Rasha Sheikh

Papers

Total Citations

H-Index

About

Rasha Sheikh is a robotics researcher whose work focuses on the intersection of computer vision, active learning, and agricultural automation. Her primary research areas include semantic segmentation for precision agriculture, visual simultaneous localization and mapping (SLAM) for humanoid robots, and efficient annotation strategies for deep learning models. Sheikh’s most impactful contribution is her work on active learning for semantic segmentation of crops and weeds, where she developed gradient and log-based methods to significantly reduce the human labeling effort required for training agricultural robots. This paper has garnered 39 citations, reflecting its practical importance in making supervised learning more scalable for real-world farming applications. She also introduced DLab, a novel binary descriptor that fuses RGB and depth information to improve robustness in visual SLAM systems for humanoid robots, addressing challenges in uniqueness and reproducibility. Through her research, Sheikh demonstrates a commitment to advancing autonomous systems that operate in complex, unstructured environments, from agricultural fields to humanoid locomotion. Her work is particularly relevant for researchers interested in reducing annotation costs and enhancing robot perception.