Abstract
With the development of RGBD sensors, the high-quality color point cloud can be obtained expediently. In this paper, we propose a novel registration method for 3D color point clouds from different views, which is a critical issue in many applications. Different from traditional feature-based methods, we design a hybrid feature representation with color moments of the point, which could be applied naturally for any color point cloud. And these features are extracted from point clouds based on the supervoxel segmentation. By jointly conducting these features for similarity measure, a weight parameter is dynamically adapted between the color and the spatial information. The registration algorithm is under a classic iterative framework for building the correspondence and estimating transformation parameters. In addition, we provide a mutual correspondence matching condition with hybrid features to build some more robust relationships for estimating transformation parameters. Experimental results demonstrate that our method can effectively reduce the number of point data for registration and achieve good matching results even in a poor initial condition.
Highlights
Point cloud registration is a classical problem in the fields of computer vision, computer graphics, and robotics
To align two point clouds with low overlapping parts, Chetverikov et al [6] proposed the trimmed Iterative Closest Point (ICP) (TrICP) algorithm, which introduced an overlapping ratio into the objective function to trim outliers
EXPERIMENTAL RESULTS we evaluate the algorithm in dealing with color point cloud registration with partial overlapping and poor initial positions
Summary
Point cloud registration is a classical problem in the fields of computer vision, computer graphics, and robotics. The goal of point cloud registration is to find an optimal spatial transformation to align two given 3D point clouds For solving this problem, the classic algorithm is the Iterative Closest Point (ICP) algorithm [5]. The correspondence is built by combining the similarity measure of color feature This easy way ignores the context information. More robust way is to extract some local appearance features, such as Scale Invariant Feature Transform (SIFT) feature [8], inspired by the feature extraction method used in image processing These features are useful for the point cloud with obvious local invariant features, whose registration results are depending on a few of robust feature points. The down-sampling is used to remain the center point in each supervoxel and calculate the color moments in the supervoxel region In this way, the point clouds become very sparse, but the local color and spatial features are still retained. A mutual correspondence matching based on the hybrid features is suggested to select the accurate correspondences between features
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
