Wei Xin (魏欣)

Doctor, Vice Dean, Lecturer

Supervisor of Software Engineering

School of Software, Nanchang University

Research Area: machine learning, deep learning, computer vision, face image processing, and medical image processing.

Email: xinwei@ncu.edu.cn

Overview:

Dr. Wei holds the position of Lecturer of Software Engineering at the School of Software, Nanchang University, and is the Vice Dean of School of Continuing Education, Nanchang University.

He obtained the master's degree in Computer Application and Technology from the School of Mathematics and Computer Science, Fujian Normal University, China, in 2016; and the PhD of Artificial Intelligence from Ulster University, UK, in 2020. He was the research associate of the School of Computing, Ulster University, UK.

He was selected for Nanchang University's 2021 "Xiangzhang Education Program". Five patents have been approved and more than 20 papers have been published, including 11 first-author papers indexed by SCI and EI, of which ten papers have been published in mainstream journals and conferences in the field of AI. His papers have been cited by top international journals in the field of artificial intelligence, such as IEEE TPAMI, IEEE TNNLS, Information Sciences, IEEE TSMC, and PR. He has served as a reviewer for top AI journals and conferences such as AAAI, IEEE Trans. Cybern., Knowledge Based Systems, etc.

Dr. Wei is the Principal Investigator of the National Natural Science Foundation of China (Youth Project): One-shot Face Identification based on Self-adaptive Balanced Learning and Semantic Understanding. He participated in two EU Horizon 2020 projects: Analysis System for Gathered Raw Data (ASGARD), Decision Support and Information Management System for Breast Cancer (DESIREE), one Northern Ireland Regional Project: The Use of Unobtrusive Technology for the Identification of Agitation, Falls and Life Sign, one National Natural Science Foundation of China Project: High-adaptive passive forensics research on deep fake face video, one National Natural Science Foundation of China Regional Project: Pedestrian Re-Identification Based on Pose Feature Matching and Cross-Domain Unsupervised Learning. He presided over one school-level teaching reform project and one first-class curriculum construction project, mentored graduate students and undergraduates to win more than ten awards in domestic and overseas science and technology competitions and guided one national innovation and entrepreneurship plan project for undergraduates. He instructed undergraduates to win the honor of the school's outstanding graduation thesis two times.

Publications:

1. Xin Wei, Hui Wang, Bryan Scotney and Huan Wan. Minimum Margin Loss for Deep Face Recognition. Pattern Recognition. 97 (2020): 107012.

2. Xin Wei, Hui Wang, Bryan Scotney and Huan Wan. Selective multi-descriptor fusion for face identification. International Journal of Machine Learning and Cybernetics. 2019: 1-13.

3. Xin Wei, Hui Wang, Gongde Guo and Huan Wan. Multiplex image representation for enhanced recognition. International Journal of Machine Learning and Cybernetics. 2018: 383-392.

4. Xin Wei, Hui Wang, Bryan Scotney and Huan Wan. Precise Adjacent Margin Loss for Deep Face Recognition. IEEE International Conference on Image Processing (ICIP) 2019, Sep 22 (pp. 3457-3461).

5. Xin Wei, Hui Wang, Bryan Scotney and Huan Wan. GicoFace: Global Information-based Cosine Optimal Loss for Deep Face Recognition. IEEE International Conference on Image Processing (ICIP) 2019, Sep 22 (pp. 3641-3645).

6. Huan Wan, Hui Wang, Gongde Guo and Xin Wei. Separability-oriented subclass discriminant analysis. IEEE transactions on pattern analysis and machine intelligence. 2017 Feb 22; 40(2): 409-22.

7. Xin Wei, Hui Wang, Bryan Scotney and Huan Wan. A Minkowski Distance-based Generalisation Method for Improving Centre Loss for Deep Face Recognition (Best Student Paper Award). Irish Machine Vision and Image Processing Conference (IMVIP) 2018, Aug (pp. 154-161). IPRCS.

8. Xin Wei, Hui Wang, Huan Wan and Bryan Scotney. Self-adaptive Feature Fusion Method for Improving LBP for Face Identification. International Conference on Computer Vision Systems (ICVS) 2017, Jul 10 (pp. 373-383). Springer, Cham.

9. Xin Wei, Hui Wang, Huan Wan and Bryan Scotney. Multiscale Feature Fusion for Face Identification. IEEE International Conference on Cybernetics (CYBCONF) 2017, Jun 21 (pp. 1-6). )

10. Xin Wei, Gongde Guo, Hui Wang and Huan Wan. A multiscale method for HOG-based face recognition. International Conference on Intelligent Robotics and Applications (ICIRA) 2015, Aug (pp. 535-545). Springer, Cham.

11. Xin Wei, Huan Wan and etc. HPS-Net: Multi-Task Network for Medical Image Segmentation with Predictable Performance. Symmetry 2021, 13, 2107.

12. Xin Wei, Wei Du and etc. GicoFace: A Deep Face Recognition Model Based on Global-Information Loss Function. Electronics 2021, 10, 2387.

13. Haoyu Zhao, Weidong Min*, Xin Wei, Qi Wang, Qiyan Fu and Zitai Wei. MSR‐FAN: Multi‐scale residual feature‐aware network for crowd counting. IET Image Processing , 1– 10 (2021) https://doi.org/10.1049/ipr2.12175.

14. Guangmin Li, Zhiwei Lin, Hui Wang, Xin Wei. A Discriminative Approach to Sentiment Classification[J]. Neural Processing Letters, 2020, 51(1): 749-758.

15. Huan Wan, Gongde Guo, Hui Wang and Xin Wei. A new linear discriminant analysis method to address the over-reducing problem[C]//International Conference on Pattern Recognition and Machine Intelligence. Springer, Cham, 2015: 65-72.

Fundings:

1. National Natural Science Foundation of China, Youth Project, 62106093, One-shot Face Identification based on Self-adaptive Balanced Learning and Semantic Understanding, January 1, 2022 to December 31, 2024, 300,000RMB, under research, program leader.

2. EU H2020 project, FCT-1-2015, Analysis System for Gathered Raw Data (ASGARD), from September 1, 2016 to November 30, 2020, 670,000 euros, principal investigator.

3. EU H2020, PHC-30-2015, Decision Support and Information Management System for Breast Cancer (DESIREE), from February 1, 2016 to January 31, 2019, 550,000 euros, principal investigator.

4. Northern Ireland Government, Lava Group and I+ Corporation, The use of unobtrusive technology for the identification of agitation, falls and life signs, 1 March 2020 to 28 February 2021, £150,000, principal investigator.

5. National Natural Science Foundation of China, General Project, 62072106, High-adaptive passive forensics research on deep fake face video, January 1, 2021 to December 31, 2024, 560,000, principal investigator.

6. National Natural Science Foundation of China, Regional Project, 62166026, Pedestrian Re-Identification Based on Pose Feature Matching and Cross-Domain Unsupervised Learning, January 1, 2022 to December 31, 2025, 350,000, principal investigator.

Awards:

1. 2016 UK PhD Full Scholarship.

2. Best Paper Awards in 2018 Irish Machine Vision and Image Processing Conference.

3. In 2022, guided students to obtain excellent graduation design (Qiaoling Huang , Yunhan Wang ).

4. 2021-2022 excellent class instructor.

5. Excellent work quality and professional ethics appraisal in 2022.