Munirah Rosbi

Papers

Total Citations

H-Index

About

Munirah Rosbi’s research focuses on the intersection of computer vision and mobile robotics, with a particular emphasis on robust human detection in dynamic, real-world environments. Her major contributions center on developing novel sensor fusion techniques to address the persistent challenge of occlusion—a critical hurdle for autonomous systems navigating crowded spaces. Rosbi pioneered the integration of thermal and depth imaging, using thermal data to initially localize human upper-bodies and depth maps to refine bounding box coordinates and resolve visual overlaps. This approach, detailed in her most-cited work, “Fusion of thermal and depth images for occlusion handling for human detection from mobile robot” (2015, 9 citations), demonstrated significant improvements over conventional RGB-based methods. She further advanced this framework in “Improved occlusion handling for human detection from mobile robot” (2015, 3 citations), where she refined the fusion pipeline to enhance detection accuracy under complex occlusion scenarios. While her citation counts reflect the niche, applied nature of her work, Rosbi’s contributions are notable for their practical impact on mobile robot perception, offering a scalable solution for service robots, surveillance, and assistive technologies. Her research remains a valuable reference for engineers developing vision systems that must operate reliably in cluttered, unpredictable settings.