用神经场压缩3D高斯点云，模型缩小91倍仍保持高清
NeuralGS: Bridging Neural Fields and 3D Gaussian Splatting for Compact 3D Representations
- 用小型MLP分簇编码高斯点属性，替代海量高斯点存储
- 在多个数据集上实现91倍模型压缩，视觉质量无损
- 适合需要轻量3D表示的部署场景，如移动端或实时应用
3D高斯点云（3DGS）虽能实现高质量渲染和快速生成，但需数百万个3D高斯点，导致存储与传输开销巨大。本文提出NeuralGS，一种简单高效的压缩方法，将原始3DGS转化为紧凑表示。观察发现，如NeRF等神经场可用仅几兆字节的MLP网络表达复杂3D场景。NeuralGS据此采用神经场表示，利用多层感知机（MLP）编码高斯点属性，即使大规模场景也仅需极小存储空间。为此，我们基于高斯点重要性得分作为拟合权重，采用聚类策略，为每簇高斯点分配独立的小型MLP。在多个数据集上的实验表明，该方法平均实现91倍模型尺寸压缩，且视觉质量未受损。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。