Qunce XU

Qunce XU is an assistant researcher in the Department of Computer Science and Technology at Tsinghua University. His research interests include computer graphics, geometric computing, and artificial intelligence, with a focus on 3D geometric reconstruction. He has contributed to areas such as point cloud denoising, surface mesh optimization, and 3D content editing. His work includes methods for 3D reconstruction using sparse neural radiance fields and deep learning for fragment assembly. XU has published in top conferences and journals, including Computer Graphics Forum and AAAI. He joined the department in 2025 and can be contacted via email at quncexu@tsinghua.edu.cn.

Qun-Ce Xu is a researcher at Tsinghua University with a focus on 3D shape generation, surface reconstruction, and procedural modeling. Their work explores deep learning techniques for generating and reconstructing 3D shapes, including diffusion models for fractured object assembly and semantic mesh generation. Xu also investigates anisotropic surface remeshing, point cloud denoising, and efficient splitting algorithms for 3D Gaussian splatting. Their research emphasizes high-accuracy reassembly of fractured objects and the use of memory cache to improve workflow system performance. Additionally, Xu contributes to generative modeling using natural languages and reflection-supervised surface reconstruction for specular scenes.