ShanghaiTech University
Visual localization, 3D reconstruction, and event-based vision.
CV
Education
Selected Research
Match-Any-Events: Zero-Shot Motion-Robust Feature Matching Across Wide Baselines for Event Cameras
Transformer-based detector-free event matching from local tracking to wide-baseline, zero-shot generalization.
A Linear N-Point Solver for Structure and Motion from Asynchronous Tracks
A unified linear solver for asynchronous multi-view observations across global-shutter, rolling-shutter, and event cameras.
An N-Point Linear Solver for Line and Motion Estimation with Event Cameras
A general linear N-point solver with improved numerical stability and over 600× faster computation.
Motion-Aware Optical Camera Communication with Event Cameras
Dynamic visual markers and a complete screen-to-camera communication system supporting localization under camera motion.
A 5-Point Minimal Solver for Event Camera Relative Motion Estimation
Established a motion-geometry model for event-camera line features and designed a 5-point minimal solver to estimate camera translation velocity with known rotation.
Experience
Changan Automobile
ShanghaiTech University
Honors
Outstanding Student (Top 10%)
Merit Student (Top 5%)
Outstanding Thesis (Top 5%)
Second Prize, Shanghai Division, China Undergraduate Mathematical Contest in Modeling
First Prize, Provincial Division, National Olympiad in Informatics (NOIP)
Service & Skills
- Reviewer
- CVPR, NeurIPS, ECCV, BMVC, ICRA, IROS, IEEE TR-O, IEEE T-MECH, IEEE RA-L
- Languages
- C++, Python
- Tools
- PyTorch, ROS, NumPy, Eigen, Ceres, CMake, Git, COLMAP, Pixi