Feng Xu

Feng Xu is an Associate Professor at the School of Software, Tsinghua University, Beijing, China. His research bridges computer science, medicine, and neuroscience, focusing on interdisciplinary applications in 3D vision, digital health, and human motion capture.

• Education: Ph.D. in Automation (2007-2012) and B.S. in Physics (2003-2007), both from Tsinghua University.
• Research Interests: 3D Vision, Computer Graphics, Digital Health, Physics-based Modeling, and Real-time Human Motion Tracking.

His work explores 3D vision and graphics through applications like IMU-based motion capture, NeRF editing, and facial geometry reconstruction. Recent publications emphasize integrating physics-aware models with deep learning for real-time performance, particularly in healthcare contexts such as mediastinal neoplasm diagnosis and radiology report generation.

Notable contributions include the EditableNeRF framework for topology editing and Transformer IMU Calibrator for dynamic sensor calibration. His work has received the Best Student Paper Award at IEEE ICME 2019.

Contact:

• Email: xufeng2003@gmail.com
• Email: feng-xu@tsinghua.edu.cn