Max Robotics — Global Service Robot Directory + US Certification

About

Ze Yang is a researcher whose work sits at the intersection of 3D computer vision, neural rendering, and autonomous systems simulation. His research focuses on two interconnected themes: realistic human and object modeling, and world model learning for autonomous driving applications. Yang's most influential contribution is his work on S³: Neural Shape, Skeleton, and Skinning Fields (2021), which has accumulated over 68 citations. This paper introduced a principled neural framework for modeling the full complexity of human shape, pose, and clothing — a critical advancement for applications in virtual reality and robotics simulation. By representing shape, skeleton, and skinning as continuous neural fields, the work significantly advanced the state of the art in animatable human avatar generation. Extending his interest in simulation realism, Yang contributed to NeuSim (2023), which reconstructs real-world objects from sparse observations for use in sensor simulation pipelines. More recently, his work on Copilot4D explored unsupervised world model learning for autonomous driving through discrete diffusion, addressing a key bottleneck in scaling predictive models for robotic agents. Together, these contributions reflect a coherent research vision: building richer, more realistic virtual environments to accelerate the development of intelligent autonomous systems.