01
FastAvatar
Instant 3D Gaussian Splatting avatars from a single unconstrained pose — an encoder–decoder that predicts a full 3DGS head in ~3 seconds, ready for novel-view synthesis and real-time animation.
Building 3D digital humans — fast, faithful avatars from a single photo with Gaussian Splatting & generative models.
I am an ECE PhD candidate at Rice University, advised by Prof. Guha Balakrishnan. Before Rice, I spent three wonderful years at Carnegie Mellon University, where I was fortunate to work closely with Prof. Gauri Joshi, Prof. Bhiksha Raj, and Prof. Giulia Fanti. I did my undergraduate studies at UESTC.
My research focuses on generative modeling, 3D human reconstruction, vision–language models, and federated learning. I like building practical systems that are simple, fast, fair, and reliable.
My research is supported by Samsung Research America. I am currently a research intern with the Apple Digital Human team.