Keming Shen / 沈可铭

Keming Shen 沈可铭

Everything happens for a reason, and there is a story to everything.

— Stephen Curry, in 14-15 NBA MVP Acceptance Speech

I am a final-year M.Sc. student at the Computer Vision Institute (CVI), Shenzhen University (SZU), advised by Prof. Linlin Shen. I received my B.Eng. in Software Engineering (Tencent Cloud Artificial Intelligence program) from College of Computer Science and Software Engineering (CSSE), SZU in 2024.

My research interests include Computer Vision, 3D Human Motion Generation, Embodied AI, and Reinforcement Learning. I have submitted papers to CVPR, NeurIPS, ICLR, ICCV, and AAAI, with two papers accepted for publication at AAAI 2026 and CVPR 2025. I also served as a reviewer for ICCV 2025 and AAAI 2026. I am seeking Ph.D. opportunities starting in Fall 2027 and research internships in human modeling, embodied AI, or AIGC, and research collaborations are welcome. For inquiries, please contact me by or via .

I excel at drawing, playing the electronic piano, and table tennis. I'm also keen on swimming, photography, cooking, and reading. I received the Bronze Prize in the 2016 "Tianyan" International Drawing Competition and hold a Level 10 certificate in electronic piano. I identify most closely with the ENFJ (Protagonist Personality) type. I have a knack for staying hopeful even in tough times, always reminding myself of what I have to be grateful for. I strive to be genuine, caring, and someone who walks the walk. Truly, nothing fulfills me more than inspiring others to do what’s right.

I share technical notes and reflections on everyday life on Keming's Eutopia / Blog.

Keming Shen avatar

Quick Facts

Supervisor Prof. Linlin Shen
Email
WeChat

News

  1. Patent

    Our smart farming patent (Patent No. 202311785633.3) has been granted!

  2. Honor

    I am honored to receive the National Scholarship for Graduate Students!

  3. Career

    I have joined Samsung R&D Institute China – Beijing (SRC-Beijing) as a research intern.

  4. Research

    FineXtrol (first author) has been accepted to AAAI 2026! Many thanks to my three co-authors: Dr. Bizhu Wu, Dr. Xiaoqin Wang, and Dr. Junliang Chen!

  5. Honor

    I received the First Prize of the Graduate Academic Scholarship at Shenzhen University.

  6. Research

    Our work MG-MotionLLM has been accepted to CVPR 2025! Congratulations to my senior labmate, Dr. Bizhu Wu!

  7. Honor

    I am honored to receive the Special Prize of the Graduate Academic Scholarship at Shenzhen University.

  8. Education

    I have officially begun my M.Sc. studies at Shenzhen University!

  9. Honor

    I was named an Outstanding Graduate of Shenzhen University.

  10. Honor

    My undergraduate thesis was selected among Shenzhen University's Top 100 Outstanding Undergraduate Theses!

  11. Education

    I graduated from Shenzhen University with a B.Eng. in Software Engineering (Tencent Cloud Artificial Intelligence Elite Class)!

Education

Shenzhen University · M.Sc. GPA 90.9/100 (3.91/4.0) · Rank #3
2024.09 — Present
Research Focus · 3D Human Motion Generation · Computer Vision · Diffusion Models · Multimodal Generation
Shenzhen University · B.Eng. GPA 87.7/100 (3.96/4.5) · Rank #7
2020.09 — 2024.06
Computer Science and Technology · Software Engineering (Tencent Cloud Artificial Intelligence Elite Class)

Publications

Think with Motion paper overview
Under Review Co-First Author
Think with Motion: A Unified Understanding and Generation Framework for Individual and Interactive Human Motion
Keming Shen (co-first author), Xiaofeng Tan, Bizhu Wu, Wanjiang Weng, Zehai Niu, Hongsong Wang, Linlin Shen
I²MoT unifies individual and interactive motion understanding and generation through relation-decoupled representations, task-specific experts, and participant-grounded chain-of-thought reasoning.
FineXtrol paper overview
AAAI 2026 First Author
FineXtrol: Controllable Motion Generation via Fine-Grained Text
Keming Shen, Bizhu Wu, Junliang Chen, Xiaoqin Wang, Linlin Shen
FineXtrol enables fine-grained text control of motion generation, using hierarchical contrastive learning to improve text-motion alignment.
MG-MotionLLM paper overview
CVPR 2025 Co-Author
MG-MotionLLM: A Unified Framework for Motion Comprehension and Generation across Multiple Granularities
Bizhu Wu, Jinheng Xie, Keming Shen, Zhe Kong, Jianfeng Ren, Ruibin Bai, Rong Qu, Linlin Shen
MG-MotionLLM unifies motion comprehension and generation across multiple granularities, supporting fine-grained editing and cross-task generalization.
Ciallo paper overview
Pre-print
Ciallo: Coordinated and Controllable 3D Talking Avatar with Holistic & Fine-Grained Face-Body Space
Yuhang Guo, Keming Shen, Kaijun Deng, Siyang Song, Zehai Niu, Jinbao Wang, Linlin Shen
Ciallo generates gestures and facial expressions in a shared representation space, enabling coordinated talking avatar synthesis with fine-grained control over individual body parts.

Honors

  • 2026
    National Scholarship for Graduate Students

    Shenzhen University

  • 2026
    Graduate Academic Scholarship — First Prize

    Shenzhen University

  • 2025
    Graduate Academic Scholarship — First Prize

    Shenzhen University · RMB 10,000

  • 2024
    Graduate Academic Scholarship — Special Prize

    Shenzhen University · RMB 15,000

  • 2024
    Outstanding Graduate (Undergraduate)

    University-level award, Shenzhen University

  • 2024
    Outstanding Undergraduate Thesis

    Top 100 undergraduate theses and projects, Shenzhen University

  • 2022
    Star of Innovation and Entrepreneurship

    Shenzhen University

  • 2021–2023
    Star of Learning

    Shenzhen University

Competitions

  • 2025
    Gold Award (Top 10)

    Shenzhen University “Liyuan Challenge” Innovation and Entrepreneurship Competition · Entrepreneurship Track

  • 2025
    Winner Award

    Workshop & Challenge on Deepfake Detection, Localization, and Interpretability, IJCAI 2025 · Generative Large Model Security Track

  • 2023
    First Prize, Guangdong Province

    China Undergraduate Mathematical Contest in Modeling

Service

  • TA
    Teaching Assistant

    University Computing, C Language Programming, and related courses at Shenzhen University.

  • 2026
    Reviewer for AAAI 2026

    Human motion modeling and computer vision.

  • 2025
    Reviewer for ICCV 2025

    Computer vision and human motion research.