About

I work on generative models for human motion, camera behavior, and dynamic visual content.

I am a first-year Ph.D. student in Control Science and Engineering at Tsinghua University, supervised by Prof. Xiu Li. Before Tsinghua, I received my B.Eng. from Chongqing University, where I ranked first in my class (1/323).

My research focuses on controllable models that connect language, music, human motion, 3D states, camera trajectories, and video.

Previously, I was a Research Intern at Tencent Games, Shenzhen, working on joint human-camera reconstruction.

I am open to research discussions, collaborations, and research internship opportunities.

News

• 2026CameraOperator was accepted at ACM MM 2026.
• 2026InfiniteDance was accepted at ECCV 2026.
• 2026WATCH was accepted.
• 2025AToM was published at CVPR 2025.

Research

Selected Publications

* denotes co-first authors. My name is underlined.

InfiniteDance: Scalable 3D Dance Generation Towards in-the-wild Generalization

Ronghui Li*, Zhongyuan Hu*, Li Siyao, Youliang Zhang, Haozhe Xie, Mingyuan Zhang, Jie Guo, Xiu Li, Ziwei Liu

ECCV 2026, accepted

Co-first author and core contributor; 100+ hours of multimodal 3D dance data.

paperproject page

Camera Operator: Object-Grounded Camera Trajectory Generation from Text and 3D Bounding Box Sequences

Zhongyuan Hu, Yue Ma, Jiangming Wang, Ronghui Li, Xiu Li

ACM MM 2026, accepted

First author and core contributor; BlockCam benchmark with 41K sequences.

DOI

WATCH: World-aware Allied Trajectory and pose reconstruction for Camera and Human

Qijun Ying*, Zhongyuan Hu*, Rui Zhang, Ronghui Li, Yu Lu, Zijiao Zeng

Accepted

Co-first author and core contributor; joint human-camera reconstruction in the wild.

paper

Controllable Video Generation: A Survey

Yue Ma*, Kunyu Feng*, Zhongyuan Hu*, Xinyu Wang*, Yucheng Wang, Mingzhe Zheng, Bingyuan Wang, Qinghe Wang, Xuanhua He, Hongfa Wang, Chenyang Zhu, Hongyu Liu, Yingqing He, Zeyu Wang, Zhifeng Li, Xiu Li, Sirui Han, Yike Guo, Wei Liu, Dan Xu, Linfeng Zhang, Qifeng Chen

ACM Computing Surveys, under review

Co-first author and core contributor; a taxonomy of controllable video generation.

paper

AToM: Aligning Text-to-Motion Model at Event-Level with GPT-4Vision Reward

Haonan Han*, Xiangzuo Wu*, Huan Liao*, Zunnan Xu, Zhongyuan Hu, Ronghui Li, Yachao Zhang, Xiu Li

CVPR 2025

Event-level alignment for text-to-motion generation with GPT-4Vision reward.

paperproject page

Education

Experience