🙋♂️ About Me
I am currently a first-year Ph.D. candidate in the CVMI Lab at the University of Hong Kong, under the supervision of Prof. Xiaojuan Qi. Prior to that, I worked as a research assistant at the Shanghai Qi Zhi Institute, advised by Prof. Li Yi. Earlier, I received my M.E. degree from the Institute of Computing Technology, Chinese Academy of Sciences in Beijing, where I was advised by Prof. Lin Gao. I obtained my B.E. degree from Tongji University in Shanghai.
My research interests include vision-language models and physics-based motion synthesis. I am particularly interested in developing embodied agents that can reason and act in 3D environments through multimodal understanding. My long-term goal is to enable simulation-trained agents to generalize to real-world settings, by leveraging the structured representations and decision-making capabilities of vision-language models. Ultimately, I aim to bridge the gap between perceptual grounding and embodied control in complex physical environments.
📝 Publications
Learning Physics-Based Full-Body Human Reaching and Grasping from Brief Walking References
Yitang Li*, Mingxian Lin*, Zhuo Lin, Yipeng Deng, Yue Cao, Li Yi
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
Deformable Radial Kernel Splatting
Yihua Huang , Mingxian Lin, Yang-Tian Sun, Ziyi Yang, Xiaoyang Lyu, Yan-Pei Cao†, Xiaojuan Qi†
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
FreeMotion: MoCap-Free Human Motion Synthesis with Multimodal Large Language Models
Zhikai Zhang, Yitang Li, Haofeng Huang, Mingxian Lin, Li Yi
European Conference on Computer Vision (ECCV), 2024.
Compositor: Bottom-Up Clustering and Compositing for Robust Part and Object Segmentation
Ju He, Jieneng Chen, Mingxian Lin, Qihang Yu, Alan Yuille
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
TM-NET: Deep Generative Networks for Textured Meshes
Lin Gao, Tong Wu, Yu-Jie Yuan, Mingxian Lin, Yu-Kun Lai, Hao Zhang
ACM Transactions on Graphics (ACM SIGGRAPH Asia 2021).
Single Image 3D Shape Retrieval via Cross-Modal Instance and Category Contrastive Learning
Mingxian Lin, Jie Yang, He Wang, Yu-Kun Lai, Rongfei Jia, Binqiang Zhao, Lin Gao
IEEE/CVF International Conference on Computer Vision (ICCV), 2021.
📖 Educations
- 2024.09 – Present, The University of Hong Kong, Ph.D. in Electrical and Electronic Engineering, Hong Kong, China
- 2020.09 – 2023.07, Institute of Computing Technology, Chinese Academy of Sciences, M.E. in Electrical and Electronic, Beijing, China
- 2015.09 – 2020.06, Tongji University, B.E. in Computer Science and Technology, Shanghai, China
💻 Internships
- 2024.11 - 2025.11, Tencent, China