NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations
This addresses storage efficiency for 3D scene rendering applications, but is incremental as it builds on existing 3DGS and neural field techniques.
The paper tackled the problem of high storage and transmission costs in 3D Gaussian Splatting (3DGS) by developing NeuralGS, a method that compresses 3DGS into a compact representation using neural fields, achieving a 91-times average model size reduction without harming visual quality.
3D Gaussian Splatting (3DGS) achieves impressive quality and rendering speed, but with millions of 3D Gaussians and significant storage and transmission costs. In this paper, we aim to develop a simple yet effective method called NeuralGS that compresses the original 3DGS into a compact representation. Our observation is that neural fields like NeRF can represent complex 3D scenes with Multi-Layer Perceptron (MLP) neural networks using only a few megabytes. Thus, NeuralGS effectively adopts the neural field representation to encode the attributes of 3D Gaussians with MLPs, only requiring a small storage size even for a large-scale scene. To achieve this, we adopt a clustering strategy and fit the Gaussians within each cluster using different tiny MLPs, based on importance scores of Gaussians as fitting weights. We experiment on multiple datasets, achieving a 91-times average model size reduction without harming the visual quality.