Papers
9
Total Citations
42
H-Index
4
About
Nathan Crombez is a robotics researcher whose work sits at the intersection of computer vision, autonomous navigation, and service robotics. His primary contributions center on visual localization and control for mobile robots, with a particular emphasis on leveraging dense 3D environment models and photometric information to improve pose estimation accuracy. Crombez has pioneered the use of dense point clouds as environment models for visual localization, a foundational approach that has garnered significant attention. His research extends into photometric visual servoing, where he has advanced the field by extending techniques to full spherical cameras and nonholonomic mobile robots, integrating deep reinforcement learning for more robust control. In the domain of service robotics, he has developed quantifiable strategies for tidy-up tasks, addressing the complex challenge of object selection and manipulation in cluttered environments. More recently, Crombez has explored hybrid AI approaches for panoptic segmentation, combining deep learning with prior spatial knowledge. With over 40 citations across his most-cited works, his research demonstrates growing impact, particularly in the areas of 3D model-based tracking and multi-robot systems. His work on direct 3D model-based tracking in omnidirectional images and multimodal 2D-to-3D registration showcases his commitment to solving real-world robotic perception challenges.
Research Focus
Key Achievements
Top Papers
- 1
- 2A Quantifiable Stratification Strategy for Tidy-up in Service Robotics8 citations · 2021
- 3
- 4Dual-Hemispherical Photometric Visual Servoing5 citations · 2024
- 5
- 6
- 7
- 8Learning Decentralized Multi-Robot PointGoal Navigation2 citations · 2025
- 9