I am pursuing an MSc in Geodesy and Geoinformation at the Technical University of Munich (TUM), which I joined in September 2026.
Previously, I was a Mid-Level Technical Engineer at the Laboratory of Artificial Intelligence and Digital Economy (Shenzhen), working on UAV photogrammetry, visual SLAM and GNSS fusion, and large-scale 3D Gaussian Splatting. Before that, I worked as an Algorithm Engineer at XGRIDS on multi-sensor calibration and handheld SLAM.
I received my master's degree in Pattern Recognition and Intelligent Systems and my bachelor's degree in Geodesy and Geomatics Engineering from Wuhan University. During my master's studies, I worked at CVRS under Prof. Jian Yao. In summer 2022, I also worked with Prof. Lingjie Liu on neural surface reconstruction.
My interests include 3D Gaussian Splatting, neural radiance fields (NeRF), multi-view stereo (MVS), structure from motion (SfM), SLAM, and multi-sensor fusion for accurate 3D reconstruction.
MSc, Geodesy and Geoinformation
MEng, Pattern Recognition and Intelligent Systems
Undergraduate Exchange Program, Computer Science
BEng, Geodesy and Geomatics Engineering
Research in 3D reconstruction, photogrammetry, and camera calibration. Representative work is highlighted.
Haonan Dong, Zhijun Wang, Qingyun Song, Yuhan Tian, Ryan Ge
Japan Society for Photogrammetry and Remote Sensing, Fall Academic Conference, 2025, pp. 25–28.
Oral presentation on November 20, 2025, in Yamagata, Japan. Supported by the ZTE Industry–Academia Collaboration Fund Project.
Haonan Dong, Jian Yao, Ye Gong, Li Li, Shaosheng Cao, Yuxuan Li
The Photogrammetric Record, 38(183), 297–319, 2023.
Paper
Encoded target detection and iterative orthorectification for accurate fisheye camera calibration. Industry–university collaboration with DiDi Global.
Ye Gong, Pengwei Zhou, Yuyue Liu, Haonan Dong, Li Li, Jian Yao
The Photogrammetric Record, 38(183), 252–296, 2023.
Paper
View-graph key-subset extraction to address data redundancy and uneven distribution in SfM. University–enterprise collaboration with Huawei.
Haonan Dong, Jian Yao
arXiv preprint, 2022.
Paper
Robust unsupervised learning losses for multi-view stereo and weakly textured surface reconstruction.
Haonan Dong, Xiaotong Ye, Congpu Hao
Journal of Geomatics, 46(Sup. 1), 40–43, 2021. (In Chinese)
Paper
Indoor evacuation planning that prioritizes rooms near the fire origin and reduces congestion. Supported by the National Undergraduate Innovation Project.
Large-scale scene reconstruction using multi-modal fusion and 3D Gaussian Splatting.
Multi-sensor calibration and visual algorithms for handheld SLAM systems.
Joint optimization of RGB-D depth and camera poses to reconstruct high-quality indoor models with lightweight hardware.
Intrinsic and extrinsic calibration for multi-camera systems, including pinhole, fisheye, monocular, and stereo configurations.
Integrated NeuS into the instant-ngp framework, achieving a 6× speedup over the official implementation.
With Hongche Yin, Lei Luo, Han Fang, and Jian Yao.
A reconstruction pipeline producing depth maps and textured meshes from multi-view images or videos.
With Hongche Yin, Lei Luo, and Jian Yao.
Recovered real-scale sparse maps using printed ArUco markers and bundle adjustment for augmented reality applications.
Languages: English — C1 in listening, reading, speaking, and writing.
Paper reviewer: IROS 2022, ISPRS.
RGB 3D Reconstruction Pipeline — Reconstruct textured meshes from videos or RGB image sequences.
Depth Fusion — Fuse depth maps from multi-view stereo or sensors with known poses.
Awesome Indoor Reconstruction — A curated list of indoor reconstruction methods and datasets.