Ismail Geles

I am a PhD Student at the Robotics and Perception Group at the University of Zurich. I am also an assiociated doctoral student at the ETH AI Center.

In 2023, I completed my Master's at Graz University of Technology in Information and Computer Engineering, focusing on AI, Computer Vision and Robotics.

For my Master's thesis I was working at the Autonomous Systems Lab (ASL), Swiss Federal Institute of Technology ETH Zurich on the topic of aerial manipulation through Reinforcement Learning.

Before that, I was a graduate exchange student at the University of Waterloo in Canada. In 2021, I received my Bachelor of Science in Electrical Engineering from Graz University of Technology.

Email  /  CV  /  Github  /  LinkedIn  /  Google Scholar

I am interested in Reinforcement Learning and Computer Vision with applications to intelligent robotic systems.

• PhD Student, University of Zurich
• Robotics Master's Thesis, ETH Zurich
• Graduate Exchange Studies, University of Waterloo
• MSc in Information and Computer Engineering, Graz University of Technology
• BSc in Electrical Engineering, Graz University of Technology

• arXiv preprint
• ”Dream to Fly: Model-Based Reinforcement Learning for Vision-Based Drone Flight”
• Angel Romero*, Ashwin Shenai*, Ismail Geles, Elie Aljalbout, Davide Scaramuzza
• RAS’25
• ”GAFAR revisited—Exploring the limits of point cloud registration on sparse subsets”
• Ludwig Mohr, Ismail Geles, Friedrich Fraundorfer
• RAL’24
• ”Multi-Task Reinforcement Learning for Quadrotors”
• Jiaxu Xing, Ismail Geles, Yunlong Song, Elie Aljalbout, Davide Scaramuzza
• RSS’24
• ”Demonstrating Agile Flight from Pixels without State Estimation”
• Ismail Geles*, Leonard Bauersfeld*, Angel Romero, Jiaxu Xing, Davide Scaramuzza
• IROS’23
• ”Learning to Open Doors with Physically Interacting Aerial Manipulators”
• Eugenio Cuniato, Ismail Geles, Weixuan Zhang, Olov Andersson, Marco Tognon, Roland Siegwart
• ECMR’23
• ”GAFAR: Graph‑Attention Feature‑Augmentation for Registration
  A Fast and Light‑weight Point Set Registration Algorithm”
• Ludwig Mohr, Ismail Geles, Friedrich Fraundorfer

PDF / code (not public) / slides

PDF (not public) / code (not public)

Submission for the Machine Learning Reproducibility Challenge 2021, reproducing results from the paper Boosting Monocular Depth Estimation Models to High-Resolution via Content-Adaptive Multi-Resolution Merging.

Literature review on Reinforcement Learning in general and re-implementing Hindsight Experience Replay in Pytorch and OpenAI Gym robotic arm environments.

• 2022: Research-Abroad Scholarship: Graz University of Technology
• 2022: Deep Learning + Reinforcement Learning (DLRL) Summer School: CIFAR, Mila, Vector Insitute, Amii
• 2021: UWaterloo Hackathon 2nd Place: University of Waterloo
• 2021: OverSEAs Scholarship: Graz University of Technology
• 2019: Merit-based Scholarship: Graz University of Technology

Design and source code from Jon Barron's website.