Hello, I am Yeji Choi, currently a postdoc fellow at KAIST, working with Prof. ***Seungryong Kim.*** My research interests include computer vision, image generation & editing, and vision-language models.

I received my PhD degree in the School of Electrical and Electronic Engineering from Yonsei University in 2025, under the supervision of Prof. Kwanghoon Sohn (Yonsei University), with co-supervision from Prof. ***Ig-Jae Kim*** (Korea Institute of Science and Technology).

E-mail 📧: yejichoi@kaist.ac.kr

LinkedIn : linkedin.com/in/yeji-choi

💡Research Interests

• Computer vision
• Image Generation & Editing
• Vision-Language models
• Time Series Anomaly Detection

🎓 Educations

• [Mar. 2019 - Aug. 2025] Ph.D. in Electrical and Electronic Engineering
  • Yonsei University, Seoul, Korea
    • Advisor: Kwanghoon Sohn(Yonsei University) & Ig-Jae Kim (KIST)
    • Dissertation: Identity-Preserving Cross-Modal Face Synthesis via Domain-Invariant Embedding and Text-Guided Stylization
• [Mar. 2014 - Feb. 2016] M.S. in Department of Mathematics
  • Ewha Womans University, Seoul, Korea
    • Advisor: Jungho Yoon
    • Dissertation: Image Zooming Method based on Data-adapted Moving Least H_d for Salt-and-Pepper Noise
• [Mar. 2010 - Feb. 2014] B.S. in Department of Mathematics
  • Ewha Womans University, Seoul, Korea

📄 Publications

[International Conference]

• Navigating Label Ambiguity for Facial Expression Recognition in the Wild
  • JunGyu Lee*, Yeji Choi*, Haksub Kim, Ig-Jae Kim, Gi Pyo Nam (*: equal contribution)
  • Association for the Advancement of Artificial Intelligence (AAAI 2025)
• Face Photo-Sketch Synthesis via Domain-Invariant Feature Embedding
  • Yeji Choi, Kwanghoon Sohn, and Ig-Jae Kim
  • IEEE International Conference on Image Processing (ICIP 2023)
• GAN-based Anomaly Detection and Localization of Multivariate Time Series Data for Power Plant
  • Yeji Choi, Hyunki Lim, Heeseung Choi, and Ig-Jae Kim
  • IEEE International Conference on Big Data and Smart Computing (BigComp 2020)

[International Journal]

• Dual Transformers with Latent Amplification for Multivariate Time Series Anomaly Detection
  • Yeji Choi, Kwanghoon Sohn, and Ig-Jae Kim
  • IEEE Access, 2025