John S. Loomis

Papers

Total Citations

H-Index

About

John S. Loomis is a leading researcher in the fields of autonomous navigation, embedded robotics, and real-time 3D perception. His work focuses on enabling efficient, high-performance visual SLAM (Simultaneous Localization and Mapping) and deep learning-based odometry for mobile robots and autonomous vehicles, particularly under the constraints of embedded systems. Loomis’s most cited paper, “Evaluating the Power Efficiency of Visual SLAM on Embedded GPU Systems” (2019, 21 citations), provides a critical benchmark for deploying ORB-SLAM2 on low-power platforms, directly addressing the battery limitations of mobile robots. He further advanced autonomous driving safety with his 2023 work on optimized deep learning for LiDAR and visual odometry fusion, tackling pose estimation in complex dynamic environments. Additionally, his 2018 research on real-time 3D scene reconstruction and localization with surface optimization demonstrates a novel approach to dense point cloud modeling using rotation-invariant feature matching and loop closure. Loomis’s contributions are essential for engineers and researchers developing robust, energy-efficient perception systems for real-world autonomous platforms.