Blending, and Mip-NeRF360 datasets show that GSICO achieves average compression factors of 20.2x with minimal loss in visual quality, as measured by PSNR, SSIM, and LPIPS. Compared with state-of-the-art GS compression methods, the proposed codec consistently yields superior rate-distortion (RD) trade-offs.

Problem

• Gaussian Splatting
• compression
• radiance fields
• storage efficiency
• multimedia systems

Innovation

• Gaussian Splatting
• image-based compression
• structured parameter mapping
• spatial coherence
• rate-distortion optimization