Sequence matching enhanced 3D place recognition using candidate rearrangement

作     者：Chi Li Fei Yan Yan Zhuang 

作者机构：School of Control Science and EngineeringDalian University of TechnologyDalianChina 

出 版 物：《IET Cyber-Systems and Robotics》 (智能系统与机器人（英文）)

年 卷 期：2022年第4卷第3期

页      面：189-199页

核心收录：

学科分类：08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：National Natural Science Foundation of China,Grant/Award Numbers:No.62173056,No.U1913201 Central University Basic Research Fund of China,Grant/Award Number:DUT21LAB114 National key research and development program of China,Grant/Award Number:2020YFC1511704 

主　　题：matching Deep recognition 

摘      要：Deep‐learning‐based 3D place recognition has received more attention since the data‐driven fashion is widely used for the 3D point cloud *** of the existing deep‐learning‐based 3D place recognition methods only utilise a single scene for place ***,a single scene may have measurement noise or observable dy-namic object differences,which may lead to a reduction in recognition *** improve the performance of 3D place recognition,a sequence matching based rear-rangement method is *** sequence matching method is based on an assignment algorithm and guides the candidate rearrangement in searching for a similar *** global descriptor extraction adapts the effective sparse tensor representation and a simple pooling layer to obtain the global descriptor.A new loss function com-bination is employed to train the *** proposed approach is evaluated on the popular 3D place recognition benchmarks,which proves the effectiveness of the proposed approach.