Reliable-loc: Robust sequential LiDAR global localization in large-scale street scenes based on verifiable cues

Xianghong Zou, Jianping Li†, Weitong Wu, Fuxun Liang, Bisheng Yang†, Zhen Dong

ISPRS Journal of Photogrammetry and Remote Sensing (IF: 10.6) 2025

We propose a LiDAR-based reliable global localization method, Reliable-loc, which achieves better robustness in complex large-scale outdoor scenes with insufficient features and incomplete coverage of the prior map. The experimental results indicate that Reliable-loc exhibits high robustness, accuracy, and efficiency in large-scale, complex street scenes, with a position accuracy of ±2.91 m, yaw accuracy of ±3.74 degrees, and achieves real-time performance.

[Project] [Papar] [Code]

PatchAugNet: Patch feature augmentation-based heterogeneous point cloud place recognition in large-scale street scenes

Xianghong Zou, Jianping Li†, Yuan Wang, Fuxun Liang, Weitong Wu, Haiping Wang, Bisheng Yang†, Zhen Dong

ISPRS Journal of Photogrammetry and Remote Sensing (IF: 12.7) 2023

We propose PatchAugNet, which utilizes patch feature augmentation and adaptive pyramid feature aggregation to achieve better performance and generalizability for Heterogeneous Point Cloud-based Place Recognition (PCPR) tasks. The comprehensive experimental results indicate that PatchAugNet achieves SOTA performance with 83.43% recall@top1% and 60.34% recall@top1 on unseen large-scale street scenes, outperforming existing SOTA PCPR methods by +9.57 recall@top1% and +15.50 recall@top1, while exhibiting better generalizability.

[Project] [Paper] [Code]

MuCoGraph: A Multi-scale Constraint Enhanced Pose Graph Framework for MLS Point Cloud Inconsistency Correction

Yuhao Li Zou, Xianghong Zou, Tian Li, Sihan Sun, Yuan Wang, Fuxun Liang, Jianping Li†, Bisheng Yang†, Zhen Dong

ISPRS Journal of Photogrammetry and Remote Sensing (IF: 12.7) 2023

We present MuCoGraph, which introduces multi-scale constraints to establish the correct correspondences for revisited areas, and formulates an enhanced pose graph for position inconsistency correction. The proposed method was used to correct the MLS point cloud position inconsistency in three datasets. The average three-dimensional distance of the checkpoints was reduced from 0.362 m, 0.108 m, and 1.027 m to 0.057 m, 0.033 m, and 0.051 m for datasets I, II, and III respectively. In addition, the root-mean-square error of all three datasets was less than 0.04 m after correction. The experiments confirmed that the proposed method can automatically locate and correct the position inconsistency of MLS point clouds, showing good robustness and effectiveness.

[Paper]

Point cloud and image 3D bounding box labeling tool for autonomous driving

Xianghong Zou.

2018.06 - 2018.08

I developed a 3D bounding box labeling tool based on point clouds and images, serving for the research and development of object detection algorithms for autonomous driving.

Large scale point clouds and panorama images visualization based on WebGL

Xianghong Zou.

2016.06 - 2017.02

I developed a large scale point clouds and panorama images visualization based on WebGL. The system was built on Potree with ThreeJS, and it supports overlay display of point clouds and images, linkage with OSM map, and simple meansurement.

Electric Grid Unmanned Crowd Intelligence Inspection System Report Software

Guided by Prof. Bisheng Yang, Prof. Zhen Dong and Prof. Chi Chen from Wuhan University.

2021.02 - 2021.08

As a core developer, I was mainly responsible for software solution designing and implementing functions such as visualization of crossover detection results. The software has been scaled up in more than ten provinces and cities in China, generating significant social and economic benefits.

Point cloud data intelligent processing software: Point2Model

Guided by Prof. Bisheng Yang and Prof. Zhen Dong from Wuhan University.

2020.09 - 2021.12

As a core developer, I was mainly responsible for software solution design and framework construction. This software has been used for new basic surveying and mapping in multiple cities.