CompGS++: Compressed Gaussian Splatting for Static and Dynamic Scene Representation

2025-04-17

AI Summary

Gaussian splatting representations suffer from significant redundancy, leading to high storage and transmission overhead—hindering their practical deployment in 3D immersive visual communication. To address this, we propose a compact Gaussian primitive modeling framework that jointly optimizes fidelity and compression. Our method introduces, for the first time, a spatio-temporal joint prediction paradigm to eliminate inter-primitive redundancy, coupled with rate-distortion-constrained optimization to suppress intra-primitive parameter redundancy. The pipeline comprises raw prediction, rate-constrained optimization, compact encoding, residual quantization, and joint entropy modeling. Evaluated on multiple benchmark datasets, our approach achieves up to 3.2× higher volumetric compression ratio while maintaining state-of-the-art rendering quality (PSNR/SSIM).

Technology Category

• Computer Vision: 3D Computer Vision
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Gaussian splatting demonstrates proficiency for 3D scene modeling but suffers from substantial data volume due to inherent primitive redundancy. To enable future photorealistic 3D immersive visual communication applications, significant compression is essential for transmission over the existing Internet infrastructure. Hence, we propose Compressed Gaussian Splatting (CompGS++), a novel framework that leverages compact Gaussian primitives to achieve accurate 3D modeling with substantial size reduction for both static and dynamic scenes. Our design is based on the principle of eliminating redundancy both between and within primitives. Specifically, we develop a comprehensive prediction paradigm to address inter-primitive redundancy through spatial and temporal primitive prediction modules. The spatial primitive prediction module establishes predictive relationships for scene primitives and enables most primitives to be encoded as compact residuals, substantially reducing the spatial redundancy. We further devise a temporal primitive prediction module to handle dynamic scenes, which exploits primitive correlations across timestamps to effectively reduce temporal redundancy. Moreover, we devise a rate-constrained optimization module that jointly minimizes reconstruction error and rate consumption. This module effectively eliminates parameter redundancy within primitives and enhances the overall compactness of scene representations. Comprehensive evaluations across multiple benchmark datasets demonstrate that CompGS++ significantly outperforms existing methods, achieving superior compression performance while preserving accurate scene modeling. Our implementation will be made publicly available on GitHub to facilitate further research.

Problem

• Reduces data volume in 3D Gaussian splatting scenes
• Compresses static and dynamic 3D scenes efficiently
• Minimizes redundancy between and within Gaussian primitives

Innovation

• Compact Gaussian primitives reduce redundancy
• Spatial-temporal prediction modules minimize redundancy
• Rate-constrained optimization enhances compactness