Yilin Zhu

About Me

I am a Machine Learning Engineer at Apple, where I work on multimodal AI for human perception and the creation of realistic digital humans.

I received my Ph.D. degree from the Computer Science Department, Stanford University, advised by Ron Fedkiw at the Stanford AI Lab and Graphics Lab. Previously, I worked at Epic Games on facial performance capture and avatar creation.

My current research centers on multimodal AI for digital humans: building models that perceive, simulate, and animate people with realism and physical grounding. Bringing together generative models, computer vision, and computational physics, I extend the same principles to embodied AI (robotics), building agents that reason about and operate in the physical world. Broadly, I aim to democratize the creation of agents that understand and act in the world, turning expensive, expert-only pipelines into tools anyone can use.

Education

Stanford University, CA, USA

Ph.D. • Computer Science • Sep. 2018 - Jun. 2024

Tsinghua University, Beijing, China

Bachelor of Engineering • Automation • Aug. 2014 - July 2018

Bachelor of Arts • Digital Media Art • Aug. 2015 - July 2018

University of Texas at Austin, TX, USA

Exchange Student • Electrical and Computer Engineering • Aug. 2016 - Dec. 2016

Work Experience

Apple, CA, USA

Machine Learning Engineer • Feb. 2024 - Current

Epic Games, CA, USA

Physics Programmer Intern • Apr. 2020 - Dec. 2023

Research

Leveraging Deepfakes to Close the Domain Gap between Real and Synthetic Images in Facial Capture Pipelines Winnie Lin, Yilin Zhu, Demi Guo, Ron Fedkiw arXiv / Paper

Learning Topological Motion Primitives for Knot Planning Mengyuan Yan, Geng Li, Yilin Zhu, Jeannette Bohg IROS 2020 / Paper / Video