Papers
1
Total Citations
18
H-Index
1
About
Hongseok Lee is a rising star in the field of energy-efficient spatial computing hardware, with a focus on real-time simultaneous localization and mapping (SLAM) for mobile and augmented reality (AR) devices. His most-cited work, "20.8 Space-Mate: A 303.5mW Real-Time Sparse Mixture-of-Experts-Based NeRF-SLAM Processor for Mobile Spatial Computing" (2024), has already garnered 18 citations, demonstrating its immediate impact. This paper introduces a groundbreaking processor that leverages a sparse mixture-of-experts architecture to enable Neural Radiance Field (NeRF)-based SLAM at ultra-low power—just 303.5mW—making high-fidelity 3D scene reconstruction and localization feasible for power-constrained platforms like autonomous robots and AR glasses. Lee’s major contribution lies in bridging the gap between computationally intensive NeRF algorithms and real-time mobile deployment, addressing critical bottlenecks in memory bandwidth and energy efficiency. His work is pivotal for advancing cyber-physical interaction, where accurate user position and environmental geometry are essential. As a researcher at the forefront of hardware-software co-design, Lee is shaping the future of mobile spatial computing, with innovations that promise to unlock seamless, immersive experiences in next-generation devices.
Research Focus
Key Achievements
Top Papers
- 1