GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting

TL;DR

GaussianSpa is a novel optimization framework that simplifies 3D Gaussian Splatting models, reducing memory usage while maintaining high rendering quality, by alternately optimizing and sparsifying the Gaussian components during training.

Contribution

We propose a new optimization-based simplification method for 3D Gaussian Splatting that enhances compactness and quality, outperforming existing approaches.

Findings

Achieves 0.9 dB higher PSNR on the Deep Blending dataset.

Uses 10 times fewer Gaussians than vanilla 3DGS.

Demonstrates superior performance across various datasets.

Abstract

3D Gaussian Splatting (3DGS) has emerged as a mainstream for novel view synthesis, leveraging continuous aggregations of Gaussian functions to model scene geometry. However, 3DGS suffers from substantial memory requirements to store the multitude of Gaussians, hindering its practicality. To address this challenge, we introduce GaussianSpa, an optimization-based simplification framework for compact and high-quality 3DGS. Specifically, we formulate the simplification as an optimization problem associated with the 3DGS training. Correspondingly, we propose an efficient "optimizing-sparsifying" solution that alternately solves two independent sub-problems, gradually imposing strong sparsity onto the Gaussians in the training process. Our comprehensive evaluations on various datasets show the superiority of GaussianSpa over existing state-of-the-art approaches. Notably, GaussianSpa achieves an average PSNR improvement of 0.9 dB on the real-world Deep Blending dataset with 10$\times$ fewer Gaussians compared to the vanilla 3DGS. Our project page is available at https://noodle-lab.github.io/gaussianspa/.