Profile of Dr FU, Yujun Eugene 符欲均 - The Education University of Hong Kong


Assistant Professor	Centre for Learning, Teaching and Technology

Contact

ORCiD

0000-0003-1048-1904

Phone

(852) 2948 7936

Fax

(852) 2948 7046

eugenefu@eduhk.hk

Address

10 Lo Ping Road, Tai Po, New Territories, Hong Kong

Website

https://pappl.eduhk.hk/rich/web/person.xhtml?pid=316523&name=FU, Yujun Eugene

Scopus ID

57189356098

Research Interests

Artificial Intelligence
Machine Learning
Multimedia Computing
Human-Centered Computing
Reliable AI
AI security

Research Interests

Artificial Intelligence
Machine Learning
Multimedia Computing
Human-Centered Computing
Reliable AI
AI security

Research Outputs


Journal Publications Huiling Hu, Zhaoyun Ding, Eugene Yujun Fu, Peter HF Ng, Andy SK Cheng (2026). The SEWAbility system: A video-based job analysis framework for understanding task-specific job demands. Scientific Reports, 16, Article 10370. https://doi.org/10.1038/s41598-026-41536-w Kangzhong Wang, Xinwei Zhai, MK Michael Cheung, Eugene Yujun Fu, Peter Qi Chen, Grace Ngai, Hong Va Leong (2025). TMAN: A temporal multimodal attention network for backchannel detection. Neurocomputing, 657, Article 131605. https://doi.org/10.1016/j.neucom.2025.131605 MK Michael Cheung, Zhongqi Yang, Xinwei Zhai, Eugene Yujun Fu, Grace Ngai, Hong Va Leong, Lily Chan, Bei Du, Ruihua Wei, Chi-wai Do (2025). Refractive error detection in smartphone images via convolutional neural network. International Journal of Medical Informatics, 205, Article 106083. https://doi.org/10.1016/j.ijmedinf.2025.106083 Zhai, X., Wang, Y., Liang, L., Wang, K., Pei, F., & Fu, E. Y. (2025). Personalized e-learning resource recommendation using multimodal-enhanced collaborative filtering. Knowledge-Based Systems, 319, 113605. https://doi.org/10.1016/j.knosys.2025.113605
Conference Papers Denglin Kang, Youqian Zhang, Wai Cheong Tam, Xiapu Luo, Eugene Yujun Fu (2025, December). Anti-ESIA: Analyzing and mitigating impacts of electromagnetic signal injection attacks on image sensing. Advances in mobile computing and multimedia intelligence, MoMM 2025, Matsue, Japan. https://doi.org/10.1007/978-3-032-11768-7_7 Kangzhong Wang, Eugene Yujun Fu, Grace Ngai, Hong Va Leong (2025, October). Multimodal learning analytics for predicting learning gains in online service-learning programs. Behavioural and social computing: 12th International Conference, BESC 2025, Hong Kong SAR, China, October 16–18, 2025, Proceedings, Part I, Hong Kong. https://doi.org/10.1007/978-981-95-7138-3_32 Kangzhong Wang, Zitong Shen, Youqian Zhang, Michael MK Cheung, Xiapu Luo, Grace Ngai, Eugene Yujun Fu (2025, October). One size fits all? A Modular Adaptive Sanitization Kit (MASK) for customizable privacy-preserving phone scam detection. Proceedings of the 33rd ACM International Conference on Multimedia, MM 2025, Dublin, Ireland. https://doi.org/10.1145/3746027.3758164 Wenhao Liao, Yuanyuan Wang, Xinwei Zhai, Sineng Yan, Kai Zhong, Eugene Yujun Fu (2025, October). Student learning engagement recognition method based on video transformer. Behavioural and social computing: 12th International Conference, BESC 2025, Hong Kong SAR, China, October 16–18, 2025, Proceedings, Part I, Hong Kong. https://doi.org/10.1007/978-981-95-7138-3_18 SDGs infomation: 9 - Industry, Innovation and Infrastructure Yuanyuan Wang, Grace Ngai, Hong Va Leong, Stephen CF Chan, Eugene Yujun Fu (2025, October). Using topic detection to analyze student reflections in service-learning: A text mining method for understanding learning gains. Behavioural and social computing: 12th International Conference, BESC 2025, Hong Kong SAR, China, October 16–18, 2025, Proceedings, Part I, Hong Kong. https://doi.org/10.1007/978-981-95-7138-3_27 Liao, W., Yan, S., Zhang, Y., Zhai, X., Wang, Y., & Fu, E. (2025, April). Is Your Autonomous Vehicle Safe? Understanding the Threat of Electromagnetic Signal Injection Attacks on Traffic Scene Perception. Proceedings of the AAAI Conference on Artificial Intelligence, Washington, DC, USA. https://doi.org/10.1609/aaai.v39i26.34958 Shen, Z., Wang, K., Zhang, Y., Ngai, G., & Fu, E. Y. (2025, April). Combating Phone Scams with LLM-based Detection: Where Do We Stand? (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, Washington, DC, USA. https://doi.org/10.1609/aaai.v39i28.35298 Shen, Z., Yan, S., Zhang, Y., Luo, X., Ngai, G., & Fu, E. Y. (2025, April). "It Warned Me Just at the Right Moment": Exploring LLM-based Real-time Detection of Phone Scams. Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, Yokohama, Japan. https://doi.org/10.1145/3706599.3720263

Journal Publications
Huiling Hu, Zhaoyun Ding, Eugene Yujun Fu, Peter HF Ng, Andy SK Cheng (2026). The SEWAbility system: A video-based job analysis framework for understanding task-specific job demands. Scientific Reports, 16, Article 10370. https://doi.org/10.1038/s41598-026-41536-w
Kangzhong Wang, Xinwei Zhai, MK Michael Cheung, Eugene Yujun Fu, Peter Qi Chen, Grace Ngai, Hong Va Leong (2025). TMAN: A temporal multimodal attention network for backchannel detection. Neurocomputing, 657, Article 131605. https://doi.org/10.1016/j.neucom.2025.131605
MK Michael Cheung, Zhongqi Yang, Xinwei Zhai, Eugene Yujun Fu, Grace Ngai, Hong Va Leong, Lily Chan, Bei Du, Ruihua Wei, Chi-wai Do (2025). Refractive error detection in smartphone images via convolutional neural network. International Journal of Medical Informatics, 205, Article 106083. https://doi.org/10.1016/j.ijmedinf.2025.106083
Zhai, X., Wang, Y., Liang, L., Wang, K., Pei, F., & Fu, E. Y. (2025). Personalized e-learning resource recommendation using multimodal-enhanced collaborative filtering. Knowledge-Based Systems, 319, 113605. https://doi.org/10.1016/j.knosys.2025.113605

Conference Papers
Denglin Kang, Youqian Zhang, Wai Cheong Tam, Xiapu Luo, Eugene Yujun Fu (2025, December). Anti-ESIA: Analyzing and mitigating impacts of electromagnetic signal injection attacks on image sensing. Advances in mobile computing and multimedia intelligence, MoMM 2025, Matsue, Japan. https://doi.org/10.1007/978-3-032-11768-7_7
Kangzhong Wang, Eugene Yujun Fu, Grace Ngai, Hong Va Leong (2025, October). Multimodal learning analytics for predicting learning gains in online service-learning programs. Behavioural and social computing: 12th International Conference, BESC 2025, Hong Kong SAR, China, October 16–18, 2025, Proceedings, Part I, Hong Kong. https://doi.org/10.1007/978-981-95-7138-3_32
Kangzhong Wang, Zitong Shen, Youqian Zhang, Michael MK Cheung, Xiapu Luo, Grace Ngai, Eugene Yujun Fu (2025, October). One size fits all? A Modular Adaptive Sanitization Kit (MASK) for customizable privacy-preserving phone scam detection. Proceedings of the 33rd ACM International Conference on Multimedia, MM 2025, Dublin, Ireland. https://doi.org/10.1145/3746027.3758164
Wenhao Liao, Yuanyuan Wang, Xinwei Zhai, Sineng Yan, Kai Zhong, Eugene Yujun Fu (2025, October). Student learning engagement recognition method based on video transformer. Behavioural and social computing: 12th International Conference, BESC 2025, Hong Kong SAR, China, October 16–18, 2025, Proceedings, Part I, Hong Kong. https://doi.org/10.1007/978-981-95-7138-3_18
SDGs infomation: 9 - Industry, Innovation and Infrastructure
Yuanyuan Wang, Grace Ngai, Hong Va Leong, Stephen CF Chan, Eugene Yujun Fu (2025, October). Using topic detection to analyze student reflections in service-learning: A text mining method for understanding learning gains. Behavioural and social computing: 12th International Conference, BESC 2025, Hong Kong SAR, China, October 16–18, 2025, Proceedings, Part I, Hong Kong. https://doi.org/10.1007/978-981-95-7138-3_27
Liao, W., Yan, S., Zhang, Y., Zhai, X., Wang, Y., & Fu, E. (2025, April). Is Your Autonomous Vehicle Safe? Understanding the Threat of Electromagnetic Signal Injection Attacks on Traffic Scene Perception. Proceedings of the AAAI Conference on Artificial Intelligence, Washington, DC, USA. https://doi.org/10.1609/aaai.v39i26.34958
Shen, Z., Wang, K., Zhang, Y., Ngai, G., & Fu, E. Y. (2025, April). Combating Phone Scams with LLM-based Detection: Where Do We Stand? (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, Washington, DC, USA. https://doi.org/10.1609/aaai.v39i28.35298
Shen, Z., Yan, S., Zhang, Y., Luo, X., Ngai, G., & Fu, E. Y. (2025, April). "It Warned Me Just at the Right Moment": Exploring LLM-based Real-time Detection of Phone Scams. Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, Yokohama, Japan. https://doi.org/10.1145/3706599.3720263

Projects


Special Teaching Assistant with Responsive Sensing (STARS): Research and Development on Multimodal AI-Driven Educational Support Systems for SEN Students The STARS project will develop an innovative multimodal AI system addressing critical support gaps for students with autism and moderate intellectual disabilities. We aim to create a platform providing personalized, real-time learning support by accurately analyzing students' behavioral and emotional states through multiple sensing channels. Our methodology integrates computer vision, audio analysis, and behavioral monitoring with our industrial partner’s (Bridge AI) established wearable and environmental sensor technology. These multimodal inputs will feed novel AI models designed to detect attention, engagement, emotion, and conversational cues in SEN students. We will collect data and validate our approach in Rhenish Church Grace (RCG) School, our partner special school, ensuring the system meets real educational needs. An interactive LLM voice agent will provide adaptive guidance to students, while a comprehensive dashboard will offer teachers data-driven insights and intervention suggestions. STARS will impact special education by: (1) Empowering teachers with objective insights into student states, reducing subjective guesswork; (2) Providing students with a responsive, individualized learning companion; (3) Creating a closed-loop "Sense → Analyze → Respond" system for real-time support. The benefits include reducing teacher burden, increasing student engagement, improving learning outcomes, and addressing the critical "knowledge transfer gap" between classroom settings and real-world applications. Project Start Year: 2025, Principal Investigator(s): FU, Yujun, Eugene SDGs Information: 9 - Industry, Innovation and Infrastructure
Research and Development of Robust Refractive Error Detection Utilizing Machine Learning Models and Smartphone Videos Refractive error is a serious medical problem in Hong Kong. When it goes undetected and uncorrected, it can potentially lead to severe vision impairment and worse academic performance. The inconvenience, professional-dependent, as well as time and financial high-cost of the conventional vision examination limit the regular refractive error detection of children, resulting in a disappointingly low vision screening rate in Hong Kong. This project proposes to research and develop a robust refractive error detection method utilising machine learning models and smartphone videos. The proposed method can eliminate the inherent limitations of image-based methods, enhancing the accuracy and robustness of smartphone-based refractive error detection in practical applications. The output of the project has potential to achieve low-cost, robust and accurate detection with smartphone as the sole device. The method will be comprehensively evaluated in the project, including the evaluation of the model detection accuracy as well as user acceptance. It is believed that this project will contribute to a better vision well-being for the whole society in the future. Project Start Year: 2025, Principal Investigator(s): DO, Chi-wai (FU, Yujun, Eugene as Co-Principal Investigator)
Modeling Individual Differences in Nonverbal Communication Cues: A Machine Learning Framework for Enhanced Backchannel and Engagement Detection Human interaction and communication analysis via nonverbal cues (e.g., head movement, eye gaze) holds transformative potential for education (e.g., student engagement detection), mental healthcare (e.g., AI-based assessment and intervention), and human-computer interaction. However, current nonverbal communication cues analysis and engagement detection often fail to account for individual differences in backchannel and engagement expressing, leading to biased or ineffective detection. Building on the PI’s prior work in video-based machine learning (ML) models for detecting nonverbal backchannel signals (e.g., head nods, eye contact), this project extends to address two critical gaps: (1) Individual differences in expressing backchannel and agreement; and (2) Adaptation of models for engagement detection in the context of novice-expert dyadic conversations. It will explore deep learning techniques, focusing on attention-based/transformer architectures trained on multimodal visual data (head, eye, body, facial features), incorporating with individual standardization framework. Comprehensive evaluation and analysis will be conducted on publicly available datasets to identify the optimal setup for the approach. Furthermore, the developed framework and models are expected to be validated on real applications through the collaborations with educational, industry, and clinical partners in the future.Objectives (1) To develop a ML framework that accounts for individual differences in nonverbal communication cues (e.g., head movement, eye gaze, facial action units) during human interactions, based on existing attention/transformer-based models. (2) To adapt the developed framework and models to detect engagement signals (beyond backchannels) in the context of novice-expert dyadic conversations. (3) To lay the groundwork for scalable applications in education, SEN support, mental health assessment, and AI-driven engagement interventions. Project Start Year: 2025, Principal Investigator(s): FU, Yujun, Eugene SDGs Information: 9 - Industry, Innovation and Infrastructure

Prizes and awards


Silver Medal, International Exhibition of Inventions Geneva 2025 Date of receipt: 12/4/2025, Conferred by: International Exhibition of Inventions Geneva
Silver Medal, International Exhibition of Inventions Geneva 2023 Date of receipt: 28/4/2023, Conferred by: International Exhibition of Inventions Geneva