Moongu Son

Gwangju Institute of Science and Technology

Papers

Total Citations

H-Index

About

Moongu Son is a computer vision researcher whose work focuses on advancing visual localization—a critical technology underpinning autonomous driving, robotics, and augmented reality. His most-cited paper, "Learning-based essential matrix estimation for visual localization" (2022), tackles the fundamental challenge of estimating camera pose from 2D images, proposing a novel learning-based approach that improves geometric accuracy in real-world environments. This contribution addresses a core bottleneck in vision-based navigation systems, where traditional methods often fail under challenging conditions like lighting changes or textureless scenes. While his citation count is still growing, Son’s research demonstrates strong potential to influence both academic theory and practical deployment in autonomous systems. His work sits at the intersection of deep learning and geometric computer vision, aiming to make localization more robust and efficient for applications ranging from robot mapping to mixed reality. As the demand for reliable visual positioning in self-driving cars and AR devices intensifies, Son’s contributions are poised to become increasingly relevant, marking him as an emerging voice in the field of visual localization and pose estimation.