Wenjin Deng

SSP = Strong + Smart + Professional

google scholar
  • CSIG Member

About Me

I'm currently R&D Engineer in Momenta (opens new window), working on MSD (Momenta Self-Driving).

Before that I got my Master's degree and Bachelor's degree in the School of Informatics at Xiamen University. Especially, I spent wonderful time from 2019 to 2023 in the Visual Computing and Graphics Lab (opens new window), supervised by associate professor Ming Zeng (opens new window). In 2022.6 -2022.9, I served as a virutal human research intern at Virtual Human Group (opens new window) of NetEase Fuxi AI Lab (opens new window), working with Zhimeng Zhang (opens new window) and Yu Ding (opens new window).

My research interests lie at Computer Vision and Computer Graphics. I'm particularly interested in the areas of human-centric perception (pose estimation, restruction), generation and animation (speech audio or natural language driven).


  • [Oct. 2023] Our work on 3D Human Reconstruction has been accepted by Graphical Models.
  • [Sep. 2023] Our two National Invention Patents from China have been officially granted.
  • [Nov. 2022] Our work on singing head generation has been accepted by CVM2023 and CVMJ.
  • [Nov. 2022] Our work on talking head generation has been accepted by AAAI2023.
  • [Oct. 2022] I and my roommate Yinglin Zheng (opens new window) won the National Scholarship for Graduate Students at the same time.
  • [Aug. 2022] Our work on talking head generation has been submitted to AAAI2023.
  • [Apr. 2022] Our work on multi-person pose estimation has been accepted by IJCAI2022.
  • [Mar. 2021] Our work on maksed face revealing is accepted by ICME 2021 as oral presentation.
  • [Feb. 2021] My zhihu survey on human pose (opens new window) is re-tweeted by several well-known AI WeChat accounts.
  • [Oct. 2020] Our work on human pose estimation is accepted by PRCV 2020.

Some of my paper co-authors

Ming Zeng (opens new window), Associate Professor at School of Informatics, Xiamen University
Yinglin Zheng (opens new window), Research Intern at Visual Computing Group, Microsoft Research Asia (MSRA)
Jinpeng Lin (opens new window), Researcher at AlibabaGroup


Vertex position estimation with spatial–temporal transformer for 3D human reconstruction

Xiangjun Zhang, Yinglin Zheng, Wenjin Deng, Qifeng Dai, Yuxin Lin, Wangzheng Shi, Ming Zeng

2023, Graphical Models

Paper (opens new window)

MusicFace: Music-driven Expressive Singing Face Synthesis

Pengfei Liu, Wenjin Deng, Hengda Li, Jintai Wang, Yinglin Zheng, Yiwei Ding, Xiaohu Guo, Ming Zeng

2022, The Computational Visual Media Conference (CVM2023)

Paper (opens new window) Data (opens new window)

DINet: Deformation Inpainting Network for Realistic Face Visually Dubbing on High Resolution Video

Zhimeng Zhang*, Zhipeng Hu*, Wenjin Deng*, Changjie Fan, Tangjie Lv, Yu Ding*

2022, Thirty-Seventh AAAI Conference on Artificial Intelligence. (AAAI2023)

paper (opens new window) Code (opens new window)

I^2R-Net: Intra- and Inter-Human Relation Network for Multi-Person Pose Estimation

Yiwei Ding*, Wenjin Deng*, Yinglin Zheng, Pengfei Liu, Jianmin Bao, Meihong Wang, Xuan Cheng, Ming Zeng, Dong Chen

2022, The 31st International Joint Conference on Artificial Intelligence (IJCAI2022)

Paper (opens new window) Code (opens new window)

Controllable Facial Caricaturization with Localized Deformation and Personalized Semantic Attentions

Ming Zeng, Yinglin Zheng, Jinpeng Lin, Xuan Cheng, Jing Liao, Zizhao Wu, Wenjin Deng

2021, IEEE Transactions on Multimedia (TMM2021).

Link (opens new window)

Real-time Masked Face Revealing for Video Conference

Jinpeng Lin, Pengfei Liu, Yinglin Zheng, Wenjin Deng, Ming Zeng

2021, IEEE International Conference on Multimedia and Expo (ICME2021), Oral presentation.

Link (opens new window)

VH3D-LSFM:Video-based Human 3D Pose Estimation with Long-term and Short-term Pose Fusion Mechanism

Wenjin Deng, Yinglin Zheng, Hui Li, Xianwei Wang, Zizhao Wu, Ming Zeng

2020, Chinese Conference on Pattern Recognition and Computer Vision (PRCV2020)

Link (opens new window)

Patents and Copyrights

[中国国家发明专利] CN202111117800.8 一种感知环境的语音驱动虚拟人姿态合成方法 (已授权)


[中国国家发明专利] CN202011092625.7 一种利用长短期信息融合的视频三维人体姿态估计算法 (已授权)


  • [中国国家发明专利] CN202110750794.3 一种利用语音信息的实时视频人脸区域时空一致合成方法 (公审中)


  • [中国国家发明专利] CN202011408727.5 一种增强医患相互信任减少医患矛盾评估预警系统 (公审中)


  • [软件著作权] 2019SR0447844 智能交互健身系统 (已授权)



A management system for case workflow

Java Backend: Wenjin Deng, Pengfei Liu (opens new window)

H5 frontend: Yiwei Ding (opens new window), Wenjin Deng


Watermeter Reader Autonomous System for paper Image-Based Automatic Watermeter Reading under Challenging Environments

Java Backend: Wenjin Deng, Jian Wang

H5 frontend: Yiwei Ding (opens new window)


Paper Link (opens new window)

AI fitness coach in web(PC/Mobile) using our Human Pose Estimation Network. Benefit from our design, it infers well pose on user's local device with 25fps.

Human Pose Estimation Algorithm & Website: Zihao Chen (opens new window), Wenjin Deng

Team: Ximeng Zhou, Zihao Chen (opens new window), Wenjin Deng, Yilin Huang


Media Report (opens new window)

A course system in web(PC/Mobile) using Springboot and VUE.

Java Backend: Shiqi Wang (opens new window), Wenjin Deng

H5 frontend: Tianyu Su, Zihao Chen (opens new window)


Project Link (opens new window)

Awards & Honors

  • National Scholarship for Graduate Students, Xiamen University, China, 2022
  • Cho Tak Wong Scholarship, Xiamen University, China, 2021
  • Outstanding prize (1st place) of the 12nd "Intel Cup" national undergraduate software innovation competition, Shanghai, China, 2019.

Education & Experiences

  • Momenta Self-Driving, Momenta, R&D Engineer.
    Jul. 2023 - Present

  • Virtual Human Group, NetEase Fuxi AI Lab, Research Intern.
    Jun. 2022 - Sept. 2022

  • School of Informatics, Xiamen University, M.S.
    Sept. 2020 - Jun. 2023

  • School of Informatics, Xiamen University, B.S.
    Sept. 2016 - Jun. 2020


Great thanks to my friend Yinglin Zheng (opens new window).

stats counter
Last Updated: 1/16/2024, 6:57:55 AM