无需SfM即可重建大场景3D高斯点云，精度更高。
ICP-3DGS: SfM-free 3D Gaussian Splatting for Large-scale Unbounded Scenes
- 用ICP+优化替代SfM估计相机位姿，适应大运动场景。
- 引入体素引导的稠密化策略，提升大场景重建质量。
- 适合无预设位姿的大规模户外/室内场景重建任务。
近年来，神经渲染方法如NeRF和3D高斯点云（3DGS）在场景重建与新视角合成方面取得显著进展。然而，这些方法严重依赖结构光（SfM）提供的预处理相机位姿和3D结构先验，在室外场景中获取困难。为此，本文提出将迭代最近点（ICP）与基于优化的精修相结合，实现大相机运动下的精确位姿估计；同时引入体素引导的场景稠密化策略，指导大规模场景的重建。实验表明，ICP-3DGS在室内与室外多尺度场景中，均在相机位姿估计与新视角合成上优于现有方法。代码已开源：https://github.com/Chenhao-Z/ICP-3DGS。
原文摘要 · Abstract (English)
In recent years, neural rendering methods such as NeRFs and 3D Gaussian Splatting (3DGS) have made significant progress in scene reconstruction and novel view synthesis. However, they heavily rely on preprocessed camera poses and 3D structural priors from structure-from-motion (SfM), which are challenging to obtain in outdoor scenarios. To address this challenge, we propose to incorporate Iterative Closest Point (ICP) with optimization-based refinement to achieve accurate camera pose estimation under large camera movements. Additionally, we introduce a voxel-based scene densification approach to guide the reconstruction in large-scale scenes. Experiments demonstrate that our approach ICP-3DGS outperforms existing methods in both camera pose estimation and novel view synthesis across indoor and outdoor scenes of various scales. Source code is available at https://github.com/Chenhao-Z/ICP-3DGS.
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。