Theodora Retzepi

Papers

Total Citations

H-Index

About

Theodora Retzepi’s research lies at the intersection of robotics, computer vision, and knowledge representation, with a particular focus on enabling autonomous object manipulation through semantic understanding. Her most-cited work, “Ontology-based 3D pose estimation for autonomous object manipulation” (2012), introduces a novel framework that bridges the gap between high-level reasoning and low-level robotic control. By employing an ontology-based approach during both training and testing phases, Retzepi’s system allows a robotic gripper to infer the 3D pose of objects not just from geometric features, but from their semantic context—a significant step toward more intelligent, adaptable manipulation in unstructured environments. While her citation count (2) reflects the niche, foundational nature of this early work, its conceptual contribution is noteworthy: it anticipates the current trend toward integrating symbolic AI with robotic perception. Retzepi’s research is particularly valuable for students and researchers exploring how ontologies can provide the structured knowledge needed for robots to generalize beyond pre-programmed tasks, making her a thoughtful contributor to the evolving field of cognitive robotics.