TAEKYUNG KI


I am a researcher at KAIST MLAI working on generative models and computer vision.

Bio: From March 2022 to March 2025, I was a researcher working on video generation as an alternative military service in South Korea. I started researching deep learning in February 2021. I received M.S in mathematics in February 2021, and B.S. in mathematics in February 2019.

I am interested in the following research topics:

  • Generative models (score-based and flow-based models, etc.)
  • 3D Gaussian splatting and neural implicit representation
  • Human video generation
  • Human motion representation
  • Audio-visual and Vision-language representation

If you have an interest in the above research topics, feel free to drop me an e-mail.

Mail Twitter Scholar Github LinkedIn

Profile picture

Highlights

  • [Mar. 2025] Finished alternative military service in South Korea.
  • [July 2024] One paper accepted to ECCV 2024.
  • [July 2023] One paper accepted to ICCV 2023.

Publications

* : Equal contribution, † : Corresponding author

FLOAT
FLOAT: Generative Motion Latent Flow Matching for Audio-driven Talking Portrait
Taekyung Ki, Dongchan Min, Gyeongsu Chae
arXiv 2024
Paper / ProjectPage / Code /
@article{ki2024float,
    title={FLOAT: Generative Motion Latent Flow Matching for Audio-driven Talking Portrait},
    author={Ki, Taekyung and Min, Dongchan and Chae, Gyeongsu},
    journal={arXiv preprint arXiv:2412.01064},
    year={2024}
    }
Project image
Export3D: Learning to Generate Conditional Tri-plane for 3D-aware Expression Controllable Portrait Animation
Taekyung Ki, Dongchan Min, Gyeongsu Chae
European Conference on Computer Vision (ECCV) 2024, Milano, Italy
Paper / ProjectPage / Supp /
@inproceedings{ki2024learning,
    title={Learning to generate conditional tri-plane for 3d-aware expression controllable portrait animation},
    author={Ki, Taekyung and Min, Dongchan and Chae, Gyeongsu},
    booktitle={European Conference on Computer Vision},
    pages={476--493},
    year={2024},
    organization={Springer}
    }
Project image
StyleLipSync: Style-based Personalized Lip-sync Video Generation
Taekyung Ki*, Dongchan Min*
International Conference on Computer Vision (ICCV) 2023, Paris, France
Paper / ProjectPage / Supp / ArXiv / Code /
@InProceedings{Ki_2023_ICCV,
    author    = {Ki, Taekyung and Min, Dongchan},
    title     = {StyleLipSync: Style-based Personalized Lip-sync Video Generation},
    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
    month     = {October},
    year      = {2023},
    pages     = {22841-22850}
    }
Project image
Deep Scattering Network with Max-pooling
Taekyung Ki, Youngmi Hur
IEEE 2021 Data Compression Conference (DCC), Snowbird UT, USA
Paper / Code /
@InProceedings{ki2021deep, 
    author = {Taekyung Ki and Youngmi Hur}, 
    title = {Deep Scattering Network with Max-pooling}, 
    booktitle = {2021 Data Compression Conference, DCC 2021}, 
    year = {2021}
    }

Awards and Honors

  • [Oct. 2021] 1st Prize Winner, NLP-based Math-Word Problem Track in 2021 AI Grand Challenge (AGC), funded by South Korea Ministry of Science and ICT.
  • [Feb. 2019] Graduation with top honors in mathematics.

Academic Services

  • Reviewer: CVPR 2025, ICCV 2025.
- This website is based on the lightweight and easy-to-use template from Michael Niemeyer.