Sara Eikerdawy

Papers

Total Citations

H-Index

About

Sara Eikerdawy is a computer vision researcher whose work focuses on real-time perception for autonomous systems, particularly in the domains of robotics, autonomous driving, and aerial monitoring. Her most cited paper, "Real-Time Segmentation with Appearance, Motion and Geometry" (2018, 14 citations), introduces a novel two-stream convolutional network that fuses appearance, motion, and geometric cues to achieve efficient, real-time motion segmentation. This contribution is critical for enabling safe navigation in dynamic environments, such as traffic monitoring from unmanned aerial vehicles and driving assistance systems. Eikerdawy’s research addresses the pressing need for high-speed, accurate scene understanding in resource-constrained platforms, bridging the gap between deep learning performance and real-world deployment. Her work has been recognized for its practical impact on autonomous vehicle perception pipelines, and she continues to advance the field by developing lightweight architectures that balance computational efficiency with segmentation accuracy. For students and researchers exploring real-time computer vision, Eikerdawy’s contributions offer a foundational approach to integrating multimodal information for robust, on-the-fly scene analysis.