Scan Context: Egocentric Spatial Descriptor for Place Recognition within 3D Point Cloud Map

1、摘要

https://mp.weixin.qq.com/s/J1pVixk0rY_hxtDq4ry1JQ

2、步骤

2.1 以激光原点为中心，均匀划分周围空间

we first divide a 3D scan into azimuthal(方位的) and radial(径向的) bins in the sensor coordinate, but in an equally spaced manner as shown in Fig. 1(a).
Ns and Nr are the number of sectors and rings, respectively.