Alexandros Delitzas

Papers

Total Citations

H-Index

About

Alexandros Delitzas is an emerging researcher at the forefront of 3D scene understanding and spatial artificial intelligence. His work focuses on bridging the gap between geometric perception and semantic reasoning in real-world environments, with a particular emphasis on making 3D representations more useful for autonomous agents and robotic systems. His most recognized contribution, "Open-Vocabulary Functional 3D Scene Graphs for Real-World Indoor Spaces" (2025, 8 citations), introduces a novel paradigm that moves beyond traditional spatial scene graphs to capture the functional relationships between objects and interactive elements within indoor environments. By leveraging posed RGB-D imagery, Delitzas advances the field's ability to reason not just about where objects are, but how they can be used and interacted with — a critical step toward embodied AI and intelligent robotics. His integration of open-vocabulary approaches signals a commitment to generalizable, language-grounded perception systems that can operate without rigid category constraints. Though early in his career, Delitzas is establishing himself as a thoughtful innovator whose research carries meaningful implications for scene understanding, human-robot interaction, and the next generation of spatially aware AI systems.