I'm a PhD student at University College London (UCL) VECG group, advised by Prof. He Wang and Prof. Niloy Mitra. My PhD study (10/2024-10/2028) is supported by the UCL Research Studentship. Before my PhD study, I worked at Adobe Research London and UCL with Dr. Duygu Ceylan, Prof. Niloy Mitra and Kinross Prof. Ray Dolan.
My research interests span Computer Graphics, Machine Learning, and Computational Physics, all of which are important for me to understand animation, learning, and physical systems. I hope to explore how intelligence learns to interact with a complex physical world and, based on that, to generate physically natural character animation. In my PhD study, I focus on physics-based character-fluid animation, such as human/humanoid swimming, diving, and artistic swimming.
Outside of research, I enjoy music, sports, and digital arts .
A reinforcement learning framework for physically-based human swimming, learned from a single reference motion. A compact, low-dimensional representation of the body–water force exchange lets one policy generalize to unseen goals, trajectories, flows, fluids, perturbations and body morphologies, across all four competitive strokes.
A 3D editing framework using generative AI. The advantages of ProteusNeRF are simple, interactive, fast and lightweight.