提出新嵌入方式，让3D高斯点云更易被神经网络学习
Learning Unified Representation of 3D Gaussian Splatting
- 用连续子流形场建模高斯点，统一参数表达
- 使颜色与几何结构保持一致，避免数据依赖
- 适合做3D重建的深度学习系统，提升泛化能力
基于3D高斯点云（3DGS）的高效显式三维重建系统依赖于良好的向量化表示。尽管3DGS能实现快速且直观的三维重建，但其基于参数的表示方式难以作为特征被神经网络有效学习，尤其是对基于神经网络的模型而言。直接将原始高斯参数输入学习框架无法解决参数化的非唯一性与异质性问题，导致模型高度依赖具体数据。为此，本文提出一种基于连续子流形场的3DGS嵌入表示方法，充分捕捉高斯原语的内在信息，在保持颜色与几何结构的同时，确保映射唯一性和通道同质性，从而显著提升3DGS在神经网络中的可学习性。代码已开源。
原文摘要 · Abstract (English)
A well-designed vectorized representation is crucial for the learning systems natively based on 3D Gaussian Splatting. While 3DGS enables efficient and explicit 3D reconstruction, its parameter-based representation remains hard to learn as features, especially for neural-network-based models. Directly feeding raw Gaussian parameters into learning frameworks fails to address the non-unique and heterogeneous nature of the Gaussian parameterization, yielding highly data-dependent models. This challenge motivates us to explore a more principled approach to represent 3D Gaussian Splatting in neural networks that preserves the underlying color and geometric structure while enforcing unique mapping and channel homogeneity. In this paper, we propose an embedding representation of 3DGS based on continuous submanifold fields that encapsulate the intrinsic information of Gaussian primitives, thereby benefiting the learning of 3DGS. Implementation available at https://github.com/cilix-ai/gs-embedding
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。