Xiaoyun Yuan is an Associate Professor at the Artificial Intelligence Institute, School of Computer Science, Shanghai Jiao Tong University. He was previously a postdoctoral researcher in the Sigma lab, Department of Electronic Engineering, Tsinghua University.He received his B.S. from the University of Science and Technology of China in 2014 and his Ph.D. from the Hong Kong University of Science and Technology in 2020. Dr. Yuan has published approximately 20 academic papers in leading journals and conferences, including Nature Biomedical Engineering, Nature Communications, Light: Science & Applications, IEEE TPAMI, and CVPR. His research interests lie in computer vision, computational imaging, photoacoustic tomography, and optical computing. Notably, his work on human brain functional photoacoustic tomography earned the Best Paper Award at SPIE Photonics West 2021.
Jun 1, 2026New
One paper has been accepted by Nature Communications 2026. Details will be announced after the official acceptance notification is received.
May 25, 2026New
We just launched RL Flappy Bird, an interactive open-source web lab for visualizing Reinforcement Learning (a study aid for our AI1201 students!).
Play hereOur GitHub
Mar. 31, 2026New
Our paper "LUCID: Learning Unified Control for Image Deflaring and Exposure Mastery in Nighttime Photography" has been accepted by SIGGRAPH 2026. Congratulations to first author Tingyu Yang (杨庭语，SJTU 大三).
Project Page
Feb. 21, 2026New
Our paper "3M-TI: High-Quality Mobile Thermal Imaging via Calibration-free Multi-Camera Cross-Modal Diffusion" has been accepted by CVPR 2026. Special congratulations to undergraduate Junzhe Wan (万骏哲, SJTU 大四) and Ph.D. candidate Minchong Chen.
Code & Data
Sep. 4, 2025
Our paper "Integrated lithium niobate photonics for sub-angstrom snapshot spectroscopy" has been accepted by Nature 2025.
Paper
Jul. 31, 2025
Our paper "GBR: Generative Bundle Refinement for High-fidelity Gaussian Splatting with Enhanced Mesh Reconstruction" has been accepted by IEEE Trans. on CSVT 2025.
Project Page
Jun. 26, 2025
Our paper "DMDiff: Degradation-Modeled Multipath Diffusion for Tunable Metalens Photography" has been accepted by ICCV 2025. Special congratulations to undergraduate Jiayi Zhu (朱嘉熠, SJTU 大三) and Ph.D. candidate Feiyu Ji.
Project Page
More News
Nov. 11, 2024
I joined the Artificial Intelligence Institute, Shanghai Jiao Tong University as an associate professor.
Visit Institute
Mar. 9, 2024
Our paper "SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction" has been accepted by CVPR 2024.
Paper
Jan. 24, 2024
Our paper "Photonic neuromorphic architecture for tens-of-task lifelong learning" has been accepted by Light: Science & Applications.
Paper
Oct. 25, 2023
Our paper "Training large-scale optoelectronic neural networks with dual-neuron optical-artificial learning" has been accepted by Nature Communications.
Paper
Follow Us
We run an official WeChat public account and Xiaohongshu account where we regularly share updates on our team's work and discuss cutting-edge research ideas. Welcome to follow us!
Bilibili
