Moemen Abdel-Razek

Papers

Total Citations

259

H-Index

About

Moemen Abdel-Razek is a leading researcher in computer vision, with a primary focus on real-time semantic segmentation for autonomous driving and robotics. His major contributions center on developing computationally efficient neural network architectures that balance accuracy with speed—a critical requirement for real-world deployment. His seminal work, "A Comparative Study of Real-Time Semantic Segmentation for Autonomous Driving" (2018), has garnered 178 citations, establishing a foundational benchmark for the field. Abdel-Razek introduced ShuffleSeg, an innovative architecture leveraging grouped convolution and channel shuffling to dramatically reduce computational cost while maintaining segmentation quality (46 citations). He further advanced the field with RTSeg, providing comprehensive comparative analyses of efficient segmentation models (21 citations), and proposed a novel two-stream convolutional network that integrates appearance, motion, and geometry for real-time motion segmentation (14 citations). His research directly addresses the gap between high-accuracy models and the practical need for lightweight, real-time solutions in autonomous systems. Through his work, Abdel-Razek has significantly influenced the development of efficient perception systems, enabling safer and more responsive autonomous vehicles and robotic platforms.