Dr Yuanbo Wu

Senior Lecturer

Computer Science

l.y.wu@swansea.ac.uk

Office - 223
Second Floor
Computational Foundry
Bay Campus

Available For Postgraduate Supervision

Research Links

https://orcid.org/0000-0001-6119-058X

Google Scholar

ResearcherID

About

Dr Lin Yuanbo Wu received her PhD from The University of New South Wales, Kensington, Sydney, Australia. She has broad research interests across computer vision tasks and machine learning with strong motivation to address real-world challenges and has intensively published more than 70 peer-reviewed research articles (including two book chapters) in premier journals and proceedings. Dr Wu is specialized in video content understanding (such as video instance/panoptic/object segmentation, object detection/tracking), target re-identification, incremental learning, Generative AI, and AI-based advancement for medical study. She was a recipient of the Award for Growth of 2021 The 4th Eureka International Innovation and Entrepreneurship Competition (Eureka IIEC 2021, Melbourne, Australia). Dr Wu is currently serving as the Associate Editor with IEEE Trans on Neural Networks and Learning Systems (IF: 14.255), IEEE Trans on Multimedia (IF: 8.182), IEEE Trans on Big Data (IF: 4.271), and Pattern Recognition Letters (IF: 4.757). She was invited to serve as Area Chair with ACM Multimedia 2023 (Ottawa), 2022 (Lisbon). Dr Wu is a Senior Member of IEEE.

Areas Of Expertise

Computer vision
Object detection
Multi-object tracking
Instance and panoptic segmentation
Machine learning
AI for bioscience
Video generation
6D articulated object estimation

Career Highlights

Research

Lin is focusing on cutting-edge research challenges in a broad array of domains with computer vision, machine learning, and transcending emerging disciplines of artificial intelligence and biosciences. Current opportunities are outlined with video generation/editing, instance segmentation on both images and videos, protein structure generation/prediction, 6D articulated object estimation and target re-identification in surveillance.

Lin’s exploratory research scheme Generative AI challenge: A Tracking-Discriminating Deep Generative Model for Real-Time Anomaly Detection in Videos, is fully funded by Airbus Endeavr, Wales (May 2023-April 2024, £150K GBP), with possible extension from May 2024-April 2026 (approx. £300K GPB).

Publications

Research Highlights

Yuan, X., Zhang, Z., Wang, X., & Wu, Y. (2023). Semantic-Aware Adversarial Training for Reliable Deep Hashing Retrieval. IEEE Transactions on Information Forensics and Security, 18, 4681-4694.
https://doi.org/10.1109/tifs.2023.3297791, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa64127
Wu, Y., Liu, L., Wang, Y., Zhang, Z., Boussaid, F., Bennamoun, M., & Xie, X. (2023). Learning Resolution-Adaptive Representations for Cross-Resolution Person Re-Identification. IEEE Transactions on Image Processing, 32, 4800-4811.
https://doi.org/10.1109/tip.2023.3305817, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa64124
http://dx.doi.org/10.1109/tip.2023.3305817
Liu, D., Wu, L., Li, B., Boussaid, F., Bennamoun, M., Xie, X., Liang, C., & Wu, Y. (2024). Jacobian norm with Selective Input Gradient Regularization for interpretable adversarial defense. Pattern Recognition, 145, 109902
https://doi.org/10.1016/j.patcog.2023.109902, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa64108
http://dx.doi.org/10.1016/j.patcog.2023.109902
Wu, Y., Hong, R., Wang, Y., & Wang, M. (2019). Cross-Entropy Adversarial View Adaptation for Person Re-identification. IEEE Transactions on Circuits and Systems for Video Technology, 1-1.
https://doi.org/10.1109/tcsvt.2019.2909549, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63560
http://dx.doi.org/10.1109/tcsvt.2019.2909549
Liu, D., Wu, Y., Hong, R., Ge, Z., Shen, J., Boussaid, F., & Bennamoun, M. (2023). Generative Metric Learning for Adversarially Robust Open-world Person Re-Identification. ACM Transactions on Multimedia Computing, Communications, and Applications, 19(1), 1-19.
https://doi.org/10.1145/3522714, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63558
http://dx.doi.org/10.1145/3522714
Wu, Y., Liu, D., Guo, X., Hong, R., Liu, L., & Zhang, R. (2022). Multi-scale Spatial Representation Learning via Recursive Hermite Polynomial Networks. In Proceedings of the Thirty-First International Joint Conference on Artificial IntelligenceInternational Joint Conferences on Artificial Intelligence Organization.
https://doi.org/10.24963/ijcai.2022/204, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63557
http://dx.doi.org/10.24963/ijcai.2022/204
Wu, Y., Wang, Y., Yin, H., Wang, M., & Shao, L. (2019). Few-Shot Deep Adversarial Learning for Video-Based Person Re-Identification. IEEE Transactions on Image Processing, 29, 1233-1245.
https://doi.org/10.1109/tip.2019.2940684, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63201
http://dx.doi.org/10.1109/tip.2019.2940684
Wu, Y., Wang, Y., Gao, J., Wang, M., Zha, Z., & Tao, D. (2021). Deep Coattention-Based Comparator for Relative Representation Learning in Person Re-Identification. IEEE Transactions on Neural Networks and Learning Systems, 32(2), 722-735.
https://doi.org/10.1109/tnnls.2020.2979190, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63200
http://dx.doi.org/10.1109/tnnls.2020.2979190
Lin, Z., Zhang, D., Shi, D., Xu, R., Tao, Q., Wu, Y., He, M., & Ge, Z. (2023). Contrastive pre-training and linear interaction attention-based transformer for universal medical reports generation. Journal of Biomedical Informatics, 138, 104281
https://doi.org/10.1016/j.jbi.2023.104281, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63181
Liu, D., Wu, Y., Zheng, F., Liu, L., & Wang, M. (2022). Verbal-Person Nets: Pose-Guided Multi-Granularity Language-to-Person Generation. IEEE Transactions on Neural Networks and Learning Systems, 1-13.
https://doi.org/10.1109/tnnls.2022.3151631, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63180
http://dx.doi.org/10.1109/tnnls.2022.3151631

All Research

Journal Articles

Liu, D., Wu, L., Li, B., Boussaid, F., Bennamoun, M., Xie, X., Liang, C., & Wu, Y. (2024). Jacobian norm with Selective Input Gradient Regularization for interpretable adversarial defense. Pattern Recognition, 145, 109902
https://doi.org/10.1016/j.patcog.2023.109902, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa64108
http://dx.doi.org/10.1016/j.patcog.2023.109902
Wu, Y., Liu, L., Wang, Y., Zhang, Z., Boussaid, F., Bennamoun, M., & Xie, X. (2023). Learning Resolution-Adaptive Representations for Cross-Resolution Person Re-Identification. IEEE Transactions on Image Processing, 32, 4800-4811.
https://doi.org/10.1109/tip.2023.3305817, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa64124
http://dx.doi.org/10.1109/tip.2023.3305817
Yuan, X., Zhang, Z., Wang, X., & Wu, Y. (2023). Semantic-Aware Adversarial Training for Reliable Deep Hashing Retrieval. IEEE Transactions on Information Forensics and Security, 18, 4681-4694.
https://doi.org/10.1109/tifs.2023.3297791, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa64127
Lin, Z., Zhang, D., Shi, D., Xu, R., Tao, Q., Wu, Y., He, M., & Ge, Z. (2023). Contrastive pre-training and linear interaction attention-based transformer for universal medical reports generation. Journal of Biomedical Informatics, 138, 104281
https://doi.org/10.1016/j.jbi.2023.104281, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63181
Hu, C., Xie, X., & Wu, Y. (2023). Face reenactment via generative landmark guidance. Image and Vision Computing, 130, 104611
https://doi.org/10.1016/j.imavis.2022.104611, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa62159
Liu, D., Wu, Y., Hong, R., Ge, Z., Shen, J., Boussaid, F., & Bennamoun, M. (2023). Generative Metric Learning for Adversarially Robust Open-world Person Re-Identification. ACM Transactions on Multimedia Computing, Communications, and Applications, 19(1), 1-19.
https://doi.org/10.1145/3522714, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63558
http://dx.doi.org/10.1145/3522714
Ji, Z., Hu, J., Liu, D., Wu, Y., & Zhao, Y. (2023). Asymmetric Cross-Scale Alignment for Text-Based Person Search. IEEE Transactions on Multimedia, 25, 7699-7709.
https://doi.org/10.1109/tmm.2022.3225754, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa62251
Wu, Y., Liu, D., Zhang, W., Chen, D., Ge, Z., Boussaid, F., Bennamoun, M., & Shen, J. (2022). Pseudo-Pair Based Self-Similarity Learning for Unsupervised Person Re-Identification. IEEE Transactions on Image Processing, 31, 4803-4816.
https://doi.org/10.1109/tip.2022.3186746, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa61771
Liu, D., Wu, Y., Zheng, F., Liu, L., & Wang, M. (2022). Verbal-Person Nets: Pose-Guided Multi-Granularity Language-to-Person Generation. IEEE Transactions on Neural Networks and Learning Systems, 1-13.
https://doi.org/10.1109/tnnls.2022.3151631, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63180
http://dx.doi.org/10.1109/tnnls.2022.3151631
Wu, Y., Wang, Y., Gao, J., Wang, M., Zha, Z., & Tao, D. (2021). Deep Coattention-Based Comparator for Relative Representation Learning in Person Re-Identification. IEEE Transactions on Neural Networks and Learning Systems, 32(2), 722-735.
https://doi.org/10.1109/tnnls.2020.2979190, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63200
http://dx.doi.org/10.1109/tnnls.2020.2979190
Wu, Y., Wang, Y., Yin, H., Wang, M., & Shao, L. (2019). Few-Shot Deep Adversarial Learning for Video-Based Person Re-Identification. IEEE Transactions on Image Processing, 29, 1233-1245.
https://doi.org/10.1109/tip.2019.2940684, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63201
http://dx.doi.org/10.1109/tip.2019.2940684
Wu, Y., Hong, R., Wang, Y., & Wang, M. (2019). Cross-Entropy Adversarial View Adaptation for Person Re-identification. IEEE Transactions on Circuits and Systems for Video Technology, 1-1.
https://doi.org/10.1109/tcsvt.2019.2909549, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63560
http://dx.doi.org/10.1109/tcsvt.2019.2909549

Conference Contributions

Wu, Y., Liu, D., Guo, X., Hong, R., Liu, L., & Zhang, R. (2022). Multi-scale Spatial Representation Learning via Recursive Hermite Polynomial Networks. In Proceedings of the Thirty-First International Joint Conference on Artificial IntelligenceInternational Joint Conferences on Artificial Intelligence Organization.
https://doi.org/10.24963/ijcai.2022/204, SU Repository: https://cronfa.swan.ac.uk/Record/cronfa63557
http://dx.doi.org/10.24963/ijcai.2022/204

Supervision

An Advanced Quantitative Framework for Unsupervised Multi-Object Instance Segmentation: Integrating Latent Dynamics, Stochastic Geometry, Algebraic Topology, Persistent Homology and more (current)

PhD

Other supervisor: Prof Xianghua Xie
The Assessment and Spotting of Generative Artificial Intelligence (current)

PhD

Other supervisor: Prof Xianghua Xie
Enhancing the designability and functional efficacy of the anti-body using deep learning methods. (current)

MSc

Other supervisor: Dr Benjamin Mora
Generative AI in Protein Design: De novo Protein Design and Multi-Motif Scaffolding (current)

MSc

Other supervisor: Dr Jonathan Mullins

Taught Modules

CS-109 Technologies for Information Presentation

This module is about the technologies and markup languages that make various forms of data presentation possible. It will explore web-based presentation languages including HTML, XML and CSS, text presentation languages including LaTeX, and mathematical / graphical presentation through R. After studying this module, students will be able to build web sites, produce professional-quality reports and typeset and visualize mathematical formulae and data, using the appropriate graphs for the data. The students will also be able to produce a crawler / scraper that can pull data from websites automatically for analysis.
CSC009 Technologies for Information Presentation

This module is about the technologies and markup languages that make various forms of data presentation possible. It will explore web-based presentation languages including HTML, XML and CSS, text presentation languages including LaTeX, and mathematical / graphical presentation through R. After studying this module, students will be able to build web sites, produce professional-quality reports and typeset and visualize mathematical formulae and data, using the appropriate graphs for the data. The students will also be able to produce a crawler / scraper that can pull data from websites automatically for analysis.
CSCM77 Computer Vision and Deep Learning

This module introduces students to the important and modern topics and concepts of computer vision and deep learning, including image processing, feature extraction, camera calibration, stereo vision, motion and tracking, recognition, deep neural network and their application to vision problems. It teaches techniques that are used to understand and interpret the contents of images and videos and dissects state-of-the-art vision systems, such as Microsoft Kinect. Practical examples in Python are provided throughout the lectures.

Dr Yuanbo Wu

Email address

Research Links

About

Areas Of Expertise

Career Highlights

Publications

Research Highlights

All Research

Journal Articles

Conference Contributions

Supervision

An Advanced Quantitative Framework for Unsupervised Multi-Object Instance Segmentation: Integrating Latent Dynamics, Stochastic Geometry, Algebraic Topology, Persistent Homology and more (current)

The Assessment and Spotting of Generative Artificial Intelligence (current)

Enhancing the designability and functional efficacy of the anti-body using deep learning methods. (current)

Generative AI in Protein Design: De novo Protein Design and Multi-Motif Scaffolding (current)

Taught Modules

CS-109 Technologies for Information Presentation

CSC009 Technologies for Information Presentation

CSCM77 Computer Vision and Deep Learning