Perception & SLAM engineer building real-time 3D perception, calibration, and mapping systems for production robotics and mobility environments.
I work on bringing real-world objects and spaces into computers through captured data. My interests include 3D object detection, 3D reconstruction, Gaussian Splatting, world models, SLAM, and spatial perception systems that improve the quality of human life.
Interests
Workplace Projects
Real-time 3D Occupancy Network using Multiple Camera Sensors
Real-time occupancy perception using multiple camera sensors, validated on the Guro testbed.
Open videoMulti Camera 3D Object and Lane Detection
Real-time 3D object and lane detection using multiple camera sensors for embedded systems.
Multi Sensor Object Detection
Trained with nuScenes, Argoverse, ONCE, PandaSet, and internal datasets. The system runs within 60 ms from raw image acquisition to inference result.
Caliberator
Automated online LiDAR-to-IMU calibration software.
Discarto
Distributed mapping software for reconstructing 3D high-definition maps using LiDAR sensors, including selected mapping results from Sejong and the Guro testbed.
Publications
CIPER: A Unified Framework for Cross-view Image-retrieval and Pose-estimation
Yurim Jeon, Dongseong Seo, and Seung-Woo Seo.