Xiaojiang Peng School of Artificial Intelligence, Shenzhen Technology University · Lead of OpenFEEL Team

Full Professor, School of Artificial Intelligence, Shenzhen Technology University · Lead of OpenFEEL

Xiaojiang Peng (IEEE Senior Member) is a full professor at the School of Artificial Intelligence, Shenzhen Technology University, and head of the Department of Intelligent Science and Technology (founder of the university's AI undergraduate program). He is a master's supervisor at Shenzhen University, a Shenzhen "Peacock" talent, and has been selected for the Stanford University top 2% world scientists list every year since 2022.

He received his Ph.D. from Southwest Jiaotong University. He was a postdoctoral researcher at Inria (LEAR/THOTH, France, with Cordelia Schmid) and Idiap Research Institute, Switzerland (with Francois Fleuret) from 2015 to 2017, and an associate professor at the Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, from 2018 to 2020.

Research

Centered on Open Frontiers in Emotion and Embodied Learning (OpenFEEL), we build a research pipeline for embodied agents that possess both emotional intelligence and artificial intelligence, spanning from multimodal interaction understanding and embodied action, emotion perception and understanding, to embodied emotional expression and proactive interaction.

Affective Computing

Facial expression recognition, multimodal emotion understanding, affective large models, and speech emotion.

Embodied Intelligence

Robotic arm manipulation, world models, vision-language navigation, humanoid control, and embodied emotional interaction.

Multimodal Large Models

Large model post-training, instruction tuning and alignment, and specialized agents.

Visual Intelligence Applications

Traffic visual intelligence, identity recognition, 3D visual measurement and processing, and industrial vision.

News

  1. 2026/02/22An overview of embodied intelligence on robotic manipulation has been accepted by Neurocomputing.
  2. 2026/01/29The multimodal emotion understanding papers "EmoPrefer" and "MME-Emotion" have been accepted by ICLR. Congratulations to everyone!
  3. 2025/10/29One paper on 3D point cloud keypoint detection for humans has been accepted by Expert Systems With Applications.
  4. 2025/10/29One paper on generative models for noise cross-domain image recognition has been accepted by Pattern Recognition.
  5. 2025/10/29One paper on pruning the Mamba model has been accepted by NeurIPS.
  6. 2025/08/29One paper on multimodal controlled clothing editing has been accepted by ICME.
  7. 2025/07/29Welcome to the 6th International Conference on Machine Learning and Computer Application (ICMLCA) — see you at SZTU, Oct. 17-19.
  8. 2025/07/29One paper on semi-supervised facial emotion recognition has been accepted by CVIU.
  9. 2025/07/16One paper on EEG-based emotion recognition has been accepted by IEEE TAFFC.
  10. 2025/07/16One paper on ViT training strategy has been accepted by ICCV 2025.
  11. 2025/02/28One paper on speeding up Diffusion model training has been accepted by CVPR 2025.
  12. 2024/12/26One paper on sEMG-based joint angle estimation has been accepted by IEEE Robotics and Automation Letters (RA-L).
  13. 2024/12/26Four papers on content emotion recognition, human avatar, emotional speech generation, and 6D pose have been accepted by ICASSP 2025.
  14. 2024/10/16The multimodal emotion understanding model Emotion-LLaMA has been accepted at NeurIPS 2024.
  15. 2024/10/16One paper on facial expression recognition has been accepted by IEEE Transactions on Multimedia.
  16. 2024/07/16Two papers on emotion recognition and stance detection have been accepted by ACM MM 2024.
  17. 2024/07/10Works on content generation authentication of various types of Stable Diffusion have been accepted by PRCV.
  18. 2024/03/10Causal spatio-temporal reasoning work has been accepted by ICLR.
  19. 2024/02/10The dialogue stance detection work has been accepted by COLING, and the team ranked third in the video dialogue emotion understanding competition at NAACL.
  20. 2023/10/25Selected for the Stanford University 2022 global top 2% scientists list.
  21. 2023/07/17One paper on multimodal emotion recognition accepted by ACM MM; runner-up in the MER 2023 Grand Challenge.
  22. 2023/04/17Two papers on vehicle smoke detection have been accepted by Applied Sciences.
  23. 2022/10/21Selected by Stanford University as one of the top 2% of scientists in 2021 (News).
  24. 2022/07/02One paper on rail detection has been accepted by ACM MM.
  25. 2022/06Special Issue "New Trends in Computer Vision, Deep Learning and Artificial Intelligence" opened in Mathematics.
  26. 2022/05/02One paper on cross-domain facial expression recognition has been accepted by Applied Sciences.
  27. 2022/03/08One paper on training large-scale face recognition data has been accepted by CVPR.
  28. 2022/01/17One paper on person re-identification has been accepted by Pattern Recognition.
  29. 2021/07/01One paper on context-aware emotion recognition has been accepted by IJCB 2021.
  30. 2021/03/22Two papers on human-object interaction have been accepted by CVPR 2021.
  31. 2021/01/25One paper on action anticipation has been accepted by Neurocomputing.
  32. 2020/07/03Three papers accepted by ECCV 2020 — noisy supervised learning, multi-label classification, and HOI recognition.
  33. 2020/04/02One paper on product image classification with noisy labels has been accepted by CVIU.
  34. 2020/02/04One paper on large-scale facial expression recognition has been accepted by CVPR 2020.

Awards

  1. 2021Digital Intelligent Construction Monitoring Platform — Key Technologies and Applications, CAA Science and Technology Progress Award, Second Prize, rank 3/15
  2. 2021Smart Construction Site IoT Multimodal Intelligent Monitoring and Management Platform, CIS Science and Technology Progress Award, Third Prize, rank 6/15
  3. 2019Deep Representation and Recognition Technology and Applications of Video, Guangdong Provincial Science and Technology Progress Award (Technological Invention), First Prize, rank 5/7
  4. 2019Deep Representation and Understanding Technology and Applications of Video Sequences, Wu Wenjun AI Science and Technology Progress Award, Second Prize (CAAI), rank 4/6
  5. 2019Deep Representation and Understanding Technology and Applications of Complex Video, Shenzhen Science and Technology Progress Award, Second Prize, rank 4/6
  6. 2019Shenzhen Peacock Talent (Category C), Nanshan Leading Talent

Professional Activity

Associate Editor

  • -IEEE Transactions on Affective Computing
  • -Empathic Computing

Conference PC / SPC

  • PCCVPR (2019-), ECCV (2018-), ICCV (2019-), AAAI (2019-), IJCAI (2018-)
  • SPCIJCAI (2021)

Courses

  1. Deep Learning Methods and ApplicationsUndergraduate
  2. Python ProgrammingUndergraduate
  3. Computer VisionUndergraduate
  4. Java ProgrammingUndergraduate
  5. Neural Networks and Deep LearningGraduate