Dongguk University, Republic of Korea
Biometric Recognition · Image Restoration · Edge AI Deployment
I am a master's student specializing in AI and computer vision, with research interests in biometric recognition, image restoration, and deployment-oriented deep learning systems.
My recent work focuses on:
- 🩸 Finger-vein recognition and image outpainting
- 🎯 Knowledge distillation for lightweight vision models
- 🕵️ Deepfake detection and robust visual classification
- 🚗 Semantic segmentation for autonomous driving
- ⚡ Edge AI deployment on embedded and resource-constrained environments
I am particularly interested in developing vision models that are not only accurate, but also practical for real-world deployment.
Deep learning-based image outpainting of finger-vein image
- 🧠 Finger-vein image outpainting using knowledge distillation and adaptive feature attention
- 🛠 Implemented in PyTorch
- 📰 Published in Expert Systems with Applications (ESWA)
- 🔗 DOI: 10.1016/j.eswa.2025.131015
Deep learning-based finger-vein image outpainting project based on knowledge distillation and adaptive feature attention for biometric image restoration.
Synthetic finger-vein image generation project using diffusion models for spoof sample generation and real/fake anti-spoofing classification.
Deepfake detection pipeline based on Swin Transformer V2 with EFFORT-style SVD-residual fine-tuning.
🏆 Leaderboard: Top 6.7% (31 / 458)
Semantic segmentation project for autonomous driving with DDRNet, teacher-student knowledge distillation, degradation-based augmentation, and rare-class-aware oversampling.
News big data hackathon project developed in an organization repository, selected as a finalist-level team entry in the 2025 News Big Data Hackathon.
- Deep learning-based image outpainting of finger-vein image
Expert Systems with Applications, 2026
🔗 DOI: 10.1016/j.eswa.2025.131015