Xuejun Hu

Email Google Scholar GitHub

I am a fourth-year Ph.D. student at Tsinghua University, advised by Prof. Kun Xu. My research lies at the intersection of real-time rendering and artificial intelligence, with a focus on applying neural networks to solve rendering problems. Apart from rendering, I am also interested in topics about Monte Carlo PDE solvers. I received my bachelor's degree from IIIS, Tsinghua University in 2023.

Publications

Pixel-Light Attention for Neural Many Light Importance Sampling

Xuejun Hu, Jinfan Lu, Kun Xu

ACM Transactions on Graphics (SIGGRAPH Asia 2026)

[PDF] [Code] [Project]

Gradient Domain Reconstruction for Monte Carlo PDE Solvers Best Paper Honorable Mention

Jiaqi Wu, Xuejun Hu, Shuang Zhao, Kun Xu

ACM Transactions on Graphics (SIGGRAPH 2026)

[PDF] [Code] [Project]

Kernel Predicting Neural Shadow Maps

Xuejun Hu, Jinfan Lu, Kun Xu

SIGGRAPH 2025 Conference Paper

[PDF] [Code] [Project]

Extended Path Space Manifolds for Physically Based Differentiable Rendering

Jiankai Xing, Xuejun Hu, Fujun Luan, Ling-Qi Yan, Kun Xu

SIGGRAPH Asia 2023 Conference Paper

[PDF] [Code] [Project]

Differentiable Rendering using RGBXY Derivatives and Optimal Transport

Jiankai Xing, Fujun Luan, Ling-Qi Yan, Xuejun Hu, Houde Qian, Kun Xu

ACM Transactions on Graphics (SIGGRAPH Asia 2022)

[PDF] [Code] [Project]