EntropyGS: An Efficient Entropy Coding on 3D Gaussian Splatting

2025-08-13

AI Summary

To address the high storage and transmission overhead of Gaussian parameters and low coding efficiency in 3D Gaussian Splatting (3DGS), this paper proposes the first end-to-end entropy coding compression framework tailored for 3DGS. We make the novel observation that spherical harmonic coefficients (AC) follow a Laplacian distribution, while other attributes—rotation, scaling, and opacity—are well-modeled by mixture Gaussians; moreover, AC exhibits weak correlation with these attributes. Leveraging these insights, we design a factorized, parameterized entropy model and introduce an attribute-adaptive quantization scheme. Evaluated on multiple standard benchmarks, our method achieves ~30× compression ratio while preserving reconstruction quality nearly identically to the original 3DGS (ΔPSNR < 0.1 dB, SSIM unchanged). Encoding and decoding operate at millisecond latency, significantly outperforming both state-of-the-art neural compression and generic entropy coding approaches.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

As an emerging novel view synthesis approach, 3D Gaussian Splatting (3DGS) demonstrates fast training/rendering with superior visual quality. The two tasks of 3DGS, Gaussian creation and view rendering, are typically separated over time or devices, and thus storage/transmission and finally compression of 3DGS Gaussians become necessary. We begin with a correlation and statistical analysis of 3DGS Gaussian attributes. An inspiring finding in this work reveals that spherical harmonic AC attributes precisely follow Laplace distributions, while mixtures of Gaussian distributions can approximate rotation, scaling, and opacity. Additionally, harmonic AC attributes manifest weak correlations with other attributes except for inherited correlations from a color space. A factorized and parameterized entropy coding method, EntropyGS, is hereinafter proposed. During encoding, distribution parameters of each Gaussian attribute are estimated to assist their entropy coding. The quantization for entropy coding is adaptively performed according to Gaussian attribute types. EntropyGS demonstrates about 30x rate reduction on benchmark datasets while maintaining similar rendering quality compared to input 3DGS data, with a fast encoding and decoding time.

Problem

• Compress 3D Gaussian Splatting data efficiently
• Analyze statistical distributions of Gaussian attributes
• Maintain rendering quality while reducing storage

Innovation

• Factorized entropy coding for 3D Gaussian attributes
• Laplace distribution modeling for spherical harmonics
• Adaptive quantization by attribute types