Mostafa Gamal

Papers

Total Citations

H-Index

About

Mostafa Gamal is a computer vision researcher whose work centers on real-time semantic segmentation, with a particular focus on developing computationally efficient deep learning solutions for robotics and autonomous systems. His most recognized contribution, **ShuffleSeg** (2018), introduced a lightweight segmentation architecture leveraging grouped convolutions and channel shuffling to achieve high-speed inference without sacrificing accuracy — a critical balance for mobile and embedded platforms. The paper has garnered 46 citations, establishing it as a notable reference in efficient network design. Gamal further broadened the field through **RTSeg**, a comparative study (21 citations) that systematically evaluated real-time segmentation models, filling an important gap in research that had largely prioritized accuracy over computational feasibility. His work on multi-stream segmentation incorporating appearance, motion, and geometric cues (14 citations) demonstrated a sophisticated understanding of how diverse visual signals can be fused for robust scene understanding in autonomous driving and aerial imagery contexts. Collectively, Gamal's research reflects a consistent commitment to bridging the gap between theoretical segmentation advances and their practical deployment in resource-constrained, real-world environments — making him a valuable contributor to the applied computer vision community.