Taebok Lee

I am an undergraduate student at Kookmin University, South Korea, and a research intern at the 3D Vision & Robotics Lab at UNIST, advised by Kyungdon Joo. My research focuses on computer vision, 3D vision, SLAM and robotics.

Email / CV / Google Scholar / Github / LinkedIn

News

2026-06	Our paper got Best Talk Award at ICRA Workshop 2026 on "From Sea to Space: Advancing Perception in Harsh Domains"
2026-05	Our paper has been accepted to ICROS 2026.
2026-04	Our paper has been accepted as spotlight at the ICRA 2026 Workshop on “From Sea to Space: Advancing Perception in Harsh Domains.”
2026-02	Our paper has been accepted to Korea Artificial Intelligence Conference 2026.
2025-12	I have joined the 3D Vision & Robotics Lab at UNIST as a research intern.
2025-09	Received the Outstanding Paper Award at the Korea Artificial Intelligence Conference 2025.
2025-08	Our paper has been accepted to Korea Artificial Intelligence Conference 2025.

Research

* indicates equal contribution, † indicates corresponding author

International Publications

Proprioceptive Reconstruction (blinded title)
Yujin Park, Seungoh Han, Taebok Lee, Chanwoo Lee, Hyungyu Park, Kyungdon Joo†

Under Review

Domestic Publications

Enhancing Foothold Planning for Locomotion on Curved Discrete Terrain
Taebok Lee*, Donghyeon Soon*, Kyungdon Joo†
ICROS, 2026

This work proposes a heightmap-based implicit path guidance method to mitigate path drift in foothold planning within a convex MPC-based locomotion framework.

Grid-based Local Homography Model for Panorama Stitching under Parallax
Donghyeon Soon, Minseok Oh, Taebok Lee
Korea Artificial Intelligence Conference, 2026
[paper]

This work analyzes the limitations of global homography under parallax and introduces a grid-based local homography model for robust panorama stitching.

Enhancing Underwater Object Detection via Conditional Diffusion Model based Preprocessing and Augmentation
Donghyeon Soon*, Taebok Lee*
Korea Artificial Intelligence Conference, 2025
[paper]
🏆 Outstanding Paper Award

We propose a conditional diffusion-based preprocessing and augmentation pipeline to improve underwater object detection under challenging visibility conditions.

Education

Kookmin University
Bachelor of Science
Major: Electrical Engineering

Mar 2021 – Feb 2027

Experience

	Research Intern, 3D Vision & Robotics Lab, UNIST Advisor: Kyungdon Joo	Jan 2026 – Present
	Research Administrative Assistant, Integrated Systems Lab, Kookmin University Advisor: KyeongSik Min	May 2024 – Dec 2025

Template from this website.