用3D高斯注册实现无姿态稀疏视图重建，精度高、泛化强。
RegGS: Unposed Sparse Views Gaussian Splatting with 3DGS Registration
- 基于前馈网络生成局部高斯，通过最优传输对齐到全局一致表示
- 在RE10K和ACID上实现高保真对齐与精准相机位姿估计
- 适合稀疏视图三维重建，尤其适用于无初始姿态数据
3D高斯点云（3DGS）已在无姿态图像场景重建中展现潜力。然而，基于优化的3DGS方法在稀疏视图下因先验知识不足而表现受限；前馈式高斯方法则受输入格式限制，难以融合更多视图。为此，我们提出RegGS，一种基于3D高斯注册的框架，用于重建无姿态稀疏视图。RegGS将前馈网络生成的局部3D高斯对齐至全局一致的3D高斯表示。技术上，我们采用熵正则化Sinkhorn算法高效求解最优传输混合2-沃瑟斯坦（MW₂）距离，作为Sim(3)空间中高斯混合模型（GMMs）的对齐度量。此外，我们设计了联合3DGS注册模块，整合MW₂距离、光度一致性与深度几何信息，实现粗到精的注册过程，同时准确估计相机位姿并完成场景对齐。在RE10K和ACID数据集上的实验表明，RegGS能以高保真度对齐局部高斯，实现精确的位姿估计与高质量的新视角合成。
原文摘要 · Abstract (English)
3D Gaussian Splatting (3DGS) has demonstrated its potential in reconstructing scenes from unposed images. However, optimization-based 3DGS methods struggle with sparse views due to limited prior knowledge. Meanwhile, feed-forward Gaussian approaches are constrained by input formats, making it challenging to incorporate more input views. To address these challenges, we propose RegGS, a 3D Gaussian registration-based framework for reconstructing unposed sparse views. RegGS aligns local 3D Gaussians generated by a feed-forward network into a globally consistent 3D Gaussian representation. Technically, we implement an entropy-regularized Sinkhorn algorithm to efficiently solve the optimal transport Mixture 2-Wasserstein $(\text{MW}_2)$ distance, which serves as an alignment metric for Gaussian mixture models (GMMs) in $\mathrm{Sim}(3)$ space. Furthermore, we design a joint 3DGS registration module that integrates the $\text{MW}_2$ distance, photometric consistency, and depth geometry. This enables a coarse-to-fine registration process while accurately estimating camera poses and aligning the scene. Experiments on the RE10K and ACID datasets demonstrate that RegGS effectively registers local Gaussians with high fidelity, achieving precise pose estimation and high-quality novel-view synthesis. Project page: https://3dagentworld.github.io/reggs/.
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。