Taeyeop Lee

Korea Advanced Institute of Science and Technology, Kyung Hee University

Papers

Total Citations

H-Index

About

Taeyeop Lee is a robotics researcher whose work bridges perception, manipulation, and 3D scene understanding. His key research areas include robotic grasping in cluttered environments, 3D object representation, and dynamic system modeling. Lee’s major contributions center on advancing robust perception for real-world robotics. His work on GraspClutter6D introduced a large-scale, real-world dataset designed to address the limitations of existing benchmarks, which often feature simplistic scenes with light occlusion. By providing diverse, heavily cluttered scenarios, this dataset enables deep learning methods to generalize to practical, unstructured environments—a critical step toward deploying robots in warehouses, homes, and industrial settings. Additionally, Lee’s research on one-shot neural fields offers a compact, unified scene representation that encodes both geometry and appearance, supporting tasks like novel view synthesis and 3D reconstruction from minimal input. This work has implications for efficient robotic perception and object interaction. With each of his most-cited papers garnering 2 citations, Lee’s contributions are gaining recognition for their focus on real-world applicability and foundational challenges in robotics. His work on the coupled three-link manipulator further demonstrates his breadth, spanning dynamic analysis and parameter estimation for flexible systems.