Gwangtae Park
Papers
2
Total Citations
34
H-Index
2
About
Gwangtae Park is a leading researcher in energy-efficient hardware design for mobile spatial computing, with a focus on real-time 3D perception and augmented reality (AR) systems. His work bridges the gap between advanced deep learning algorithms and practical mobile deployment, achieving remarkable efficiency without sacrificing performance. Park's major contributions include the development of the "Space-Mate" processor, a 303.5mW real-time sparse mixture-of-experts-based NeRF-SLAM processor that enables accurate 3D scene reconstruction and localization on power-constrained devices. This work, published in 2024, has already garnered 18 citations for its innovative approach to neural rendering and simultaneous localization and mapping (SLAM). Additionally, his "DSPU" (Depth Signal Processing Unit) demonstrated a 281.6mW real-time solution for dense RGB-D data acquisition and 3D bounding box extraction, achieving over 30fps performance critical for autonomous robots and AR glasses. With 16 citations, this work showcases his ability to optimize complex computer vision pipelines for mobile platforms. Park's research directly enables next-generation spatial computing applications, from autonomous navigation to immersive AR experiences, making him a key figure in the push toward truly mobile, intelligent systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2