A cylindrical shape descriptor for registration of unstructured point clouds from real-time 3D sensors
- PDF / 2,961,211 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 43 Downloads / 218 Views
SPECIAL ISSUE PAPER
A cylindrical shape descriptor for registration of unstructured point clouds from real‑time 3D sensors Yu He1 · Shengyong Chen2 · Hongchuan Yu3 · Thomas Yang4 Received: 17 March 2020 / Accepted: 6 October 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract To deal with data sets from real-time 3D sensors of RGB-D or TOF cameras, this paper presents a method for registration of unstructured point clouds. We firstly derive intrinsic shape context descriptors for 3D data organization. To replace the Fast-Marching method, a vertex-oriented triangle propagation method is applied to calculate the ’angle’ and ’radius’ in descriptor charting, so that the matching accuracy at the twisting and folding area is significantly improved. Then, a 3D cylindrical shape descriptor is proposed for registration of unstructured point clouds. The chosen points are projected into the cylindrical coordinate system to construct the descriptors. The projection parameters are respectively determined by the distances from the chosen points to the reference normal vector, and the distances from the chosen points to the reference tangent plane and the projection angle. Furthermore, Fourier transform is adopted to deal with orientation ambiguity in descriptor matching. Practical experiments demonstrate a satisfactory result in point cloud registration and notable improvement on standard benchmarks. Keywords Cylindrical shape descriptor · Unstructured point cloud · 3D registration · RGB-D data · Depth image
1 Introduction Three-dimensional (3D) real-time sensors such as RGB-D sensors or TOF cameras, are emerging in recent years and found wide applications, e.g. augmented reality on mobile devices. Point cloud registration is a fundamental task for * Shengyong Chen [email protected] Yu He [email protected] Hongchuan Yu [email protected] Thomas Yang [email protected] 1
Department of Computer Science, Tsinghua University, Beijing 100084, China
2
School of Computer Science and Engineering, Tianjin University of Technology, Tianjin 300384, China
3
National Centre for Computer Animation, Bournemouth University, Poole BH12 5BB, UK
4
Department of Electrical Engineering and Computer Science, Embry-Riddle Aeronautical University, Daytona Beach, Florida 32114, USA
data fusion from different views. It is also a key technique in 3D reconstruction [12], object detection and recognition [10, 22]. In these applications, registration is indispensable for the point clouds obtained from different objects or views. With the rapid development of 3D sensing technology, a host of 3D sensing devices emerged, such as the Microsoft Kinect based on structured light [25] and the MESA 4500 time-of-flight (TOF) camera based on TOF technique [11]. The registration of unstructured point clouds obtained by these sensing devices has also become a critical problem to be solved. In the wide range of registration algorithms, feature-based methods have gained good performance and popularity in many applications. Accordin
Data Loading...
