统一求解点云配准与形状组装，无需对称标签
Rectified Point Flow: Generic Point Cloud Pose Estimation
- 用连续速度场引导点云向目标位置移动
- 在6个基准上达到新最好性能，尤其擅长对称结构
- 适合需要联合训练的复杂3D形状重建任务
我们提出矩形点流（Rectified Point Flow），将成对点云配准与多部件形状组装统一为一个条件生成问题。给定未对齐的点云，该方法学习一个连续的点级速度场，将噪声点逐步迁移至目标位置，从而恢复各部件姿态。相比以往需手动处理对称性的方法，本方法能内在地学习装配对称性，无需对称标签。结合专注于重叠区域的自监督编码器，该方法在六个涵盖成对配准与形状组装的基准上取得当前最优性能。值得注意的是，统一框架支持跨不同数据集的联合训练，有助于学习共享几何先验，进而提升精度。项目主页：https://rectified-pointflow.github.io/
原文摘要 · Abstract (English)
We introduce Rectified Point Flow, a unified parameterization that formulates pairwise point cloud registration and multi-part shape assembly as a single conditional generative problem. Given unposed point clouds, our method learns a continuous point-wise velocity field that transports noisy points toward their target positions, from which part poses are recovered. In contrast to prior work that regresses part-wise poses with ad-hoc symmetry handling, our method intrinsically learns assembly symmetries without symmetry labels. Together with a self-supervised encoder focused on overlapping points, our method achieves a new state-of-the-art performance on six benchmarks spanning pairwise registration and shape assembly. Notably, our unified formulation enables effective joint training on diverse datasets, facilitating the learning of shared geometric priors and consequently boosting accuracy. Project page: https://rectified-pointflow.github.io/.
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。