Chanhyuk Jung

seagullcjung [at] gmail [dot] com
CV / Google Scholar / Github

About Me

I am a researcher at the CVPR Lab at Keimyung University in South Korea. My research interests include robot learning, imitation learning, visuomotor policies, and world models. Recently, I have been working on diffusion and flow-based approaches for behavior cloning.

Publications

I started with action recognition with graph neural networks and transitioned to motion genration. I'm currently working on imitation learning for manipulation.

* = equal contribution.

simDP: Sim-to-Real Transfer with Shared Action Spaces

Chanhyuk Jung, Jongbin Choi, Sungkeunn Yoo, and Byoung Chul Ko
Under Review
website

Shifted Flow Policy: Uncertainty-aware Time Reparameterization for Visuomotor Learning

Dasom Ahn*, Chanhyuk Jung*, Joonki Baek, Sungkeun Yoo, and Byoung Chul Ko
ICRA, 2026
Oral Presentation (Top 4%)
paper / website / code

Rolling Diffusion Policy for Robotic Action Prediction: Enhancing Efficiency and Temporal Awareness

Chanhyuk Jung*, Sangwon Kim*, Dasom Ahn, In-su Jang, Kwang-Ju Kim, Sungkeun Yoo, and Byoung Chul Ko
ICRA FMNS Workshop, 2025
paper / poster

Flow-Guided Policies: Overcoming Diffusion Limitations for Robust Robot Imitation Learning

Chanhyuk Jung*, Sangwon Kim*, Dasom Ahn, In-su Jang, Kwang-Ju Kim, Sungkeun Yoo, and Byoung Chul Ko
ICCV Workshop, 2025
paper

Lightweight Semantic-Guided Neural Networks Based on Single Head Attention for Action Recognition

Seon-Bin Kim, Chanhyuk Jung, Byeong-Il Kim, and Byoung Chul Ko
Sensors, 2022
paper

Education

Recently grduated with a B.S. in Computer Engineering.

Keimyung University (Mar. 2021 - Feb. 2026)
B.S. in Computer Engineering