Junyu Xie

Papers

Total Citations

H-Index

About

Junyu Xie is a researcher at the intersection of robotics, computer vision, and human-robot interaction. Their work focuses on enhancing the perceptual and communicative abilities of autonomous systems, with key contributions in humanoid robotics and 3D scene understanding. In their 2023 study on a humanoid reading system with voice lip synchronization, Xie developed a framework that enables service robots to perform educational tasks with greater realism and engagement, addressing the growing demand for socially interactive robots. This work has garnered early attention with 3 citations, reflecting its relevance in the robotics community. More recently, in 2025, Xie introduced *FacaDiffy*, a novel diffusion-based method for inpainting unseen facade parts in high-detail semantic 3D building models. This work tackles the critical challenge of incomplete 2D conflict maps—a common issue in geoinformatics and robotic mapping—by leveraging generative AI to reconstruct missing architectural details. With 2 citations and growing interest, *FacaDiffy* demonstrates Xie’s ability to apply state-of-the-art generative techniques to real-world spatial perception problems. Junyu Xie’s research is shaping how robots perceive and interact with both human users and complex built environments.