Multi-instance Point Cloud Registration by Efficient Correspondence Clustering

We address the problem of estimating the poses of multiple instances of the source point cloud within a target point cloud. Existing solutions require sampling a lot of hypotheses to detect possible instances and reject the outliers, whose robustness and efficiency degrade notably when the number of instances and outliers increase. We propose to directly group the set of noisy correspondences into different clusters based on a distance invariance matrix. The instances and outliers are automatically identified through clustering. Our method is robust and fast. We evaluated our method on both synthetic and real-world datasets. The results show that our approach can correctly register up to 20 instances with an F1 score of 90.46% in the presence of 70% outliers, which performs significantly better and at least 10x faster than existing methods. (Source code: https://github.com/SITU-ViSYSlmulti-instant-reg).

1. Introduction

Three-dimensional point cloud registration [12] [48] [45] mainly focuses on estimating one single transformation between the source point cloud and the target point cloud. However, we may sometimes want to estimate multiple transformations between point clouds. For instance, we have a 3D scan of an object and may want to find the poses of the same objects on the table within the target point cloud as shown in Figure 1. This problem, named multi-instance point cloud registration here, has been less investigated in the literature. It is nontrivial to extend existing point cloud registration methods to solve this problem.