Hojun Son

Papers

Total Citations

H-Index

About

Hojun Son is a researcher focused on advancing efficient computer vision for resource-constrained systems, with a primary emphasis on semantic segmentation for low-compute embedded devices. His work addresses a critical challenge: enabling mobile robots, drones, and wearable systems to perform real-time semantic scene understanding despite severe limitations in power and computational capacity. Son’s most cited paper (2023, 8 citations) introduces a lightweight semantic segmentation network designed specifically for low-compute platforms, demonstrating how onboard cameras can provide rich semantic information to improve robot navigation without requiring high-end hardware. His earlier foundational work (2022, 3 citations) further explores deploying deep convolutional neural networks on embedded devices, highlighting the growing importance of efficient AI for autonomous systems. By optimizing deep learning models for practical, real-world deployment, Son contributes directly to making intelligent, context-aware robotics more accessible and functional in field applications. His research is particularly relevant for students and engineers working on edge AI, autonomous navigation, and embedded vision systems, where balancing accuracy with computational efficiency is paramount.