Duotun Wang
HKUST(GZ), China.
"When you get there, you will find an unsuspecting people and a spacious land that God has put into your hands, a land that lacks nothing whatever" (Judges. 18:10).
I am a PhD candidate of Computational Media and Arts at The Hongkong University of Science and Technology (Guangzhou) and I am fortunate to be co-supervised by Prof. Mingming Fan and Prof. Zeyu Wang. During Apr. to Sept. 2024, I was a research intern mentored by Dr. Xiaohang Zhan at Tencent AI Lab.
Before returning to research, I gained industrial experience as a software engineer in Virtual Reality at ByteDance in 2023, and as a research engineer in Simulation Platform for AI at Xiaomi from 2021 to 2022.
I obtained a master’s degree in computer science at University of Maryland, College Park with a full scholarship in 2021 and a bachelor’s degree in computer science with an outstanding graduate award at Beijing Institute of Technology in 2019.
During my graduate study, I was supervised by Prof. Huaishu Peng and worked as research intern within Document Intelligent Lab at Adobe.
During Mar. to Jul. 2019, I was a research intern in the Team of User Experience Design(UXD) at Lenovo Research. In 2016 - 2018, I pursued research experiences under the guidance of Prof. Wei Liang.
My research interests include differential geometry for AI-generated content and interactive design in mixed reality.
selected publications
- DesignMemo
DesignMemo: Integrating Discussion Context into Online Collaboration with Enhanced Design Rationale TrackingProceedings of the ACM on Human-Computer Interaction, 2025 - Text2VDM
Text2VDM: Text to Vector Displacement Maps for Expressive and Interactive 3D SculptingInternational Conference on Computer Vision (ICCV), 2025 - FocalSelect
FocalSelect: Improving Occluded Objects Acquisition with Heuristic Selection and Disambiguation in Virtual RealityIEEE Transactions on Visualization and Computer Graphics, 2025 - DEGAS
DEGAS: Detailed Expressions on Full-Body Gaussian AvatarsInternational Conference on 3D Vision (3DV), 2025 - DreamMapping
DreamMapping: High-Fidelity Text-to-3D Generation via Variational Distribution MappingPacific Graphics (PG), 2024 - HeadEvolver
HeadEvolver: Text to Head Avatars via Expressive and Attribute-Preserving Mesh DeformationInternational Conference on 3D Vision (3DV), 2025 - SplattingAvatar
SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian SplattingComputer Vision and Pattern Recognition (CVPR), 2024
