About Me
I’m open to new roles and collaborations. Feel free to reach out if my background aligns with your team’s goals.
I earned my doctoral degree from the Advanced Interactive Technologies lab within the Institute for Intelligent Interactive Systems at ETH Zurich, supervised by Prof. Otmar Hilliges and Prof. Marc Pollefeys. In February 2026, I successfully defended my thesis, "Advancing Digital Humans Toward Personalized Human-Centric AI." My doctoral committee included Prof. Marc Pollefeys (ETH Zurich), Prof. Siyu Tang (ETH Zurich), Prof. Ming-Hsuan Yang (UC Merced), and Prof. Lingjie Liu (University of Pennsylvania).
My research focuses on human-centric computer vision and machine perception powered by Generative AI. Prior to my PhD, I completed my MSc degree in Computer Science at ETH Zurich and BSc in Electrical Engineering at National Taiwan University. I have joined Meta Reality Labs in Redmond and Google XR in Zurich as a research scientist intern, working on diffusion models for human pose estimation and 3D human generation in practical industrial settings.
Research Interests
To enable advanced AI systems to interact naturally in our world, they must accurately perceive 3D human shape, state, and appearance. My research focuses on developing these capabilities in three intersecting domains of human-centric perception: reconstruction, human modeling, and pose estimation.
- • In reconstruction, I address the problem of recovering 3D geometry and appearance of humans and clothing from unconstrained 2D observations.
- • In human modeling, I focus on designing novel 3D representations to capture the dynamic deformations of humans and clothing in 4D.
- • In pose estimation, I work on developing real-world and causal pose estimation methods for challenging in-the-wild scenarios.
Beyond passive machine perception, I am excited about the emerging opportunity to actively reasoning about human intent and interacting with them. My long-term vision is to transfer the capabilities of 3D perception to future human-centric reasoning AI.