匹配距离与几何分布辅助学习的多视角点云配准
Matching Distance and Geometric Distribution Aided Learning Multiview Point Cloud Registration
AI总结：
本文针对多视角点云配准，提出利用匹配距离识别可靠配对以构建位姿图，并通过考虑几何分布和改进注意力的神经网络以数据驱动方式同步运动，在多个数据集上验证了有效性与泛化性。
摘要：
多视角点云配准在机器人、自动化和计算机视觉领域发挥着至关重要的作用。本文聚焦于多视角配准中的位姿图构建与运动同步。以往的位姿图构建方法通常对全连接图进行剪枝，或利用从局部描述子聚合得到的全局特征构建稀疏图，这些方法并不总能产生可靠的结果。为了在位姿图构建中识别可靠的配对，我们设计了一个网络模型，从点云对之间的匹配距离中提取信息。对于运动同步，我们提出了另一个神经网络模型，以数据驱动的方式计算绝对位姿，而不是优化不准确的手工设计的损失函数。我们的模型考虑了几何分布信息，并采用改进的注意力机制来促进灵活可靠的特征交互。在多种室内和室外数据集上的实验结果证实了我们方法的有效性和泛化能力。源代码可在 https://github.com/Shi-Qi-Li/MDGD 获取。
英文摘要：
Multiview point cloud registration plays a crucial role in robotics, automation, and computer vision fields. This paper concentrates on pose graph construction and motion synchronization within multiview registration. Previous methods for pose graph construction often pruned fully connected graphs or constructed sparse graph using global feature aggregated from local descriptors, which may not consistently yield reliable results. To identify dependable pairs for pose graph construction, we design a network model that extracts information from the matching distance between point cloud pairs. For motion synchronization, we propose another neural network model to calculate the absolute pose in a data-driven manner, rather than optimizing inaccurate handcrafted loss functions. Our model takes into account geometric distribution information and employs a modified attention mechanism to facilitate flexible and reliable feature interaction. Experimental results on diverse indoor and outdoor datasets confirm the effectiveness and generalizability of our approach. The source code is available at https://github.com/Shi-Qi-Li/MDGD.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
