Low-Rank Approximation for Efficient Compression of Gaussian Splatting Spherical Harmonics
- 3 March 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (7) , 9225-9238
Abstract
3D Gaussian Splatting (3DGS) enables real-time, high-fidelity rendering but suffers from large model sizes, mainly due to the Spherical Harmonics (SH) coefficients used for view-dependent appearance modeling, whose parameter count scales quadratically (O(L2)) with the degree L. This paper presents the first systematic study that reveals and leverages the intrinsic low-rank structure of SH coefficients in 3DGS. Unlike prior approaches that truncate spectral energy, the proposed low-rank paradigm compactly preserves spectral information, achieving high visual quality with substantially reduced storage. Two complementary approaches are introduced. (1) SHAC-PCA (Principal Component Analysis) is a plug-and-play post-hoc compressor that retains principal spectral variance for high-fidelity compression. (2) SHAC-LST (Learned Subset Transformation) is a training-integrated approach that decomposes Alternating Current (AC) of SH coefficients into low-dimensional subset coefficients and a shared transformation matrix, offering superior compression and even quality improvements through regularization. Both methods effectively reduce SH coefficients storage complexity to O(L). Extensive experiments demonstrate that our approaches significantly reduce memory usage while maintaining or even improving rendering quality. The proposed techniques are highly versatile: SHAC-PCA can be applied to any pre-trained 3DGS model, while SHAC-LST supports end-to-end training or fine-tuning. Both methods are compatible with existing 3DGS compression pipelines, providing a practical and general solution for efficient compression of SH coefficients in 3DGS.Keywords
Funding Information
- National Natural Science Foundation of China (62571485)
This publication has 36 references indexed in Scilit:
- Compact 3D Scene Representation via Self-Organizing Gaussian Grids2024
- End-to-End Rate-Distortion Optimized 3D Gaussian Representation2024
- MesonGS: Post-training Compression of 3D Gaussians via Efficient Attribute Transformation2024
- BungeeNeRF: Progressive Neural Radiance Field for Extreme Multi-scale Scene Rendering2022
- D-NeRF: Neural Radiance Fields for Dynamic Scenes2021
- Deep low-rank subspace ensemble for multi-view clusteringInformation Sciences, 2019
- Structured Low-Rank Matrix Factorization for Point-Cloud Denoising2018
- The Unreasonable Effectiveness of Deep Features as a Perceptual Metric2018
- Lossy image compression using singular value decomposition and wavelet difference reductionDigital Signal Processing, 2014
- Random Fields on the Sphere2011