Hi, I’m Zhiyuan Ma (马致远), a third year Ph.D. student in the Department of Computing (COMP) at Hong Kong Polytechnic University (PolyU). I am jointly advised by Prof. Lei Zhang at PolyU and Prof. Zhen Lei, Xiangyu Zhu at Institute of Automation, Chinese Academy of Sciences (CASIA).

My research interest is 3D generation.

🔥 News

• 2025.02:  🎊 3 papers accepted by CVPR 2025, including 1 first-author paper.
• 2024.09:  🥳 1 paper accepted by NeurIPS 2024.
• 2024.07:  🎊 2 papers accepted by ECCV 2024, including 1 first-author paper.
• 2024.02:  🥳 1 paper accepted by CVPR 2024.
• 2023.02:  🎊 1 first-author paper accepted by CVPR 2023.

📝 Publications

CVPR 2025

Progressive Rendering Distillation: Adapting Stable Diffusion for Instant Text-to-Mesh Generation without 3D Data

Zhiyuan Ma, Xinyue Liang, Rongyuan Wu, Xiangyu Zhu, Zhen Lei, Lei Zhang

• First work to adapt Stable Diffusion as a text-to-mesh generator without 3D data as ground truths.
• Achieved sub-second text-to-mesh generation in 4 steps.
• Introduced only 2.6% more parameters to the original Stable Diffusion model.
• | HuggingFace Demo

ECCV 2024

ScaleDreamer: Scalable Text-to-3D Synthesis with Asynchronous Score Distillation

Zhiyuan Ma, Yuxiang Wei, Yabin Zhang, Xiangyu Zhu, Zhen Lei, Lei Zhang

• First work to propose a score distillation method to train text-to-3D generator across multiple text prompts.
• Validated on 100K+ prompts using hyper-network, CNN and Transformer architectures.
• | Project Page

CVPR 2023

OTAvatar : One-shot Talking Face Avatar with Controllable Tri-plane Rendering

Zhiyuan Ma*, Xiangyu Zhu*, Guojun Qi, Zhen Lei, Lei Zhang

• First work to adapt pre-trained 3D Face GANs for one-shot talking face generation.
• Explicit 3D rendering can handle extreme head movements.
• | Demo Video | Report Recording

• Arxiv 2024 DiffSpeaker: Speech-Driven 3D Facial Animation with Diffusion Transformer | Zhiyuan Ma, Xiangyu Zhu, Guojun Qi, Chen Qian, Zhaoxiang Zhang, Zhen Lei |

• CVPR 2025 MVBoost: Boost 3D Reconstruction with Multi-View Refinement | Xiangyu Liu*, Xiaomei Zhang*, Zhiyuan Ma, Xiangyu Zhu, Zhen Lei |

• CVPR 2025 Pixel-level and Semantic-level Adjustable Super-resolution: A Dual-LoRA Approach | Lingchen Sun, Rongyuan Wu, Zhiyuan Ma, Shuaizheng Liu, Qiaosi Yi, Lei Zhang |

• NeurIPS 2024 One-Step Effective Diffusion Network for Real-World Image Super-Resolution | Rongyuan Wu*, Lingchen Sun*, Zhiyuan Ma*, Lei Zhang |

• CVPR 2024 Dual memory networks: A versatile adaptation approach for vision-language models | Yabin Zhang, Wenjie Zhu, Hui Tang, Zhiyuan Ma, Kaiyang Zhou, Lei Zhang |

• ECCV 2024 Dense Multimodal Alignment for Open-Vocabulary 3D Scene Understanding | Ruihuang Li, Zhengqiang Zhang, Chenhang He, Zhiyuan Ma, Vishal M Patel, Lei Zhang |

📖 Educations

• 2022.02 - now, Ph.D. in Computer Science, Hong Kong Polytechnic University, Hong Kong, China.
• 2020.09 - 2022.01, Master of Science in Electronic Engineering, Columbia University, New York, USA.
• 2016.09 - 2020.06, Bachelor of Science in Information Engineering, Xi’an Jiaotong University, Xi’an, China.

💻 Internships

• 2022.01 - 2022.05, CBSR, CASIA, Beijing, China.
• 2021.09 - 2022.01, AI Research Lab, JD, Beijing, China.