Computer Vision · Generative AI

Taewoo Kim

Postdoctoral Researcher at INSAIT

Ph.D., KAIST · Computer Vision Researcher

Working on visual generation and low-level vision, including image and video generation, editing, and restoration.

Email ↗ Google Scholar ↗ GitHub ↗ LinkedIn ↗

About Me

I am a Postdoctoral Researcher at INSAIT, focusing on controllable video generation. I received my Ph.D. from KAIST under the supervision of Prof. Kuk-Jin Yoon, where I worked on computer vision, machine learning, and computational imaging. Before joining INSAIT, I worked in the camera algorithm team at Qualcomm, developing ML-based computational imaging and camera systems.

My work focuses on generative AI (image and video generation), computational photography, and multimodal data fusion. I welcome collaborations and discussions—please feel free to reach out.

Publications (Google Scholar)

ICCV 2025

Event-guided Unified Framework for Low-light Video Enhancement, Frame Interpolation, and Deblurring

Taewoo Kim , Kuk-Jin Yoon

NeurIPS 2024

A Benchmark Dataset for Event-Guided Human Pose Estimation and Tracking in Extreme Conditions

Hoonhee Cho*, Taewoo Kim* , Yuhwan Jeong, Kuk-Jin Yoon

ECCV 2024

Towards Real-world Event-guided Low-light Video Enhancement and Deblurring

Taewoo Kim , Jaeseok Jeong, Hoonhee Cho, Yuhwan Jeong, Kuk-Jin Yoon

ECCV 2024

CMTA: Cross-Modal Temporal Alignment for Event-guided Video Deblurring

Taewoo Kim* , Hoonhee Cho*, Kuk-Jin Yoon

CVPR 2024

Frequency-aware Event-based Video Deblurring for Real-world Motion Blur

Taewoo Kim* , Hoonhee Cho*, Kuk-Jin Yoon

CVPR 2024

TTA-EVF: Test-Time Adaptation for Event-based Video Frame Interpolation via Reliable Pixel and Sample Estimation

Hoonhee Cho, Taewoo Kim , Yuhwan Jeong, Kuk-Jin Yoon

ICCV 2023

Non-coaxial Event-Guided Motion Deblurring with Spatial Alignment

Hoonhee Cho, Yuhwan Jeong, Taewoo Kim , Kuk-Jin Yoon

CVPR 2023 ★ Highlight Paper (Top 2.5%)

Event-based Video Frame Interpolation with Cross-Modal Asymmetric Bidirectional Motion Fields

Taewoo Kim , Yujeong Chae, Hyun-Kurl Jang, Kuk-Jin Yoon

ECCV 2022 ★ Oral Presentation (Top 2.7%)

Event-guided Deblurring of Unknown Exposure Time Videos

Taewoo Kim , Jeongmin Lee, Wang Lin, Kuk-Jin Yoon

RA-L + IROS 2020 ★ Oral Presentation

Loop-Net: Joint Unsupervised Disparity and Optical Flow Estimation of Stereo Videos with Spatiotemporal Loop Consistency

Taewoo Kim , Kwonyoung Ryu, Kyeongseob Song, Kuk-Jin Yoon

Education

Ph.D. in Mechanical Engineering, KAIST

Thesis: Video Enhancement with Event Cameras

Advisor: Prof. Kuk-Jin Yoon

Feb. 2025

M.S. in Robotics, KAIST

Thesis: Joint Unsupervised Disparity and Optical Flow Estimation

Advisor: Prof. Kuk-Jin Yoon

Feb. 2020

B.S. in Mechanical Engineering, Yonsei University

Feb. 2017

Academic Service

Conferences: CVPR, ICCV, ECCV, WACV, NeurIPS, SIGGRAPH, ICLR, AAAI.
Journals: IEEE TPAMI, IJCV, Neural Networks, RA-L, IJAT