A Hierarchical Compression Technique for 3D Gaussian Splatting Compression

TL;DR

The paper tackles the large storage demands of explicit 3D Gaussian Splatting representations used in novel view synthesis. It introduces Hierarchical GS Compression (HGSC), which prunes Gaussians by a joint global/local importance score, compresses geometry with an Octree, and performs hierarchical attribute compression via KD-tree blocks, FPS-based anchor primitives, RAHT for near-lossless anchor attributes, and LoD-based prediction with LZ77 residual encoding. HGSC achieves over $4.5\\times$ data size reduction and at least $6\\%$ BD-rate improvement across small and large scenes, with decoding times around 2 seconds, indicating strong practical potential for streaming and deployment. These results demonstrate a scalable compression framework that preserves rendering quality while dramatically reducing data size for 3D GS pipelines.

Abstract

3D Gaussian Splatting (GS) demonstrates excellent rendering quality and generation speed in novel view synthesis. However, substantial data size poses challenges for storage and transmission, making 3D GS compression an essential technology. Current 3D GS compression research primarily focuses on developing more compact scene representations, such as converting explicit 3D GS data into implicit forms. In contrast, compression of the GS data itself has hardly been explored. To address this gap, we propose a Hierarchical GS Compression (HGSC) technique. Initially, we prune unimportant Gaussians based on importance scores derived from both global and local significance, effectively reducing redundancy while maintaining visual quality. An Octree structure is used to compress 3D positions. Based on the 3D GS Octree, we implement a hierarchical attribute compression strategy by employing a KD-tree to partition the 3D GS into multiple blocks. We apply farthest point sampling to select anchor primitives within each block and others as non-anchor primitives with varying Levels of Details (LoDs). Anchor primitives serve as reference points for predicting non-anchor primitives across different LoDs to reduce spatial redundancy. For anchor primitives, we use the region adaptive hierarchical transform to achieve near-lossless compression of various attributes. For non-anchor primitives, each is predicted based on the k-nearest anchor primitives. To further minimize prediction errors, the reconstructed LoD and anchor primitives are combined to form new anchor primitives to predict the next LoD. Our method notably achieves superior compression quality and a significant data size reduction of over 4.5 times compared to the state-of-the-art compression method on small scenes datasets.

Figures (3)

• Figure 1: Framework of HGSC. $Q$ and $Q^{-1}$ denote the processes of quantization and dequantization, respectively.
• Figure 2: Cumulative distribution curves of importance.
• Figure 3: 2D example of RAHT