Motivated by the intuition that the critical step of localizing a 2D image in the corresponding 3D point cloud is establishing 2D-3D correspondence between them, we propose the first feature-based dense correspondence framework for addressing the image-to-point cloud registration problem, dubbed CorrI2P, which consists of three modules, i.e., feature embedding, symmetric overlapping region detection, and pose estimation through the established correspondence. Specifically, given a pair of a 2D image and a 3D point cloud, we first transform them into high-dimensional feature space and feed the resulting features into a symmetric overlapping region detector to determine the region where the image and point cloud overlap each other. Then we use the features of the overlapping regions to establish the 2D-3D correspondence before running EPnP within RANSAC to estimate the camera's pose. Experimental results on KITTI and NuScenes datasets show that our CorrI2P outperforms state-of-the-art image-to-point cloud registration methods significantly. We will make the code publicly available.

翻译：以直觉为动力,在相应的 3D 点云中将 2D 图像本地化的关键步骤是在它们之间建立 2D-3D 对应关系,我们提出了第一个基于地基的密集通信框架,用于解决图像到点云登记问题,称为CorrI2P, 由三个模块组成, 即: 特征嵌入、 对称重叠区域探测, 并通过固定通信进行估计。 具体地说, 如果有一对 2D 图像和 3D 点云, 我们首先将它们转换成高维特征空间, 并将由此产生的特征输入一个对称重叠区域探测器, 以确定图像和点云相互重叠的区域。 然后我们利用重叠区域的特征来建立 2D-3D 通信, 然后再在 RANSAC 内运行 EP 之前对相机的外观进行估计。 KITTI 和 NuScenes 数据集的实验结果显示, 我们的 CorrI2P 将显著地表现为最新图像到点的云层登记方法。