我们在两个大型室外数据集上评估了拟议的 MSPR-Net:KITTI 里程计数据集 [26] 和 NuScenes 数据集 [27]。 结果表明 MSPR-Net 取得了最先进的性能 总之,我们的主要贡献如下: 我们提出了一种名为 MSPR-Net 的新型点云注册网络,它在室外 LiDAR 数据集中实现了最先进的性能 我们提出了局部相似度估计模块和全局相似...
In this paper, we propose a multi-scale features-based point cloud registration network named MSPR-Net for large-scale outdoor LiDAR point cloud registration. The main motivation of the proposed MSPR-Net is that the features of two keypoints from a true correspondence must match in different ...
MSPR-Net is an encoder-decoder network, as shown in Figure 2. The input of MSPR-Net is a pair of point clouds 𝑃𝑆,𝑃𝑇∈ℝ𝑁×3PS,PT∈RN×3, where N is the number of points. Firstly, a Siamese multi-scale backbone is utilized to process the input data, and outputs the...