Communications in Information and Systems

Volume 23 (2023)

Number 1

TransLoc3D: point cloud based large-scale place recognition using adaptive receptive fields

Pages: 57 – 83

DOI: https://dx.doi.org/10.4310/CIS.2023.v23.n1.a3

Authors

Tian-Xing Xu (Tsinghua University, China)

Yuan-Chen Guo (Tsinghua University, China)

Zhiqiang Li (National Defence University of PLA, China)

Ge Yu (Chinese Academy of Sciences, China)

Yu-Kun Lai (Cardiff University, United Kingdom)

Song-Hai Zhang (Tsinghua University, China)

Abstract

Place recognition plays an essential role in the field of autonomous driving and robot navigation. Point cloud based methods mainly focus on extracting global descriptors from local features of point clouds. Despite having achieved promising results, existing solutions neglect the following aspects, which may cause performance degradation: (1) huge size difference between objects in outdoor scenes; (2) moving objects that are unrelated to place recognition; (3) long-range contextual information. We illustrate that the above aspects bring challenges to extracting discriminative global descriptors. To mitigate these problems, we propose a novel method named TransLoc3D, utilizing adaptive receptive fields with a pointwise reweighting scheme to handle objects of different sizes while suppressing noises, and an external transformer to capture longrange feature dependencies. As opposed to existing architectures which adopt fixed and limited receptive fields, our method benefits from size-adaptive receptive fields as well as global contextual information, and outperforms current state-of-the-arts with significant improvements on popular datasets.

The full text of this article is unavailable through your IP address: 185.93.229.3

This work was supported by the National Key Technology R&D Program (Project Number 2017YFB1002604), by the National Natural Science Foundation of China (Project Numbers 61772298, 61832016), by the Key Research Projects of the Foundation Strengthening Program under Grant No. 2020JCJQZD01412, by a Research Grant of Beijing Higher Institution Engineering Research Center, and by the Tsinghua–Tencent Joint Laboratory for Internet Innovation Technology.

Received 26 February 2022

Published 17 April 2023