Kangyeol Kim

Co-founder & AI Team Leader at Letsur inc.
Ph.D. Student at KAIST
Contact: kangyeolk at {letsur dot ai | kaist dot ac dot kr}


About Me

I am co-founder and AI team leader of Letsur inc., developing automated solutions for AI development. Our company strives to make AI development process easier and faster with our product. In parallel, I am a Ph.D. student at Korea Advanced Institute of Science and Technology (KAIST), working on machine learning and its applications under the supervision of Jaegul Choo. As a researcher, I have an interest in understanding and controlling visual information in images and videos.


News

[Jul. 2022] 2 papers accepted to ECCV 2022.
[Mar. 2021] I will start my Ph.D. course at Korea Advanced Institute of Science and Technology (KAIST).
[Jan. 2021] I started a research internship at NAVER WEBTOON Corp.
[Dec. 2020] Our paper got accepted to AAAI 2021.
[Jun. 2018] I started a research internship at NAVER Corp.


Publications

Reference-based Image Composition with Sketch via Structure-aware Diffusion Model
Kangyeol Kim, Sunghyun Park, Junsoo Lee, and Jaegul Choo
AI for Content Creation Workshop at CVPR, 2023
[Paper] [Code]

AnimeCeleb: Large-Scale Animation CelebHeads Dataset for Head Reenactment
Kangyeol Kim*, Sunghyun Park*, Jaeseong Lee*, Sunghyo Chung, Junsoo Lee, and Jaegul Choo
European Conference on Computer Vision (ECCV), 2022, Tel Aviv, Accepted
Best Paper Award, Korean Artificial Intelligence Association 2022
[Paper] [Code]

Style Your Hair: Latent Optimization for Pose-Invariant Hairstyle Transfer via Local-Style-Aware Hair Alignment
Chaeyeon Chung*, Taewoo Kim*, Yoonseo Kim*, Sunghyun Park, Kangyeol Kim, and Jaegul Choo
European Conference on Computer Vision (ECCV), 2022, Tel Aviv, Accepted
[Paper] [Code]

Continuous-Time Video Generation via Learning Motion Dynamics with Neural ODE
Kangyeol Kim*, Sunghyun Park*, Junsoo Lee, Joonseok Lee, Sookyung Kim, Jaegul Choo, and Edward Choi
British Machine Vision Conference (BMVC), 2021, Accepted.
[Paper] [Project]

Vid-ODE: Continuous-Time Video Generation with Neural Ordinary Differential Equation
Sunghyun Park*, Kangyeol Kim*, Junsoo Lee, Jaegul Choo, Joonseok Lee, Sookyung Kim, and Edward Choi
AAAI Conference on Artificial Intelligence (AAAI), 2021, Accepted (21% acceptance rate).
[Paper] [Project] [Code & Dataset] [PPT] Qualcomm Innovation Fellowship Korea 2021

Hurricane Nowcasting with Irregular Time-step using Neural-ODE and Video Prediction
Sunghyun Park*, Kangyeol Kim*, Sookyung Kim*, Joonseok Lee, Junsoo Lee, Jiwoo Lee, and Jaegul Choo
International Conference on Learning Representations Workshop (ICLRW), 2020, Accepted as Spotlight Presentation.
[Paper]

Unpaired Image Translation via Adaptive Convolution based Normalization
Wonwoong Cho*, Kangyeol Kim*, Eungyeup Kim, Hyunwoo Kim, and Jaegul Choo CoRR, abs/1911.13271, 2019
[Paper]


Work Experiences

Letsur Inc. Aug., 2021 - Current
Co-founder & AI Team Leader

NAVER Webtoon Corp. Jan., 2021 - Mar., 2021
AI Research, as a Research Internship

NAVER Corp. Jun., 2018 - Aug., 2018
AI Research,, as a Research Internship
- Worked at CLOVA Multimedia team
- Developed the neural networks that embed the trackable watermarks into webtoon contents against the copyright infringement of illegal attacks

Seoul National University Hospital. Jun., 2017 - Jan., 2018
Data Analyst, as a Research Internship
- Worked at Division of Clinical Bioinformatics
- Developed a cancer diagonisis model


Education

Korea Advanced Institute of Science and Technology. Mar., 2020 - Present
M.S./Ph.D. in Artificial Intelligence

Korea University. Mar., 2019 - Feb., 2020
M.S., Department of Computer Science
- Moved to KAIST due to affiliation change of my advisor, Prof. Choo

Korea University. Mar., 2013 - Feb., 2019
B.E., Department of Statistics
B.S., Department of Computer Science


Academic Activities

Reviewer. ICRL 2021, CVPR 2022 ECCV 2022