JINYOUNG LEE
Portrait of Jinyoung Lee

Jin-Young Lee

Graph Neural Networks · Generative Models · Temporal & Spatial Dynamics

Elevator Pitch

I study Graph Neural Networks and Generative Models to efficiently process Temporal & Spatial Dynamics, under the supervision of Dr. Chul-Ho Lee.

I have always believed that science and technology can lead humanity toward a better world, and I have pursued this vision through diverse experiences. I began in Economics, motivated by the idea that accurate forecasting could prevent people from suffering in economic crises. While learning programming for data analysis, I developed a strong interest in information security, and worked as an intern at KEPCO KDN, a Korean public power corporation. Inspired by AlphaGo’s historic matches, I shifted my focus to AI and transferred to the Department of Software at Soongsil University. During an AI course, the Universal Approximation Theorem convinced me of AI’s world-changing capability. I gained industry experience as a machine vision engineer, developing automatic visual inspection systems delivered to factory production lines of companies such as Samsung/LG Display, LG Chem, and Hanwha QCELLS, and later pursued a master’s degree in computer vision, where I conducted several studies, with my main focus on applying time-series forecasting models to video data. Over time, I became increasingly interested in generative models for approximating statistical distributions and in graph algorithms for capturing complex temporal and spatial dynamics. These interests led me to my current doctoral research at Texas State University.

Research

Areas of Focus

  • Graph Generation
  • Dynamic Graph Modeling
  • Time Series/Spatiotemporal Forecasting

Research Papers

  • "Segment Anything Model for Anomaly Detection with Lightweight Domain Adapter", Y Kim, J Lee, G Kim, ITPM Journal, Jun 2025
  • "CNN-Based Time Series Decomposition Model for Video Prediction", J Lee, G Kim, IEEE ACCESS, Sep 2024
  • "Leveraging Time Series Decomposition for Efficient Spatiotemporal Feature Map Processing to Improve Video Prediction", J Lee, Master’s Thesis, Dec 2023
  • "Multi-View 3D Human Pose Estimation Based on Transformer", S Choi, J Lee, G Kim, Smart Media Journal, Dec 2023
  • "Few-Shot Learning of Tire Defect Image Classification with Deep Learning Model using Triplet Loss", J Lee, Y Kim, G Kim, ICROS-KROS joint conference, Dec 2022
  • "A Study on Tire Defect Detection Using Industrial Anomaly Detection Techniques in Unsupervised Learning", Y Kim, J Lee, G Kim, ICROS-KROS joint conference, Dec 2022

Experience

  • Researcher at Soongsil University (Computer Vision Lab), Mar 2024 – Feb 2025
  • Machine Vision Software Engineer at FAVION, Oct 2019 – Mar 2021
  • Cyber Security Operater Intern at KEPCO KDN, Jan 2016 – May 2016

Education

  • Texas State University — Ph.D. in Computer Science, Aug 2025 – Present
  • Soongsil University — M.S. in Software, Mar 2022 - Feb 2024
  • Soongsil University — B.S. in Software, Mar 2017 - Aug 2019
  • Kyung Hee University — B.S. in Economics, Mar 2009 - Feb 2016

Teaching

  • Coming soon

Service

  • Coming soon

Contact

Get in touch

Email: zby22@txstate.edu