Senior Research Scientist — Vision & Robotics
Jaehyun Lee
Current focus: Hybrid factor-graph + neural-prior SLAM that survives tunnel-to-bright transitions without re-initialization.
Background
PhD Carnegie Mellon Robotics Institute 2019, advised by Sebastian Scherer at the AirLab; thesis on resilient LiDAR-inertial-visual SLAM in GPS-denied environments. Research Scientist at Skydio Autonomy (2019–2023) leading the visual-inertial state estimator on the X10. Joined Blueberry to fuse classical SLAM with neural scene priors.
Education
- PhD · Robotics — Carnegie Mellon University (2019), advisor: Sebastian Scherer
- MS · Robotics — Carnegie Mellon University (2015)
- BS · Mechanical and Aerospace Engineering — Seoul National University (2013)
Selected publications
- Lee et al., 'Resilient LiDAR-Inertial-Visual SLAM in GPS-Denied Subterranean Environments', JFR 2020
- Lee & Scherer, 'Factor-Graph Fusion of Heterogeneous Driving Sensors', RA-L 2021
- Lee et al., 'Neural-Prior SLAM for Urban Driving', CoRL 2023
Reading list
- Campos et al., 'ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM', T-RO 2021
- Shan et al., 'LVI-SAM: Tightly-coupled Lidar-Visual-Inertial Odometry via Smoothing and Mapping', ICRA 2021
- Qin et al., 'VINS-Mono / VINS-Fusion: A Robust and Versatile Monocular Visual-Inertial State Estimator', T-RO 2018
- Sucar et al., 'iMAP: Implicit Mapping and Positioning in Real-Time', ICCV 2021
- Zhu et al., 'NICE-SLAM: Neural Implicit Scalable Encoding for SLAM', CVPR 2022
Lead experiments
All experiments →No experiments led by this member yet.