Fernando Amodeo

Papers

Total Citations

H-Index

About

Fernando Amodeo is a robotics researcher whose work sits at the intersection of computer vision, knowledge representation, and human-robot interaction. His primary research areas include scene graph generation, ontology-guided learning, and people detection using non-vision sensors. Amodeo’s most notable contribution is **OG-SGG (Ontology-Guided Scene Graph Generation)**, a framework that integrates formal ontological knowledge into the scene graph generation pipeline. This approach improves the semantic richness and transferability of visual representations, particularly for telepresence robotics and Visual Question Answering (VQA). With 11 citations since 2022, this work has been recognized for bridging the gap between symbolic AI and deep learning in robotics. More recently, Amodeo introduced **FROG**, a novel dataset designed for detecting people using knee-high 2D range finders—a sensor modality often overlooked in favor of cameras. This dataset addresses a critical gap in safe human-aware navigation for mobile robots, especially in low-light or privacy-sensitive environments. By expanding the perceptual toolkit for robots, Amodeo’s work contributes to more robust, context-aware autonomous systems.