SectionKey: 3-D semantic point cloud descriptor for place recognition in large-scale environments

• Main Author: Jin, Shutong
• Other Authors: Wang Dan Wei
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
• Online Access: https://hdl.handle.net/10356/156767
• Institution: Nanyang Technological University

Place recognition is seen as a crucial factor to correct cumulative errors in Simultaneous Localization and Mapping (SLAM) applications. Most existing place recognition studies focus on vision-based approaches, which are sensitive to environmental changes such as illumination, weather, and seasons due to their inherent limitations. More recent attention has been attracted to use 3-D Light Detection and Ranging (LiDAR) scans, where most of the LiDAR-based methods work by leveraging accurate geometric information that demonstrates more credibility. However, these pure geometric-based methods are tend to fail in repetitive and ambiguous scenarios in the urban environments, which may result in low place recognition accuracy and efficiency. Considering these facts, this dissertation proposes a novel global descriptor, named SectionKey, which leverages both semantic and geometric information to tackle the problem of place recognition in large-scale urban environments. In addition to the robustness against moving objects, the proposed descriptor is also invariant to viewpoint changes and able to return translation to correct the drifting errors. Specifically, the encoded three-layers key serves as a pre-selection step and a ‘candidate center’ selection strategy is deployed before calculating the similarity score, thus improving the accuracy and efficiency significantly. Then, a two-step semantic iterative closest point (ICP) algorithm is applied to acquire the 3-D pose (x, y,q) that is used to align the candidate point clouds with the query frame, and calculate the similarity score. Extensive experiments have been conducted on public Semantic KITTI dataset and self-collected dataset in NTU, and the results demonstrate the superior performance of the proposed system over state of-the-art baselines.