B.E. @ Tsinghua UniversityI'm pursuing an undergraduate degree at Tsinghua University, where I research 3D scene representation and reconstruction under the guidance of Professor Bin Wang. My work focuses on advancing geometric deep learning methods to bridge 3D computer vision and graphics for scalable digital applications.
") does not match the recommended repository name for your site ("").
", so that your site can be accessed directly at "http://".
However, if the current repository name is intended, you can ignore this message by removing "{% include widgets/debug_repo_name.html %}" in index.html.
",
which does not match the baseurl ("") configured in _config.yml.
baseurl in _config.yml to "".

Jiankun Zhong, Zitong Zhan, Quankai Gao, Haozhe Lou, Ziyu Chen, Jiageng Mao, Ulrich Neumann, Yue Wang
International Conference on Robotics & Automation (ICRA) 2026 Spotlight
We accelerate Structure-from-Motion by redesigning critical stages (bundle adjustment/global positioning) with GPU-optimized sparse Levenberg-Marquardt, achieving 40× speedup over COLMAP on 5k-image datasets while preserving high qualities.
Jiankun Zhong, Zitong Zhan, Quankai Gao, Haozhe Lou, Ziyu Chen, Jiageng Mao, Ulrich Neumann, Yue Wang
International Conference on Robotics & Automation (ICRA) 2026 Spotlight
We accelerate Structure-from-Motion by redesigning critical stages (bundle adjustment/global positioning) with GPU-optimized sparse Levenberg-Marquardt, achieving 40× speedup over COLMAP on 5k-image datasets while preserving high qualities.