Alexandra(Yongkang) Cheng

🤓About Me

Hello, everyone. I’m Yongkang Cheng. I’m an incoming Ph.D. student at MBZUAI, and my supervisor is Prof. Mingming Gong. I obtained my master’s and bachelor’s degrees from NWAFU and NJAU respectively. Currently, I’m a researcher at an Embodied AI startup. Before this, I worked as a research intern at Tencent AILab, CV-Lab, and the Digital Human Research Center, supervised by Dr. Shaoli Huang. My research interests lie in 3D motion generation, Motion Capture, and Embodie AI.

“Fortunately, my passion lies in the essence of artificial intelligence itself, rather than the mere trappings it brings ! “ — Yongkang

🎉 Research Interests

• Computer Vision: 3D human pose estimation, motion generation, objects tracking
• Machine Learning: multimodel learning
• Embodied Artificial Intelligence

👣 Education Experience

• 2017.09 - 2021.06: Nanjing Agriculture University (Bachelor 211).
• 2021.09 - 2024.12: Northwest A&F University (Master 985).
• 2025.08 - Future: MBZUAI (PHD student, CS Ranking TOP 10 in the world)

👣 Research Experience

• 2022.08 - 2024.12: Research Intern, Tencent AI Lab.
• 2025.01 - 2025.07: Researcher, Embodied AI.

🔥 News

• [2025 to present]
• 3 papers are under review.
• [2023 to 2024]
• [Nov. 2024] 1 paper is accepted to 3DV 2025.
• [Dec. 2024] 1 paper is accepted to AAAI 2025.
• [Aug. 2024] 3 papers are accepted to WACV 2025.
• [Jul. 2024] 1 paper is accepted to ECCV 2024.
• [Mar. 2024] 1 paper is accepted to ICME 2024.
• [Jan. 2024] 1 paper is accepted to ICASSP 2024.
• [Before 2023]
• [Dec. 2023] 1 paper is accepted to TCSVT.
• [Feb. 2021] 1 paper is accepted to TCSAM.

📝 Selected Publications (* indicates co-authors. † indicates the corresponding author/project leader.)

HoleGest: Decoupled Diffusion and Motion Priors for Generating Holisticly Expressive Co-speech Gestures (3DV 2025 & China3DV2024 video pre)

Yongkang Cheng, Shaoli Huang†

DIDiffGes: Decoupled Semi-Implicit Diffusion Models for Real-time Gesture Generation from Speech (AAAI 2025)

Yongkang Cheng, Shaoli Huang†, Xuelin Chen, Jifeng Ning, Mingming Gong

BoPR: Body-aware Part Regressor for Human Shape and Pose Estimation (Under Review)

arxiv project code

Yongkang Cheng, Shaoli Huang, Jifeng Ning, Ying Shan

Conditional GAN for Enhancing Diffusion Models in Efficient and Authentic Global Gesture Generation from Audio (WACV 2025)

Yongkang Cheng, Mingjiang Liang, Shaoli Huang†, Gaoge Han, Jifeng Ning, Wei Liu.

RopeTP: Global Human Motion Recovery via Integrating Robust Pose Estimation with Diffusion Trajectory Prior (WACV 2025)

code

Yongkang Cheng *, Mingjiang Liang *, Hualin Liang, Shaoli Huang†, Wei Liu

SignAvatars: A Large-scale 3D Sign Language Holistic Motion Dataset and Benchmark (ECCV 2024)

arxiv paper code project

Zhengdi Yu, Shaoli Huang†, Yongkang Cheng, Tolga Birdal1

ExpGest: Expressive Full-Body Gesture Generation Using Diffusion Model and Hybrid Audio-Text Guidance (ICME 2024)

arxiv paper code

Yongkang Cheng, Mingjiang Liang, Shaoli Huang†, Gaoge Han, Wei Liu, Jifeng Ning†

🎖 Honors and Awards

• [May. 2021] Nanjing Agraiculture University Outstanding Graduate
• [Dec. 2020] National Undergraduate Innovation Project (10000 RMB)
• [Aug. 2019] 2019 National University Artificial Intelligence Competition Champion (12000 RMB)
• [Apr. 2019] Second-class Scholarship of Nanjing Agricultural University
• [Feb. 2019] Second Prize in the Asia Region of ASC 2018

👀 Services

Conference Reviewer ICME 2024,2025; ICPR 2024; ACM MM 2024; WACV 2025; ICCV 2025;

Journal Reviewer IJCV; TVCG; PR;