I am a researcher working at the intersection of computer vision and AI for science. I earned an M.S. in Mathematics from Harbin Institute of Technology (Shenzhen), after completing a B.S. in Information and Computing Science at Fuzhou University.
My work focuses on building learning systems that remain reliable in difficult, data-constrained, or physically complex settings. This includes robust 3D human pose estimation, robot-assisted medical imaging, and physics-informed AI agents for scientific analysis.
How I work
I enjoy moving between mathematical formulation, model design, evaluation, and engineering delivery. Across research and industry collaborations, I have built computer-vision models, scientific data pipelines, evaluation systems, and end-to-end AI research workflows.
Current interests
- Robust perception under occlusion and uncertainty
- Vision-guided robotics for medical imaging
- Foundation models and agents for scientific instruments
- Physics-informed learning for spectroscopy and materials research
Selected highlights
- PICFormer accepted by ICASSP 2026
- Gold Award for NexusNMR in the Bohr + SciMaster AI Research Agent Competition
- Research published in IEEE Transactions on Robotics
- Reviewer for NeurIPS, ICML, and AAAI
I am open to PhD opportunities, research collaborations, and R&D conversations involving reliable AI systems for science.