Marah Halawa

Papers

Total Citations

H-Index

About

Marah Halawa is a rising researcher in autonomous systems and human-robot interaction, with a focused expertise in trajectory prediction—a critical component for safe navigation in environments like autonomous driving. Her most cited work, "Action-Based Contrastive Learning for Trajectory Prediction" (2022), has garnered 48 citations, establishing her as a notable contributor to this field. In this research, Halawa tackles the challenging problem of forecasting pedestrian trajectories from a first-person, moving-camera perspective—a scenario far more complex than static camera setups. Her key innovation lies in introducing an action-based contrastive learning framework, which leverages pedestrian actions (e.g., walking, stopping) to improve prediction accuracy by learning more discriminative feature representations. This approach not only enhances the model's ability to anticipate future paths in dynamic, ego-centric views but also addresses the scarcity of labeled trajectory data. Halawa's work is particularly impactful for real-world applications like autonomous vehicles and collaborative robots, where understanding human intent is paramount. Her contributions represent a meaningful step toward more robust and context-aware prediction systems, marking her as a promising voice in the next generation of robotics and AI researchers.