Real-Time LiDAR Gaussian Splatting SLAM
Seungjun Tak, Yewon Jeon, Jaeik Hwang, SukMin Hwang, Seongbo Ha, and Hyeonwoo Yu
In European Conference on Computer Vision (ECCV), 2026
SLAM · 3D Scene Representation · Robot Perception
Seongbo Ha (하성보)
Integrated M.S.–Ph.D. Program
Sungkyunkwan University
I am a graduate student in Intelligent Robotics, advised by Prof. Hyeonwoo Yu. Previously, I received my B.S. in Electrical Engineering with a minor in Future Mobility from the University of Ulsan.
My research interests lie in SLAM, 3D scene representation, and robot perception. I have mainly worked on real-time dense SLAM using expressive 3D scene representations. I am also interested in building SLAM systems that operate on real robotic platforms and remain practical beyond controlled laboratory settings. Recently, I've been exploring what information a robot should keep in its map, how that information should be represented, and how such representations can be coupled with SLAM to support downstream tasks.
Education
2024 — Present
Sungkyunkwan University
Integrated M.S.–Ph.D. Program in Intelligent Robotics
Lab of Artificial Intelligence & Robotics (LAIR) · Advisor: Prof. Hyeonwoo Yu
2018 — 2024
University of Ulsan
B.S. in Electrical Engineering, Minor in Future Mobility (GPA: 4.4 / 4.5)
Papers
2026
GSO-SLAM: Bidirectionally Coupled Gaussian Splatting and Direct Visual Odometry
Jiung Yeon*, Seongbo Ha*, and Hyeonwoo Yu (* equal contribution)
IEEE Robotics and Automation Letters (RA-L), 2026
LangGS-SLAM: Real-Time Language-Feature Gaussian Splatting SLAM
Seongbo Ha, Sibaek Lee, Kyeongsu Kang, Joonyeol Choi, Seungjun Tak, and Hyeonwoo Yu
arXiv preprint arXiv:2602.06991, 2026
2025
LEGO-SLAM: Language-Embedded Gaussian Optimization SLAM
Sibaek Lee, Seongbo Ha, Kyeongsu Kang, Joonyeol Choi, Seungjun Tak, and Hyeonwoo Yu
arXiv preprint arXiv:2511.16144, 2025
Spatial Coordinate Transformation for 3D Neural Implicit Mapping
Kyeongsu Kang, Seongbo Ha, Sibaek Lee, and Hyeonwoo Yu
IEEE Robotics and Automation Letters (RA-L), 2025
Bayesian NeRF: Quantifying Uncertainty With Volume Density for Neural Implicit Fields
Sibaek Lee, Kyeongsu Kang, Seongbo Ha, and Hyeonwoo Yu
IEEE Robotics and Automation Letters (RA-L), 2025
2024
RGBD GS-ICP SLAM
Seongbo Ha, Jiung Yeon, and Hyeonwoo Yu
In European Conference on Computer Vision (ECCV), 2024
A dense RGB-D SLAM system combining Gaussian Splatting with ICP for high-quality, real-time mapping.
2023
Vehicle Tracking System in Drone Imagery with YOLOv5 and Histogram
Jehwan Choi, Seongbo Ha, Youlkyeong Lee, and Kanghyun Jo
In the 62nd Annual Conference of the Society of Instrument and Control Engineers (SICE), 2023
2022
Optimal Path Driving Through Real-Time Obstacle Recognition and Map Creation in Top-View Images
Seongbo Ha, Seongmin Kim, and Kanghyun Jo
In KSIC, 2022
Automatic Parking System Using a Spatially Fixed Single Camera
Seongbo Ha and Kanghyun Jo
In ICROS, 2022
2021
Target Tracking and Mapping of Tracking Routes Using a Drone
Seongmin Kim, Seongbo Ha, Kyeongmin Yu, and Kanghyun Jo
In KSIC, 2021
Projects
Challenges
POSCO Industrial AI-Robot Challenge
First-prize-winning robotic system developed for an industrial AI challenge.
Industry Collaborations
AMR Localization for Automated Parking Using an RGB-D Camera
Early-stage project — development underway.
Real-Time 3D Reconstruction and Localization for Humanoid Robots and Low-Profile AMRs
Early-stage project — development underway.
Drone-Based Cart Monitoring System Using SLAM
Low-latency LiDAR-inertial SLAM and localization running onboard a drone with an embedded computer.
Physical AI–Based Integrated Autonomous Navigation for AMRs
Localization on 2D CAD drawings by aligning 3D SLAM maps to the CAD coordinate frame.
Real-Time High-Resolution 3D RGB-D SLAM
Real-time RGB-D SLAM on an embedded platform, building high-fidelity 3D Gaussian Splatting maps within minutes.
3D LiDAR Mapping & Robust Localization
A framework for robust localization in environments that have changed since mapping.
Experience
Research Projects
2025 — 2030
AI Semiconductor Innovation Center
Sungkyunkwan University · IITP grant funded by MSIT
2024 — 2027
Multi-Modal 3D Perception & Task Planning
Robotic manipulation of thin, non-rigid objects for battery-pack recycling · MOTIE
2024 — 2028
Unmanned Autonomous Operation Technology
Visualization for field deployment and 6-axis rotational angle sensing · MOTIE
2024 — 2027
Platform-Agnostic Embodied AI
Spatial and task generalization framework · NRF
Awards & Honors
POSCO Industrial AI-Robot Challenge 2025
First Prize
Institute of Embedded Engineering of Korea (IEMEK)
Invited Talk
Yangyoung Foundation
Full Scholarship
KSIC Fall Conference 2022, Korean Society of Industrial Convergence
Outstanding Paper Presentation Award
University of Ulsan Capstone Design Competition
First Prize
U-Challenge Festival (BTS Practical Problem Research), UNIST & University of Ulsan
Second Prize
KSIC Fall Conference 2021, Korean Society of Industrial Convergence
Outstanding Paper Presentation Award