I am a Ph.D candidate student supervised by Jaegul Choo, at Kim Jaechul Graduate School of AI, KAIST. My research interest lies in reconstructing photo-realistic human heads with controllability. Particularly, my research topics include neural head avatar synthesis with rigging or relighting. I have worked on 2D face swapping with generative models, before.

Currently, I am a research intern at Meta Reality Labs with Carlos Aliaga. I collaborated deeply with Marcel Buehler from ETH Zurich. I was a a research intern at Naver AI Lab with Jin-Hwa Kim and Kakao Brain with Hyunjoon Lee. I completed a Masters in Computer Vision and Artificial Intelligence in KAIST, and a Bachelors in Engineering at DGIST.

Contact

  • wintermad1245 [at] kaist.ac.kr

    soulvip1245 [at] gmail.com

  • Bldg N1, Rm 211, 291 Daehak-ro, Yuseong-gu, Daejeon, Korea, 34141

Education

  • Ph.D in Artificial Intelligence, 2022~

    KAIST, Korea

    Advisor: Prof. Jaegul Choo

  • BS in Engineering, 2021

    DGIST, Korea

  • Summer Exchange Student, 2018

    UC Berkeley, United States

Research Experiences

  • Meta RL Research, California, USA
    Jun 2025 - Present

    Research Intern, Gemini Team.
  • Naver Cloud, Seongnam, Korea
    Aug 2024 - Feb 2025

    Research Intern, Generation Team.
  • Kakao Brain, Seongnam, Korea
    Jul 2022 - Sep 2022

    Research Intern, Neural Rendering TF team.
  • KAIST, Daejeon, Korea
    Dec 2020 - Feb 2022

    Research Intern, DAVIAN Lab
  • POSTECH, Pohang, Korea
    Jun 2020 - Aug 2020

    Research Intern, Computer Graphics Lab.
  • DGIST, Daegu, Korea
    Dec 2019 - Feb 2020

    Research Intern, Computer Vision Lab.

Publications

  • TexAvatars: Hybrid Texel-3D Representations for Stable Rigging of Photorealistic Gaussian Head Avatars

    Jaeseong Lee*, Junyeong Ahn*, Taewoong Kang, Jaegul Choo

    Under Review

    [ paper ]

  • SurFhead: Affine Rig Blending for Geometrically Accurate 2D Gaussian Surfel Head Avatars

    Jaeseong Lee*, Taewoong Kang*, Marcel C. Bühler, Min-Jung Kim, Sungwon Hwang, Junha Hyung, Hyojin Jang, Jaegul Choo

    ICLR 2025

    [ paper ]

  • Effective Rank Analysis and Regularization for Enhanced 3D Gaussian Splatting

    Junha Hyung, Susung Hong, Sungwon Hwang, Jaeseong Lee, Jaegul Choo, Jin-Hwa Kim

    NeurIPS 2024

    [ paper ]

  • SelfSwapper: A Simple yet Robust Face Swapping Model against Attribute Leakage

    Jaeseong Lee*, Junha Hyung*, Sohyun Jung, Jaegul Choo(* equal contribution)

    ECCV 2024

    [ paper ] [ project page ]

  • PixelHuman: Animatable Neural Radiance Fields from Few Images

    Gyumin Shim, Jaeseong Lee, Junha Hyung, Jaegul Choo

    arxiv preprint

    [ paper ]

  • Expression Domain Translation Network for Cross-domain Head Reenactment

    Taewoong Kang*, Jeongsik Oh*, Jaeseong Lee, Sunghyun Park, Jaegul Choo(* equal contribution)

    ICASSP 2024

    [ paper ] [ project page ]

  • RobustSwap: A Simple yet Robust Face Swapping Model against Attribute Leakage

    Jaeseong Lee*, Taewoo Kim*, Sunghyun Park, Younggun Lee, Jaegul Choo(* equal contribution)

    arxiv preprint

    [ paper ] [ project page ]

  • AnimeCeleb: Large-Scale Animation CelebHeads Dataset for Head Reenactment

    Kangyeol Kim*, Sunghyun Park*, Jaeseong Lee *, Sunghyo Chung, Junsoo Lee, Jaegul Choo(* equal contribution)

    ECCV 2022

    [ paper ] [ github ]

Awards & Honors

Academic Activities

Reviewer

  • 3DV 2026
  • ICLR 2025
  • ICCV 2023, 2025
  • CVPR 2022, 2023, 2024
  • ECCV 2022, 2024
  • ACCV 2024

Talks

  • 2023 SKT Computer Vision Research Seminar
    • Topic: Image Animation with Deep Learning
  • 2022 NAVER & KAIA Autumn Joint Academic Conference
    • Topic: Cross-domain Neural Talking Head Synthesis