MARIUS LEORDEANU

Introduction

I am a Full Professor of Computer Science at the University Politehnica of Bucharest (UPB), a Senior Researcher at the Norwegian Research Centre (NORCE), and a researcher at the Simion Stoilow Institute of Mathematics of the Romanian Academy (IMAR). My broader scientific interests center on understanding the nature of intelligence, consciousness, and life, while my research focuses on computer vision, machine learning, robotics, and artificial intelligence.

At UPB, I introduced the Computer Vision and Robotics courses in 2014. At IMAR, I founded the Computer Vision Reading Group, which has met weekly since 2016. I received my Ph.D. in 2009 from the Robotics Institute at Carnegie Mellon University Robotics Institute and earned a Bachelor's degree in Computer Science and Mathematics in 2003 from Hunter College.

My work on unsupervised learning for graph matching was recognized with the Grigore Moisil Prize in 2014, one of the highest distinctions awarded by the Romanian Academy in mathematics. In 2024, I received the Grigore Constantin Moisil Prize, the highest honor awarded to a Romanian computer scientist or mathematician at the Romanian Research Gala.

I currently lead research projects across a broad spectrum of topics in computer vision, machine learning, and deep learning. While diverse in scope, these research directions are connected by a common objective: enabling coherent semantic understanding of the visual world across space and time, together with semantic visual control, navigation, and human–AI interaction for intelligent robotic systems operating in real time and with minimal human supervision.

My research encompasses unsupervised and self-supervised learning, optimization methods for graphs and deep neural networks, object tracking, semantic segmentation, Earth observation, video understanding and activity recognition, as well as multimodal learning that integrates vision and language.

I am particularly drawn to research that combines strong theoretical foundations with real-world impact. This perspective has led me to develop intelligent vision and learning systems for autonomous drones, self-driving vehicles, medical applications, and industrial processes, including solutions for the wood industry. Many of these projects are carried out in collaboration with leading technology companies and international research organizations.

A complete list of publications is available through my  Google Scholar and on my DBLP profiles.