Hyunwoo Ha

Hyunwoo Ha's profile image

I am a Ph.D. student in Electrical Engineering at POSTECH, advised by Prof. Tae-Hyun Oh. I work with the Advanced Machine Intelligence Lab at KAIST.

My research interests lie in physical AI and visual intelligence, including motion understanding, robust perception, 3D reconstruction, and multimodal learning.

hyunwooha@postech.ac.kr | CV | Google Scholar | GitHub | LinkedIn

News

07/2026 Our paper Fast-MM was published in TMLR.
2026 I received the Best Paper Award at IPIU 2026.
2026 I joined KIST as a Student Researcher, working with Prof. In So Kweon.
2026 Our papers LA-CD and DarkQA were published in IEEE RA-L.
11/2024 Our paper MeTTA received the Best Poster Award at BMVC 2024.
07/2024 Our paper Learning-based Axial Motion Magnification was accepted to ECCV 2024.
06/2023 Our paper Sound2Scene was presented at CVPR 2023.

Education

POSTECH, Electrical Engineering2023 - Present
Ph.D. Student, advised by Prof. Tae-Hyun Oh.
POSTECH, Electrical Engineering2021 - 2023
M.S., advised by Prof. Tae-Hyun Oh.
POSTECH, Electrical Engineering2016 - 2021
B.S.

Research Experiences

Korea Institute of Science and Technology (KIST), Seoul, Korea2026
Student Researcher, working with Prof. In So Kweon.
RIST, Pohang, Korea2021 - 2023
Student Researcher.
AMI Lab, POSTECH, Pohang, Korea2020 - 2021
Undergraduate Research Student, advised by Prof. Tae-Hyun Oh.
LG Display, South Korea2019
Undergraduate Intern.

Publications

Fast-MM speed and quality comparison

Revisiting Learning-based Video Motion Magnification for Real-time Processing
Hyunwoo Ha*, Oh Hyun-Bin*, Kim Jun-Seong, Kwon Byung-Ki, Kim Sung-Bin, Linh-Tam Tran, Ji-Yun Kim, Sung-Ho Bae, Tae-Hyun Oh
Transactions on Machine Learning Research (TMLR), 2026
Project page | Paper | OpenReview | Code

LA-CD overview

LiDAR-Anchored Collaborative Distillation for Robust 2D Representations
Wonjun Jo, Hyunwoo Ha, Kim Ji-Yeon, Hawook Jeong, Tae-Hyun Oh
IEEE Robotics and Automation Letters (RA-L), 2026
Project page | Paper

DarkQA benchmark examples

DarkQA: Benchmarking Vision-Language Models on Visual-Primitive Question Answering in Low-Light Indoor Scenes
Yohan Park, Hyunwoo Ha, Wonjun Jo, Tae-Hyun Oh
IEEE Robotics and Automation Letters (RA-L), 2026
Project page | Paper

Policy drift mitigation overview

When Fine-Tuning Drifts: Mitigating Policy Drift with Base Action-Field Regularization
Wonjun Jo, Nam Hyeon-Woo, Yohan Park, Hyunwoo Ha, Tae-Hyun Oh
Under review

Collaborative distillation overview

Self-Supervised Collaborative Distillation: Enhancing Lighting Robustness and 3D Awareness
Wonjun Jo, Hyunwoo Ha, Kim Ji-Yeon, Hawook Jeong, Tae-Hyun Oh
ICCV Workshop on Wild3D, 2025
Paper

MeTTA reconstruction results

MeTTA: Single-View to 3D Textured Mesh Reconstruction with Test-Time Adaptation
Kim Yu-Ji, Hyunwoo Ha, Kim Youwang, Jaeheung Surh, Hyowon Ha†, Tae-Hyun Oh†
BMVC, 2024 (Best Poster Award)
Project page | Paper | Code | Video

Axial motion magnification architecture

Learning-based Axial Motion Magnification
Kwon Byung-Ki, Oh Hyun-Bin, Kim Jun-Seong, Hyunwoo Ha, Tae-Hyun Oh
ECCV, 2024
Project page | Paper | Code

Sound2Scene results

Sound to Visual Scene Generation by Audio-to-Visual Latent Alignment
Kim Sung-Bin, Arda Senocak, Hyunwoo Ha, Andrew Owens, Tae-Hyun Oh
CVPR, 2023
Project page | Paper | Code | Video

Mask-KLT method framework

Mask-KLT: Sub-pixel Accurate Directional Motion Estimation by Stripe Masking
Hyunwoo Ha, Tae-Hyun Oh
VCIP, 2023
Paper

* Equal contribution. † Corresponding author.

Awards and Honors

Academic Services