Dr. Shengyong Ding

Associate professor, doctoral supervisor/master supervisor 

Email：syding@cityu.edu.mo

Tel: (853)85902428

Office address:  Room S401, Stanley Ho Building, City University of Macau (Taipa)

Academic Qualifications

2015: PhD in Computer Applications, Sun Yat-sen University, China

2008: Master of Applied Mathematics, Sun Yat-sen University, China

2001: Bachelor of Information Management, University of Science and Technology of China, China

Previously Taught Subjects

Introduction to Operating System

Visualization and Computer Graphics

Research Interests

Fundamental Research: Integrating large model technology with traditional computer vision (CV) techniques to solve key challenges in machine vision, such as pose estimation, 3D reconstruction and generation, and spatial navigation, with a focus on model geometric quality.

• 3D Avatar: Identity (ID)-preserving 3D avatar reconstruction and generation. Research on how to obtain high-quality 3D realistic or stylized models from a small number of RGB images by integrating traditional CV techniques, large model technology, and reinforcement learning to solve key challenges.
• Video3D: Integrating video generation with 3D vision. Research on how to introduce 3D constraints into video generation and organically merge 2D video generation with 3D model reconstruction/generation processes, with a focus on maintaining 3D consistency of human figures.
• MAgent: Multi-agent systems for multimodal generation aim to explore how to use agent technologies to automatically orchestrate language, speech, and video models for creating multimedia content.

Applied Research: Applying advanced AI models/tools to solve specific application problems in fields such as education and healthcare.

• Multiagent Story Generation: Apply large generative models such as language models, voice models and video models to generate video stories based on users’ simple inputs.
• Educational and Medical Data Analysis: Utilize facial, body posture, and multimodal perception technologies to analyze students' attention in real-time, providing key core capabilities for smart education; explore the applications of artificial intelligence in fields such as traditional Chinese medicine heritage and vestibular therapy.
• Smart Transportation: Explore the applications of traditional vision technologies such as stereo vision, LiDAR, and multimodal large models in smart transportation inspection.

Research & Publications

Refereed Journal Articles

Liu, Y., Liu, X., Ding, S., & Yang, Z. (2025). Robustly solving PnL problem using Clifford tori. Pattern Recognit., 166, 111659.

Xiao, Zelin & Lin, Hongxin & Li, Renjie & Geng, Lishuai & Chao, Hongyang & Ding, Shengyong. (2020). Endowing Deep 3d Models With Rotation Invariance Based On Principal Component Analysis. 1-6. 10.1109/ICME46284.2020.9102947.

Lin, Hongxin & Xiao, Zelin & Tan, Yang & Chao, Hongyang & Ding, Shengyong. (2019). Justlookup: One Millisecond Deep Feature Extraction for Point Clouds By Lookup Tables. 326-331. 10.1109/ICME.2019.00064.

Tan, Yang & Lin, Hongxin & Xiao, Zelin & Ding, Shengyong & Chao, Hongyang. (2019). Face Recognition from Sequential Sparse 3D Data via Deep Registration. 1-8. 10.1109/ICB45273.2019.8987284.

Wang, Guangrun & Lin, Liang & Ding, Shengyong & Li, Ya & Wang, Qing. (2016). DARI: Distance Metric and Representation Integration for Person Verification. Proceedings of the AAAI Conference on Artificial Intelligence. 30. 10.1609/aaai.v30i1.10462.

Ding, Shengyong & Lin, Liang & Wang, Guangrun & Chao, Hongyang. J(2015). Deep Feature Learning with Relative Distance Comparison for Person Re-identification. Pattern Recognition. 48. 10.1016/j.patcog.2015.04.005.

Academic Awards

2018 Pattern Recognition Best Paper