Papers
1
Total Citations
5
H-Index
1
About
Gijae Lee is a researcher focused on advancing robotic perception, particularly in the domain of 6D object pose estimation for robotic grasping. His work addresses a critical bottleneck in deploying learning-based pose estimation methods in real-world environments: the heavy reliance on expensive ground-truth annotations. Lee’s most cited paper, "6D Object Pose Estimation Using a Particle Filter With Better Initialization" (2023), introduces a novel approach that improves pose estimation accuracy by refining initial hypotheses through particle filtering. This method offers a practical alternative to purely deep learning-based techniques, reducing the need for extensive labeled training data while maintaining robust performance. Although early in his career, Lee’s contributions are gaining traction, with his work accumulating citations that highlight its relevance to the robotics community. By bridging the gap between high-performance learning models and real-world applicability, Lee is helping to make robotic grasping more efficient and accessible, a key step toward fully autonomous manipulation systems. His research continues to inspire further exploration into hybrid probabilistic and learning-based methods.
Research Focus
Key Achievements
Top Papers
- 1