Ph.D. Student @ Yonsei University Graduate School of Artificial Intelligence
M.S. in AI-Robotics, KIST School (UST)
BS in Big-Data Analytics, Kookmin University
I'm currently researching Trajectory Prediction and Crowd Behavior Modeling.
My work focuses on building human-centered AI systems that understand and predict complex human movements.
Previously, I worked on Facial Expression Recognition, Face Recognition, and real-time vision-based applications during my M.S. program.
I aim to develop AI technologies that genuinely help people and contribute to society through meaningful, human-focused research.
A novel method that addresses both label ambiguity and class imbalance in FER using adaptive weighting and consistency regularization.
📄 Published at AAAI 2025
🔗 Official Code
An extension of NLA into the video domain. V-NAW leverages temporal dynamics and noise-aware sample emphasis for robust video-based FER.
📄 Published at CVPRW 2025
🔗 Official Code
Study on Facial Composite Feature Analysis for Determining Subject Anxiety Levels on Low-Power Computing Modules (IEIE 2024)
A study on estimating subject anxiety using eye blinking and facial expression recognition on embedded systems.
📄 Published at IEIE Summer Annual Conference 2024
🔗 Official Code
A PyQt-based app for facial expression recognition from images, videos, or real-time webcam input. Supports multi-face detection, expression classification, and visualizes softmax probability distributions.
Real-time facial expression recognition using softmax-based classification.
Includes face detection, preprocessing, and visualization.
Extension of expression recognition to continuous affect modeling via valence–arousal space.
Includes real-time video interface and visual affect mapping.
Real-time eye blink detection tailored for behavioral analysis and human state monitoring. Optimized for lightweight deployment on embedded systems and suitable for integration into anxiety estimation pipelines.
Lightweight face detector for real-time webcam input.
Based on SSD and RetinaFace; tuned for single-person detection scenarios.
More project repositories and demos are available on my GitHub profile.
Feel free to check out the code, raise issues, or collaborate!
- Email: [email protected]
- GitHub: github.com/jungyu0413
- Google Scholar: scholar.google.com/citations?user=2pnIJggAAAAJ&hl=ko
- LinkedIn: linkedin.com/in/jungyu-lee-0315sb