用流匹配方法直接生成对齐点云，实现高效全局一致的3D点云配准。
Register Any Point: Scaling 3D Point Cloud Registration by Flow Matching
- 将配准问题转化为条件生成，通过连续速度场从噪声点云恢复目标场景。
- 在多视图跨域基准上实现最优零样本性能，支持不同视角数、尺度和传感器。
- 无需初始猜测，直接输出注册点云，兼顾效率与点级全局一致性。
点云配准将多个无姿态的点云对齐到统一参考系，是3D重建与机器人定位的核心步骤。本文将配准建模为条件生成：学习一个连续的逐点速度场，将含噪点云传输至已对齐场景，并从中恢复各视角的位姿。不同于以往先匹配对应点估计两两变换再优化位姿图的方法，本模型直接生成注册后的点云，兼具效率与点级全局一致性。通过扩大训练数据并引入测试时刚性约束，该方法在现有两两配准基准及提出的跨域多视图配准基准上均达到当前最优性能。其优异的零样本表现表明，该方法可泛化至不同视图数量、场景尺度与传感器模态，即使重叠度较低亦有效。代码已开源：https://github.com/PRBonn/RAP。
原文摘要 · Abstract (English)
Point cloud registration aligns multiple unposed point clouds into a common reference frame and is a core step for 3D reconstruction and robot localization without initial guess. In this work, we cast registration as conditional generation: a learned, continuous point-wise velocity field transports noisy points to a registered scene, from which the pose of each view is recovered. Unlike prior methods that perform correspondence matching to estimate pairwise transformations and then optimize a pose graph for multi-view registration, our model directly generates the registered point cloud, yielding both efficiency and point-level global consistency. By scaling the training data and conducting test-time rigidity enforcement, our approach achieves state-of-the-art results on existing pairwise registration benchmarks and on our proposed cross-domain multi-view registration benchmark. The superior zero-shot performance on this benchmark shows that our method generalizes across view counts, scene scales, and sensor modalities even with low overlap. Source code available at: https://github.com/PRBonn/RAP.
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。