PCGS: Progressive Compression of 3D Gaussian Splatting

2025-03-11

AI Summary

While 3D Gaussian Splatting (3DGS) enables high-quality real-time rendering, its massive parameter count incurs prohibitive storage and transmission overhead. Existing compression methods lack progressive decoding capability, hindering deployment in bandwidth-constrained, on-demand scenarios. This paper introduces PCGS—the first progressive compression framework tailored for 3DGS. It proposes a novel co-design mechanism integrating Gaussian count masking and attribute quantization to enable fine-grained, controllable progressive reconstruction. Additionally, it incorporates history-aware probabilistic prediction and context-adaptive entropy coding to significantly enhance multi-rate compression efficiency. Experiments demonstrate that PCGS achieves compression ratios comparable to state-of-the-art non-progressive methods while enabling dynamic bandwidth adaptation—thereby substantially improving resource utilization in streaming and edge-deployment settings.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Learning on the Edge & Model Compression
• Data Mining & Knowledge Management: Data Compression

Application Category

• 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
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

3D Gaussian Splatting (3DGS) achieves impressive rendering fidelity and speed for novel view synthesis. However, its substantial data size poses a significant challenge for practical applications. While many compression techniques have been proposed, they fail to efficiently utilize existing bitstreams in on-demand applications due to their lack of progressivity, leading to a waste of resource. To address this issue, we propose PCGS (Progressive Compression of 3D Gaussian Splatting), which adaptively controls both the quantity and quality of Gaussians (or anchors) to enable effective progressivity for on-demand applications. Specifically, for quantity, we introduce a progressive masking strategy that incrementally incorporates new anchors while refining existing ones to enhance fidelity. For quality, we propose a progressive quantization approach that gradually reduces quantization step sizes to achieve finer modeling of Gaussian attributes. Furthermore, to compact the incremental bitstreams, we leverage existing quantization results to refine probability prediction, improving entropy coding efficiency across progressive levels. Overall, PCGS achieves progressivity while maintaining compression performance comparable to SoTA non-progressive methods. Code available at: github.com/YihangChen-ee/PCGS.

Problem

• Addresses large data size of 3D Gaussian Splatting for practical use.
• Introduces progressive compression for efficient on-demand applications.
• Enhances fidelity and quality with adaptive Gaussian control.

Innovation

• Progressive masking strategy for incremental anchor incorporation
• Progressive quantization for finer Gaussian attribute modeling
• Entropy coding efficiency improvement using existing quantization results