Kyungjune Baek
Download CV
Please contact me for the detailed information.
Interests
Experience
🎯
AI Research Engineer
Upstage
OCR / Synthetic Dataset
2023.03 ~ 2024.02
🎯
Visiting researcher
Clova AI Research (CLAIR), Naver Corp.
Image-to-Image translation
2019.09 ~ 2020.02
🎯
Development
SIT, Yonsei Univ.
Real time remote lecture room system construction
2018.09 ~ 2019.02
Education
🎓
Ph.D.
Yonsei Univ.
Integrated Technology
2018.03 ~ 2023.02
🎓
B.S.
Yonsei Univ.
Electrical & Electronic Eng. and Computer Science
2014.03 ~ 2018.02
Service
🎯
Reviewer
CVPR, IJCV, ICML-Workshop-SSL, Neurips-Workshop-SSL, ICPR, WACV

Publication

Logit Mixing Training for More Reliable and Accurate Prediction

Duhyeon Bang*, Kyungjune Baek*, Jiwoo Kim, Yunho Jeon, Jin-Hwa Kim, Jiwon Kim, Jongwuk Lee, Hyunjung Shim (* indicates an equal contribution)
IJCAI 2022
PDF

Proposing to utilize mixed logits for regularizing a classification model

International Joint Conference on Artificial Intelligence (IJCAI), 2022

Commonality in Natural Images Rescues GANs: Pretraining GANs with Generic and Privacy-free Synthetic Data

Kyungjune Baek, Hyunjung Shim
CVPR 2022
PDF

Proposing to utilize a synthetic pretraining dataset for low-shot generation of GANs

Computer Vision and Pattern Recognition (CVPR), 2022

Code
Rethinking the Truly Unsupervised Image-to-Image Translation

Kyungjune Baek, Yunjey Choi, Youngjung Uh, Jaejun Yoo, Hyunjung Shim
ICCV 2021
PDF

Unsupervised image-to-image translation even without domain labels.

International Conference on Computer Vision (ICCV), 2021

Code
GridMix: Strong Regularization Through Local Context Mapping

Kyungjune Baek*, Duhyeon Bang*, Hyunjung Shim (* indicates an equal contribution)
PR 2021
PDF

Regularization based on mixing samples at the level of patches.

Pattern Recognition (PR, IF=7.196), 2021

PsyNet: Self-Supervised Approach to Object Localization Using Point Symmetric Transformation

Kyungjune Baek*, Minhyun Lee*, Hyunjung Shim (* indicates an equal contribution)
AAAI 2020
PDF

Self-supervised approach to co-localization problem.

AAAI Conference on Artificial Intelligence (AAAI), 2020

Code
Learning from Better Supervision: Self-distillation for Learning with Noisy Labels

Kyungjune Baek*, Seungho Lee*, Hyunjung Shim (* indicates an equal contribution)
ICPR 2022
PDF

Improving a classifier with the knowledge distillation against noisy labels.

International Conference on Pattern Recognition (ICPR), 2022

Code
Editable Generative Adversarial Networks: Generating and Editing Faces Simultaneously

Kyungjune Baek, Duhyeon Bang, Hyunjung Shim
ACCV 2018
PDF

Attribute editing based on the inversion of the latent space of GANs.

Asian Conference on Computer Vision (ACCV), 2018
Oral presentation (Best paper Runner-up)

Code