Stamatis Alexandropoulos

I am a 1st year Ph.D student in Computer Science at Princeton University, working with Prof. Jia Deng as part of the Princeton Vision & Learning Lab. Previously, I completed with highest honors (Valedictorian) my Diploma M.Eng Degree at the School of Electrical and Computer Engineering of the National Technical University of Athens, where I worked with Prof. Petros Maragos (CVSP Group) and Dr. Christos Sakaridis (Computer Vision Lab, ETH Zurich).

My research interests lie in the interface of Machine Learning, Computer Vision, Computer Graphics as well as Robotics. Consequently, I would like to delve into the above areas and develop artificially intelligent systems able to make an impact on humanity and reason visual world. My research is supported by the Onassis Foundation Scholarship and the Princeton University Fellowship.

Email  /  Github

profile photo


• [Oct 2023] We got a paper accepted in IEEE/CVF WACV 2024: "OVeNet: Offset Vector Network for Semantic Segmentation"

• [Sep 2023] I have been awarded the Onassis Foundation Scholarship for my graduate studies!

• [Aug 2023] I have been awarded the Princeton Stanley J. Seeger Hellenic Studies Award

• [Aug 2023] I am starting my Ph.D at Princeton in Fall 2023, advised by Professor Jia Deng!

• [June 2023] I have been awarded the Tzafestas, Chrisovergi and Kondouli Honorary Awards for graduating 1st in class among all students of the Department of ECE (Valedictorian) and across all Departments of National Technical University of Athens, respectively.

• [June 2023] I have been awarded the Kontaxis Honors for graduating 1st in Computer Science major

clean-usnob OVeNet: Offset Vector Network for Semantic Segmentation
Stamatis Alexandropoulos, Christos Sakaridis, Petros Maragos,

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024
[Pdf] [Code]

Based on knowledge about the high regularity of real-world scenes, we propose a method for improving class predictions by learning to selectively exploit information from neighboring pixels.

This guy has an awesome website.