• 12/2024: Congratulations, a paper titled "BIM-Based Indoor Navigation Using End-to-End Visual Localization and ARCore" has been accepted in Transactions In Gis.
• 12/2024: Congratulations! Tang Shengjun has been invited to serve as a member of the editorial board for the journal Photogrammetric Engineering and Remote Sensing(PE&RS) of the American Society for Photogrammetry and Remote Sensing (ASPRS).
• 05/2024: Congratulations, a paper titled "Back to Geometry: Efficient Indoor Space Segmentation from Point Clouds by 2D-3D Geometry Constrains" has been accepted in International Journal of Applied Earth Observation and Geoinformation
• 08/2024: 祝贺，汤圣君老师获得国家自然科学基金（面上项目）的资助，项目题目为“多模态特征增强的室内点云三维场景图智能构建方法”。（Congratulations to Tang Shengjun on receiving funding from the National Natural Science Foundation of China (General Program) for the project titled "Multimodal Feature-Enhanced Intelligent Construction of Indoor 3D Scene Graphs from Point Clouds.）
• 07/2024: 祝贺，团队论文“面向室内空间智能的三维场景图表达与应用”被测绘学科顶级中文期刊《测绘学报》接收。（Congratulations, the team’s paper titled “Three-Dimensional Scene Graph Representation and Application for Intelligent Indoor Spaces” was accepted by the top-tier Chinese journal in surveying science and technology, Journal of Geodesy and Geomatics.）
• 05/2024: 祝贺，团队论文“BavsNeRF: Batch Active View Selection for Neural Radiance Field Using Scene Uncertainty”被CVPR Workshop条件接收。（Congratulations, the team's paper "BavsNeRF: Batch Active View Selection for Neural Radiance Field Using Scene Uncertainty" has been accepted to The First Workshop on Populating Empty Cities -- Virtual Humans for Robotics and Autonomous Driving at CVPR 2024.）
• 05/2024: Congratulations, Tang Shengjun's paper wins the Best Paper Award for the paper "Accurate Segmentation Method for Roadside Lampposts Based on Vehicle Mounted Lidar Point Clouds" at the ISPRS International Society for Photogrammetry and Remote Sensing Conference.
• 05/2024: Congratulations, a paper titled "TreeNet3D: A Large Scale Tree Benchmark for 3D Tree Modeling, Carbon Storage Estimation and Tree Segmentation" has been accepted in International Journal of Applied Earth Observation and Geoinformation
• 04/2024: 祝贺，汤圣君老师发表在《武汉大学学报(信息科学版)》论文《超体素随机森林与LSTM神经网络联合优化的室内点云高精度分类方法》，入选2023年度优秀论文。(Congratulations to Shengjun Tang on the publication of his paper "A High-Precision Indoor Point Cloud Classification Method Jointly Optimized by Super Voxel Random Forest and LSTM Neural Network" in Geomatics and Information Science of Wuhan University, which was selected as an outstanding paper of the year 2023.)
• 04/2024: 祝贺，团队论文“AsymFormer: Asymmetrical Cross-Modal Representation Learning for Mobile Platform Real-Time RGB-D Semantic Segmentation”被CVPR Workshop条件接收。（Congratulations, the team's paper "AsymFormer: Asymmetrical Cross-Modal Representation Learning for Mobile Platform Real-Time RGB-D Semantic Segmentation" has been conditionally accepted to the CVPR Workshop.）
• 02/2024: 祝贺，汤圣君老师获得亚热带建筑与城市科学全国重点实验室自主研究课题支持，项目题目为“城市卫星拒止环境车载系统高精度定位与三维测图研究”。（ Congratulations to Tang Shengjun for receiving support from State Key Laboratory of Subtropical Building and Urban Science for his independent research project titled "Research on High Precision Positioning and 3D Mapping of Urban Satellite Jamming Environmental Vehicle Systems".）

This paper proposes an as-built BIM-based indoor navigation using End-to-End visual localization and ARCore. The experimental results demonstrate that in publicly accessible datasets, the global positioning error consistently remains within the 20-centimeter range, with the majority of these datasets achieving a positioning error of 10 centimeters or less.

We introduce a new framework, OptiViewNeRF, which leverages scene uncertainty to guide the view selection process. Initially, an uncertainty estimation model of the entire scene is developed based on a preliminary NeRF model. This model then informs the selection of new perception viewpoints using a batch view selection strategy, allowing the entire process to be completed in a single iteration. By selecting viewpoints that provide informative data, this approach improves novel view synthesis results and accurately reconstructs 3D scenes.

This paper introduces an efficient indoor space segmentation metnod from point clouds. By shifting the focus back to geometric features, the method aims to demonstrate the effectiveness and generalization capabilities of geometry-based approaches in handling diverse indoor structures.

本文提出面向室内空间智能的三维场景图表达模型，系统性介绍了室内三维场景图在要素层级化组织、几何表达、语义描述、关系描述方法，建立了一个可统一描述室内要素几何、语义及关系的室内三维场景图概念模型。同时，该图模型可以与现有的三维场景表达方法进行融合表达，具有良好的数据兼容性。

This paper presents a novel fully automated approach for generating structured 3D synthetic tree models, addressing the limitations of existing datasets used in applications like digital twin construction, carbon stock calculation, and environmental assessments. The method allows for the automated creation of a large-scale dataset containing 13,000 tree models of ten common species, each featuring a detailed 3D point cloud with hierarchical structures, precise parameters, and separate branch and leaf information. The datasets are publicly available, accessible via the TreeNet3D Dataset link.

AsymFormer uses an asymmetrical backbone for multimodal feature extraction, reducing redundant parameters by optimizing computational resource distribution. To fuse asymmetric multimodal features, a Local Attention-Guided Feature Selection (LAFS) module is used to selectively fuse features from different modalities by leveraging their dependencies. Subsequently, a Cross-Modal Attention-Guided Feature Correlation Embedding (CMA) module is introduced to further extract cross-modal representations.

In this paper, we propose the TransCNNLoc framework, which consists of an encoding-decoding network designed to learn more robust image features for camera pose estimation.Experiments were conducted on the publicly available 7scenes dataset as well as a dataset collected under changing lighting conditions and dynamic scenes for accuracy validation and analysis. The experimental results demonstrate that the proposed TransCNNLoc framework exhibits superior adaptability to dynamic scenes and lighting changes.

In this paper, a novel method for fine-grained 3D tree model reconstruction from incomplete point clouds is proposed. This approach addresses the structural reconstruction of trees, tackling two main problems: reconstructing the main trunk and branches based on data completeness.

This paper proposed a SLAM method base on multiple ground constraints pose optimization (MGCPO) which uses a backpack LiDAR system. The proposed method includes two novel modules. The first, a regression analysis-based scenarios recognition (RASR) module provides a reference for the construction of ground constraints. The second, based on different scene detection results, the MGCPO module constrains the sensor pose using the floor plane to reduce localization errors and effectively decrease loop closure detection errors.

In this paper, we present a novel approach for robustly calibrating the extrinsic parameters of a solid-state(SS) lidar-camera system in a natural environment. We conducted a performance study to compare our proposed method against existing targetless calibration methods on various natural scenarios. The experimental results demonstrate that our proposed method achieves higher robustness, accuracy, and consistency, making it suitable for real-world applications.

In this study, a fully automatic method to generate synthetic noisy point clouds from as-built building information modeling (BIM) models is presented and it assesses the potential of these synthetic point clouds to improve deep neural network training. All simulation datasets are publicly available, including original BIM models, full synthetic point clouds, and point clouds after IHPR processing, accessible via the following link: BIMSyn Dataset.

本文面向虚拟地理环境高逼真场景构建需求，提出一种基于高精度激光扫描点云数据的树木三维模型高保真仿生重建方法，可实现形态特征保持的树木三维模型自动化重建。(In this paper, we propose a bionic reconstruction method for 3D tree models based on high-precision laser scanning point cloud data for building realistic scenes in virtual geographic environments, which enables the automated reconstruction of 3D tree models at multiple levels of detail while preserving morphological features.)

针对现有三维点云数据分割分类方法存在分类目标内部不一致的问题，提出一种超体素随机森林与长短期记忆神经网络(long short-term memory，LSTM)联合优化的室内点云高精度分类方法。 (In response to the problem of internal inconsistency in the classification targets of existing 3D point cloud data segmentation methods, we propose a high-precision indoor point cloud classification method that jointly optimizes super-voxel random forests and Long Short-Term Memory (LSTM) neural networks.)

In this paper, we proposed an unsupervised aquaculture ponds extraction method based on hyperspectral imagery super-resolution, feature fusion and stepwise extraction strategy.

We propose a method based on improved 3D morphological analysis for extracting street trees from mobile laser scanner (MLS) point clouds.

This paper presents a novel parametric modeling method for reconstructing semantic volumetric building interiors from the unstructured point cloud of a building. Unlike existing partitioning-based methods, our proposed method overcomes the limitations by providing a flexible framework for combining 3D Deep Learning and an improved morphological approach for inverse BIM modeling.

In this paper, we propose a robust and effective point cloud classification approach that integrates point cloud supervoxels and their locally convex connected patches into a random forest classifier, which effectively improves the point cloud feature calculation accuracy and reduces the computational cost.

RGB-D semantic segmentation algorithm is a key technology for indoor semantic map construction. The traditional RGB-D semantic segmentation network, which always suffer from redundant parameters and modules. In this paper, an improved semantic segmentation network PSCNet is designed to reduce redundant parameters and make models easier to implement. Based on the DeepLabv3+ framework, we have improved the original model in three ways, including attention module selection, backbone simplification, and Atrous Spatial Pyramid Pooling (ASPP) module simplification.

This paper proposes a deep cross-scale feature propagation network (CSNet), to effectively learn and fuse multi-scale features for robust semantic segmentation of indoor scene images. The proposed CSNet is deployed as an encoder-decoder engine. During encoding, the CSNet propagates contextual information across scales and learn discriminative multi-scale features, which are robust to large object scale variation and indoor occlusion. The decoder of CSNet then adaptively integrates the multi-scale encoded features with fusion supervision at all scales to generate target semantic segmentation prediction.

In this survey, we provide an overview of recent advances in indoor scene modeling methods, public indoor datasets and libraries which can facilitate experiments and evaluations, and some typical applications using RGB-D devices including indoor localization and emergency evacuation.

In this paper, we describe the trajectory drift–compensated strategy that we designed to eliminate the influence of time drift between sensors, remove the inconsistency between the sequences from various sensors, and thereby generate a coarse-to-fine procedure for robust camera-tracking based on two-dimensional–3D observations from stereo sensors.

We present a complete paradigm for pole-like street furniture segmentation and classification using mobile LiDAR (light detection and ranging) point cloud.

This paper presents a combined adjustment approach to integrate multi-source multi-resolution satellite imagery for improved geo-positioning accuracy without the use of ground control points (GCPs).

Considering HRSI and LiDAR datasets taken from metropolitan areas as a case study, this paper presents a novel approach to the geometric integration of HRSI and LiDAR data to reduce their inconsistencies and improve their geopositioning accuracy.

为实现大规模地形景观和精细工程设施模型在三维GIS中的无缝集成管理,并支持工程设施全生命周期的共享应用,提出了一种可根据设计参数自动建立复杂设施三维模型并交互式编辑修改的方法,扩展了三维GIS数据模型,实现了三维几何模型与其参数信息的有机集成与同步更新,并以桥梁模型的构建为例验证了该方法的可行性和有效性。(To realize the seamless integrated management of large-scale terrain landscape and detailed engineering facility models in 3D GIS, and to support the shared application of the whole lifecycle of engineering facilities, a method is proposed. This method can automatically build complex facility 3D models based on design parameters and allows interactive editing and modification. This method extends the 3D GIS data model, achieves the organic integration and synchronous update of the 3D geometric model and its parameter information. The feasibility and effectiveness of this method have been verified using the construction of a bridge model as an example.)

• 国家自然科学基金（面上基金），多模态特征增强的室内点云三维场景图智能构建方法 ⏬， 42471442， 2025/01/01 至 2028/12/31, RMB$460,000, 主持
• 深圳市科技计划-科技重大专项，城市数字孪生三维场景建模关键技术研究及应用示范-课题 ⏬， KJZD20230923115508017， 2024/01/01 至 2026/12/31, RMB$1800,000, 主持
• 广东省自然科学基金-青年提升项目，空地跨视角协同感知的移动车载平台精准定位与位姿优化方法 ⏬， 2024A1515030061， 2024/01/01 至 2026/12/31, RMB$300,000, 主持
• 亚热带建筑与城市科学全国重点实验室，自主研究课题，城市卫星拒止环境车载系统高精度定位与三维测图研究 ⏬, 2023ZB18， 2024/01/01 至 2026/12/31, RMB$30,000,0, 主持
• 中华人民共和国科学技术部，国家重点研发计划项目子课题，超大城市绿色低碳发展监测与诊断优化应用示范 ⏬, 2022YFB3903700， 2023/12 至 2026/11, RMB$58,800,0, 主持
• 广东省面上基金，联合视觉SLAM与深度神经网络的室内场景语义分类与自动化建模技术 ⏬，2021A1515012574，2021/01/01 至 2023/12/31，RMB$100,000，主持
• 深圳市基础研究面上项目，面向机器人自主导航的类脑视觉场景理解与实时位置计算方法 ⏬，JCYJ20210324093012033，2021/10/28 至 2024/10/27, RMB$600,000，主持
• 2020年佛山市促进高校科技成果服务产业发展扶持项目，室内全空间三维测图于标准化建模技术研究 ⏬，2020/10/13 至 2022/12/31,RMB$500,00，主持
• 中交第二航务工程勘察设计院有限公司，雄安新区孝义河倾斜摄影技术服务 ⏬，GT09200ZB(F006)，2019/01/01 至 2020/12/30,RMB$148,000，主持
• 国家自然科学基金（青年基金），多RGB-D传感器联合的在线室内高精度三维测图方法 ⏬，41801392，2019/01/01 至 2021/12/31, RMB$265,000，主持
• 中国博士后科学基金，多元特征混合优化的RGB-D室内高精度三维测图方法 ⏬，2018M633133，2018/05/01 至 2019/09/08,RMB$500,00，主持
• 深圳市科创委自由探索项目，基于便携式深度传感器的城市封闭/半封闭空间快速三维测图技术研究 ⏬，JCYJ20180305125131482，2019/01/01 至 2021/12/31, RMB$300,000，主持
• 自然资源部城市自然资源监测与仿真重点实验室开放基金, 基于室内高精度三维测图的BIM关键部件自动化重建方法 ⏬，KF-2019-04-010, 2020/01/17 至 2021/12/30，RMB$200,000，主持
• 武汉大学测绘遥感信息工程国家重点实验室开放基金，集成视觉与几何特征的RGB-D SLAM方法 ⏬，17E04，2018/01/01 至 2019/12/31, RMB$500,00，主持

• 2023年度自然资源科学技术奖，实景三维城市环境智能构建关键技术及应用 ⏬ ⏬ ⏬，一等奖，排名：2/15，2024。
• 论文《Accurate Segmentation Method for Roadside Lampposts Based onVehicle.Mounted Lidar Point Clouds》获得ISPRS TC1 Symposium Intelligent Sensing and Remote Sensing Application 国际会议最佳论文奖 ⏬, 2024.
• 发表在《武汉大学学报(信息科学版)》论文《超体素随机森林与LSTM神经网络联合优化的室内点云高精度分类方法》 ⏬，入选2023年度优秀论文, 2024.
• 论文“Robust Calibration of Vehicle Solid-state Lidar-Camera Perception System Using Line-Weighted Correspondences in Natural Environments”获得2023非暴露空间国际学术论坛最佳论文奖 ⏬, 2023.
• 论文“BIM generation from 3D point clouds by combining 3D deep learning and improved morphological approach”入选2023年深圳市第三届优秀科技学术论文成果集 ⏬。（"BIM generation from 3D point clouds by combining 3D deep learning and improved morphological approach" was selected into the 2023 Shenzhen Third Excellent Science and Technology Academic Papers Collection）, 2023.
• The Best Workshop Presentation Award at the ISPRS Geospatial Week 2023 International Conference for 'Tree-GPT: Modular Large Language Model Expert System for Forest Remote Sensing Image Understanding and Interactive Analysis' ⏬, 2023.
• The Second Place of the IPIN 2022 Competition ONLINE Track 2 -Camera in the TWELFTH INTERNATIONAL CONFERENCE ONINDOOR POSITIONING AND INDOORNAVIGATION held in BEIJING ⏬, CHINA, on 5-7 SEPTEMBER 2022.
• 深圳市高层次人才C类(C-Class High-Level Talents in Shenzhen City) ⏬, 2019.
• 广东省高校科技成果转化大赛电子信息组二等奖 ⏬，排名2/5，2020。
• 深圳市南山区"2020创业之星大赛"互联网与移动互联网初创组一等奖 ⏬，排名：1/3，2020。
• 全国优秀测绘工程奖铜奖 ⏬，排名：9/16，2022。
• 深圳大学教学成果奖 ⏬，一等奖，排名：11/12，2022。

Committee Member:

• 美国摄影测量与遥感学会会刊Photogrametry Engineering and Remote Sensing(PE&RS)期刊，编委, ⏬, From 2024
• Guest Editor for Remote Sensing, Special issue on “Remote Sensing for 2D/3D Mapping”, From 2023
• Guest Editor for IEEE JSTARS, Special issue on “High-fidelity Urban 3D Modeling and Scene Simulation”, From 2023
• Junior Editor for The Photogrammetry Record , From 2023
• 国际数字地球学会中国虚拟地理环境委员会委员(Member of the China Virtual Geographic Environment Committee of the International Society for Digital Earth), From 2022
• 中国图学学会BIM专委会委员 ⏬(Member of the BIM Special Committee of the Chinese Society for Graphics), From 2022
• 中国测绘学会位置服务工委会委员 ⏬(Member of the Location Service Working Committee of the Chinese Society for Surveying and Mapping), From 2022
• 国际数字地球学会全国激光雷达专业委员会委员 ⏬ (Member of the National LiDAR Professional Committee of the International Society for Digital Earth), From 2024
• Expert of ISO Technical Committee 59, From 2022

Journal Reviewer:

• ISPRS Journal of Photogrammetry and Remote Sensing
• The Photogrammetry Record
• IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
• Automation In Construction
• International Journal of Applied Earth Observation and Geoinformation
• IEEE Transactions on Industrial Informatics
• IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2023)