Ye's research aims to create next-generation perception systems capable of accurately modeling the physical world and reasoning about the behavior and interaction of various agents (e.g., humans, robots, and vehicles). Towards this goal, Ye has leveraged physics simulation, reinforcement learning, and generative models to solve computer vision problems such as human pose estimation, human motion synthesis, and multi-agent trajectory forecasting.
Ye is a fourth-year PhD student in the Robotics Institute at Carnegie Mellon University, working with Prof. Kris Kitani. Prior to this, he received his M.S. in computer science from CMU and B.E. in computer science from Zhejiang University. He has interned at Facebook Reality Lab Pittsburgh and Disney Research Pittsburgh.