Giljoo Nam
Papers
1
Total Citations
1
H-Index
1
About
Giljoo Nam is a researcher whose work centers on advancing dense visual SLAM (Simultaneous Localization and Mapping) systems through innovative geometric representations. His primary research areas include 3D computer vision, dense mapping, and efficient scene reconstruction for robotics and augmented reality. Nam's most notable contribution is the development of LRSLAM, a dense visual SLAM system that leverages low-rank representation of signed distance fields (SDFs). This approach significantly reduces the computational and memory overhead of traditional volumetric SDF methods, enabling real-time, high-fidelity 3D mapping on resource-constrained devices. By exploiting the low-rank structure inherent in many indoor environments, LRSLAM achieves robust camera tracking and dense reconstruction while maintaining a compact model. Although his work is early-stage, the LRSLAM framework represents a promising step toward practical, scalable dense SLAM. Nam’s research bridges the gap between theoretical geometric optimization and real-world deployment, offering a pathway for more efficient spatial AI systems. His contributions are particularly relevant for autonomous navigation, AR/VR, and robotics applications where accurate, lightweight mapping is critical.
Research Focus
Key Achievements
Top Papers
- 1