Hi, this is Yinghan Xu (徐英瀚). I am currently a Ph.D. student at the Graphics and Vision Research Group, Trinity College Dublin, supervised by John Dingliana.

My research interests lie in human-centric 3D vision, including volumetric video dataset, human reconstruction, motion capture, and avatar generation. Currently, I am focusing on realistic garment reconstruction and rendering using the 3D Gaussian splatting technique.

I earned my Master’s degree in Interactive Media Technology from Trinity College Dublin in 2017, where I was supervised by John Dingliana. Before starting my Ph.D. program, I worked as a Computer Vision Engineer at Mobvoi and STEPVR in Beijing, China. My work focused on developing monocular and multiple-view camera motion capture systems.

Research

LayerGS: Decomposition and Inpainting of Layered 3D Human Avatars via 2D Gaussian Splatting Yinghan Xu, John Dingliana Under Review Projectpage / Paper / Video

Evaluating CrowdSplat: Perceived Level of Detail for Gaussian Crowds Xiaohan Sun*, Yinghan Xu*, John Dingliana, Carol O’Sullivan (*=equal contribution) IEEE Virtual Reality (workshop paper), 2025 Projectpage / Paper