I am a postdoctoral researcher at The Movement Lab of Stanford University, working with Prof. C. Karen Liu. My research interests include artificial intelligence, computer graphics, and computer vision with a focus on motion planning and reinforcement learning for physics-based character control and agent navigation. I am also interested in applying computer science techniques to other disciplines, like bioengineering and environmental science.
Before joining Stanford University, I was a research assistant professor at Clemson University, working at the Big Data Analytics Lab with Prof. Feng Luo. I received my Ph.D. in computer science from Clemson University under the supervision of Prof. Ioannis Karamouzas. Prior to that, I received an M.S. in electrical engineering from University of Minnesota at Twin Cities.
Publications
2023
AdaptNet: Policy Adaptation for Physics-Based Character Control
Pei Xu, Kaixiang Xie, Sheldon Andrews, Paul G. Kry, Michael Neff, Morgan McGuire, Ioannis Karamouzas, Victor Zordan
ACM Transactions on Graphics (Proceedings of SIGGRAPH Asia 2023).
Context-Aware Timewise VAEs for Real-Time Vehicle Trajectory Prediction
Pei Xu, Jean-Bernard Hayet, Ioannis Karamouzas
IEEE Robotics and Automation Letters.
Also going to appear in IEEE International Conference on Robotics and Automation, 2024.
Composite Motion Learning with Task Control
Pei Xu, Xiumin Shang, Victor Zordan, Ioannis Karamouzas
ACM Transactions on Graphics (Proceedings of SIGGRAPH 2023). In technical papers trailer.
Too Stiff, Too Strong, Too Smart: Evaluating Fundamental Problems with Motion Control Policies
Kaixiang Xie, Pei Xu, Sheldon Andrews, Victor Zordan, Paul G. Kry
PACM on Computer Graphics and Interactive Techniques.
In ACM SIGGRAPH/Eurographics Symposium on Computer Animation, 2023.
2022
SocialVAE: Human Trajectory Prediction using Timewise Latents
Pei Xu, Jean-Bernard Hayet, Ioannis Karamouzas
In the 17th European Conference on Computer Vision, 2022.
2021
PFPN: Continuous Control of Physically Simulated Characters using Particle Filtering Policy Network
Pei Xu, Ioannis Karamouzas
In ACM SIGGRAPH Conference on Motion, Interaction and Games, 2021. Best paper nomination.
Also in NeurIPS Deep Reinforcement Learning workshop, 2021.
Human Inspired Multi-Agent Navigation using Knowledge Distillation
Pei Xu, Ioannis Karamouzas
In IEEE/RSJ International Conference on Intelligent Robots and Systems, 2021.
A GAN-Like Approach for Physics-Based Imitation Learning and Interactive Character Control
Pei Xu, Ioannis Karamouzas
PACM on Computer Graphics and Interactive Techniques. Cover article.
In ACM SIGGRAPH/Eurographics Symposium on Computer Animation, 2021.
Bioengineering & Biomedical Engineering
Osteoarthritis and Cartilage OpenMask R-CNN Provides Efficient and Accurate Measurement of Chondrocyte Viability in the Label-Free Assessment of Articular Cartilage, Hongming Fan, Pei Xu, Xun Chen, Yang Li, Zhao Zhang, Jennifer Hsu, Michael Le, Emily Ye, Bruce Gao, Harry Demos, Hai Yao, Tong Ye, 2023SPIE BiOS 2022Automated Chondrocyte Viability Analysis of Articular Cartilage based on Deep Learning Segmentation and Classification of Two-Photon Microscopic Images, Hongming Fan, Pei Xu, Michael Le, Jennifer Hsu, Xun Chen, Yang Li, Zhao Zhang, Bruce Gao, Shane Woolf, Tong Ye.
Prior Work
MIG 2019Low Dimensional Motor Skill Learning Using Coactivation, Avinash Ranganath, Pei Xu, Ioannis Karamouzas, Victor Zordan.- Gesture-based Human-robot Interaction for Field Programmable Autonomous Underwater Robots, Pei Xu, 2017.
- A Real-Time Hand Gesture Recognition and Human-Computer Interaction System, Pei Xu, 2017.