About Me

Hi there! I’m Shian Du. I am a 3rd-year Master Student at Tsinghua University, advised by Prof. Xiangyang Ji. Currently, I’m a research intern at the Kling Team, Kuaishou Technology, focusing on Generative Video Super-resolution and Multi-modal Video Generation. Before that, I was fortunate to work with Prof. Delu Zeng on Deep Generative Models at South China University of Technology. Feel free to reach out to me at shiandu@qq.com.

I will be graduating in the summer of 2026 and am currently seeking job opportunities. Please feel free to contact me via email.

Recent News

• 2025/09: 1 paper accepted by PR'25.
• 2025/06: 1 paper accepted by CVPR'25.
• 2025/04: 1 paper accepted by AISTATS'25.
• 2024/08: 1 paper accepted by UAI'25.
• 2024/03: Start a research internship at Kling Team, advised by Prof. Menghan Xia and Prof. Xintao Wang.
• 2024/03: 1 paper accepted by ECCV'24.
• 2023/09: Join Prof. Xiangyang Ji's lab at Tsinghua University.
• 2023/08: 1 paper accepted by TNNLS'24.
• 2022/03: 1 paper accepted by CVPR'22.
• 2021/03: Join Prof. Delu Zeng's lab at South China University of Technology.

Selected Publications

	Shian Du, Menghan Xia, Chang Liu et al. "UniMMVSR: A Unified Multi-Modal Framework for Cascaded Video Super-Resolution". In Arxiv 2025. PDF Project page
	Wei Chen, Shian Du, Shigui Li et al. "Entropy-Informed Weighting Channel Normalizing Flow for Deep Generative Models". In Pattern Recognition 2025. PDF GitHub
	Liangbin Xie, Yu Li, Shian Du et al. "SimpleGVR: A Simple Baseline for Latent-Cascaded Video Super-Resolution". In Arxiv 2025. PDF Project page
	Shian Du, Menghan Xia, Chang Liu et al. "PatchVSR: Breaking Video Diffusion Resolution Limits with Patch-wise Video Super-Resolution". In CVPR 2025. PDF Project page
	Qiran Zou, Shangyuan Yuan, Shian Du et al. "ParCo: Part-Coordinating Text-to-Motion Synthesis". In ECCV 2024. PDF GitHub
	Jian Xu, Shian Du, Junmei Yang et al. "Neural Operator Variational Inference based on Regularized Stein Discrepancy for Deep Gaussian Processes". In TNNLS 2024. PDF GitHub
	Shian Du, Yihong Luo, Wei Chen et al. "TO-FLOW: Efficient Continuous Normalizing Flows with Temporal Optimization adjoint with Moving Speed". CVPR 2022. PDF GitHub

Experience

	Research Intern, Kling, Kuaishou Technology May. 2024 - Present Advisor: Menghan Xia, Xintao Wang Interests: Generative Video Super-resolution and Multi-modal Video Generation
	Master Student, Tsinghua University Sep. 2023 - Present Advisor: Xiangyang Ji Interests: Generative Video Super-resolution and Multi-modal Video Generation
	Undergraduate, South China University of Technology Sep. 2019 - Jun. 2023 Advisor: Delu Zeng Interests: Deep Generative Models