Visual SLAM Enhancement
Integrated SSC-based ANMS into the ORB keypoint selection stage of ORB-SLAM3 to promote spatially balanced feature selection and improve tracking stability.
•
1 min read
Research Profile
I’m currently pursuing an M.S. in Artificial Intelligence at UNIST and working in the 3D Vision & Robotics Lab advised by Prof. Kyungdon Joo.
I received B.S. in Robotics from Hanyang University ERICA, South Korea in 2024.
My research focuses on sensor calibration, visual SLAM, and event-based vision for robotics.
Education
M.S. in Artificial Intelligence
2025-03-02
Ulsan National Institute of Science and Technology (UNIST)
B.S. in Robotics
2019-03-02
2024-08-31
Hanyang University ERICA
Interests
Sensor Calibration
Visual SLAM
Event-based Robot Vision
3D Vision
Publications
From Corners to Fiducial Tags: Revisiting Checkerboard Calibration for Event Cameras
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026
CVPR Highlight
We propose the first checkerboard-based calibration method for event cameras with a mathematical analysis of the event rate on checkerboard corner.
Table Robot System for Complete Restaurant Hall Automation
Institute of Control, Robotics and Systems (ICROS), 2024
Undergraduate Student Paper Award
This paper presents a table robot system for restaurant hall automation that combines ceiling-camera perception with dynamic path planning to improve service speed and reduce reliance on human servers.
Experience
Graduate Research Intern
Ulsan National Institute of Science and Technology (UNIST)
I conducted the industry-academia project “Development and Research of a Multi-Camera-Based SLAM System” at the 3D Vision & Robotics Lab, supported by CLOBOT.
Education
M.S. in Artificial Intelligence
Ulsan National Institute of Science and Technology (UNIST)
B.S. in Robotics
Hanyang University ERICA
Minor in Artificial Intelligence. Early graduation. 2 years of military service. GPA: 4.18/4.5.
Recent Projects
Table Robot System for Complete Restaurant Hall Automation
Developed a hybrid localization framework combining robot-internal EKF estimates (IMU, wheel encoder, and LiDAR) with ceiling-camera observations.
•
1 min read