Chunran ZHENG

Chunran ZHENG | 郑纯然

I received my Ph.D. from The University of Hong Kong (HKU) in 2025, where I was supervised by Prof. Fu Zhang at the MaRS Lab. Prior to that, I obtained my B.Eng. degree in Automation from Xi'an Jiaotong University (XJTU) in 2020. Recently, I was chosen as TopMinds (天才少年) at the highest level and joined Huawei Intelligent Vehicle BU.

My research interests center on spatial intelligence powered by LiDAR-vision fusion. I have published 20 papers in leading journals and conferences, including T-RO, IJRR, RA-L, ICRA, and IROS. Among these, I am first or co-first author on five papers, two of which are in T-RO. For the full list, please visit my Google Scholar📈.

Beyond academia, I am passionate about open-source contributions. My research code on GitHub has collectively accumulated over 10.3k stars⭐. In particular, my first-authored FAST-LIVO series alone has surpassed 7.8k stars⭐ and garnered over 758k cross-platform video plays, setting trends in the field and becoming a widely adopted framework in both industry and academia.

Email / Github / Scholar / Zhihu / LinkedIn / Wiki

News

[03/2026] 🏆 I was honored to receive the 2025 IEEE Transactions on Robotics (T-RO) King-Sun Fu Memorial Best Paper Award [Email].
[12/2025] 🔓 Open-sourced Global-LVBA, a globally consistent LiDAR-Visual Bundle Adjustment (LVBA) system for post-refinement.
[09/2025] 🚀 Led the algorithm team to complete the R&D of the SHARE S20 3D scanner from scratch within three months, enabling product launch and achieving 1,500+ unit sales to date [Poster].
[06/2025] 📄 Two papers, GS-SDF and Mesh-Learner, are accepted for presentation at IROS 2025.
[06/2025] 🎉 Our paper "FAST-LIVO2 on ARM Platforms: LiDAR-Inertial-Visual Odometry with Efficient Memory and Computation" is accepted for publication in Robotics and Automation Letters (RA-L) [arXiv].
[05/2025] 🌟 I was honored to be selected as Huawei's TopMinds [Letter].
[05/2025] 🎉 Our paper "GS-LIVO: Real-Time LiDAR, Inertial, and Visual Multi-sensor Fused Odometry with Gaussian Mapping" is accepted for publication in Transactions on Robotics (T-RO) [arXiv].
[04/2025] 🔓 Open-sourced FAST-Calib, bringing factory calibration-room capabilities into the research lab.
[03/2025] 🔥 FAST-LIVO2 Ranks 4th 🏆 on IEEE T-RO's Most Popular List.
[01/2025] 📄 Three papers, M2Mapping, iBTC, and LAMM, are accepted for presentation at ICRA 2025.
[10/2024] 🎉 Our paper "FAST-LIVO2: Fast, Direct LiDAR-Inertial-Visual Odometry" is accepted for publication in Transactions on Robotics (T-RO) [arXiv].

Academic Service

Reviewer of:

• Journals: IEEE Robotics and Automation Letters (RAL) * 23, IEEE Transactions on Robotics (TRO) * 9, International Journal of Robotics Research (IJRR) * 3, IEEE Transactions on Image Processing (TIP) * 2, International Journal of Computer Vision (IJCV) * 2

• Conferences: IEEE International Conference on Robotics and Automation (ICRA), IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

	FAST-Calib: LiDAR-Camera Extrinsic Calibration in One Second Chunran Zheng, Fu Zhang In Revision Paper / Code An efficient target-based extrinsic calibration tool for LiDAR-camera systems (e.g., FAST-LIVO2), making extrinsic calibration as straightforward as intrinsic calibration.
	FAST-LIVO: Fast and Tightly-coupled Sparse-Direct LiDAR-Inertial-Visual Odometry Chunran Zheng, Qingyan Zhu, Wei Xu, Xiyuan Liu, Qizhi Guo, Fu Zhang IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022 Paper / Code / Bilibili / YouTube A fast, sparse-direct LiDAR-inertial-visual odometry system built upon two tightly-coupled and direct odometry subsystems: a VIO subsystem and a LIO subsystem.
	FAST-LIVO2: Fast, Direct LiDAR-Inertial-Visual Odometry Chunran Zheng, Wei Xu, et al IEEE Transactions on Robotics (TRO), Best Paper Award, 2025 Paper / Code / Bilibili / YouTube FAST-LIVO2 is a supercharged upgrade of FAST-LIVO, integrating visual and LiDAR data within a single unified voxel map. The system achieves pixel-level reconstruction and robust onboard state estimation, even under severe LiDAR degeneration. It further demonstrates its capability through diverse downstream applications, including textured mesh, 3DGS, navigation, and airborne mapping, establishing a solid foundation for LiDAR-visual spatial AI.
	FAST-LIVO2 on ARM: State Estimation with Efficient Memory and Computation Bingyang Zhou, Chunran Zheng (co-first author), Ziming Wang, Fangcheng Zhu, Yixi Cai, Fu Zhang IEEE Robotics and Automation Letters (RA-L), 2025 Paper A lightweight adaptation of FAST-LIVO2 for ARM platforms (e.g., RK3588 and Orin NX).
	GS-LIVO: Real-Time LiDAR, Inertial, and Visual Multi-sensor Fused Odometry with Gaussian Mapping Sheng Hong, Chunran Zheng (co-first author), Fu Zhang, Tong Qin, Shaojie Shen IEEE Transactions on Robotics (TRO), 2025 Paper / Code The first real-time Gaussian-based LIVO system enabling photorealistic and high-fidelity scene representation.
	LiDAR-Inertial-Visual Handhold and UAV Device Chunran Zheng, Sheng Hong, Tianyong Ye, Junjie He Open-source project Code Hardware-synchronized Handheld and UAV device for FAST-LIVO, delivering the first proven open-source solution for Modular 3D Scanner.

News

Featured Research

Academic Service