About
I am a Computer Vision and AI researcher with several years of experience developing deep learning solutions for real-world industrial applications.
I currently work as a Project Assistant at the Institute of Visual Computing at Graz University of Technology, Austria, while pursuing a PhD in Computer Science under the supervision of Prof. Dr. Horst Bischof and Dr. Horst Possegger. I received my Master’s degree in Mathematics and Applied Mathematics from M.V.Lomonosov Moscow State University (MSU), graduating from the Faculty of Mechanics and Mathematics.
My work focuses on computer vision systems that bridge cutting-edge research and industrial deployment. My research interests include visual foundation models, vision-language models, multiple object detection and tracking, domain adaptation, and automated visual inspection for industrial environments.
In recent years, I have worked on AI solutions for intelligent transportation systems, visual surveillance, automated textile inspection, and autonomous driving. At TU Graz, we collaborate with industrial partners including Infineon Technologies, AVL, ASFINAG, KNAPP, and Saubermacher, contributing to projects that transfer state-of-the-art AI technologies into practical applications. My experience spans the complete AI development pipeline – from data acquisition and dataset creation to model design, evaluation, and deployment in industrial and public-sector projects.
I am passionate about developing reliable and scalable AI systems that create measurable value beyond the laboratory and help solve challenging real-world problems.
My Projects
Damage Detection in Second-Hand Garments
Multimodal Domain Adaptation for Air Traffic Management
Fog Simulation for Testing Vision-Based AI Systems
Monocular Depth Estimation in Real-World Environments
Tracking Road Users for Improved Traffic Safety
Understanding Pedestrian Behavior at Crosswalks
Safety Crosswalks Planning Using Movement Patterns
Compliance with Traffic Light Signals for Pedestrian Safety
Pedestrian Awareness of Approaching Vehicle and Traffic Risks
Pedestrian Distractions in Traffic Environments
Traffic Safety for People with Limited Mobility
Enhancing Pedestrian Monitoring with ToF Sensors
Improving Detection of Pedestrians and Cyclists with LiDAR
Camera – LiDAR Calibration for Multisensor Perception Systems
Cross-View Calibration Between In-Vehicle and Roadside Cameras
Automated Robust Anonymization of Vehicle License Plates in Videos
My Publications
-
A Synthetic Multi-View Tracking and 3D Pose Dataset for Automated Airport Visual Surveillance
In Proc. of the Middle-European Conference on Applied Theoretical Computer Science (MATCOS), IS-2025
Proceedings Cite this paperCodeBibTeX:
@inproceedings{Beleznai2025,
title = {{A Synthetic Multi-View Tracking and 3D Pose Dataset for Automated Airport Visual Surveillance}},
author = {Mansour, Ahmed and Beleznai, Csaba and Francisco Oberweger, Fabio and Widhalm, Verena and Kirillova, Nadezda and Possegger, Horst},
booktitle = {In Proc. of the Middle-European Conference on Applied Theoretical Computer Science (MATCOS)},
year = {IS-2025}
}
-
Long-term observation to evaluate cyclist behaviour at RTOR junction in Austria
In Proc. of International Cycling Safety Conference (ICSC), 2025
Cite this paperBibTeX:
@inproceedings{Tomasch2025,
title = {{Long-term observation to evaluate cyclist behaviour at RTOR junction in Austria}},
author = {Tomasch, Ernst and Schneider, Bernd and Hoschopf, Heinz and Kirillova, Nadezda and Possegger, Horst},
booktitle = {In Proc. of International Cycling Safety Conference (ICSC)},
year = {2025}
}
-
Into the Fog: Evaluating Robustness of Multiple Object Tracking
In Proc. of the British Machine Vision Conference (BMVC), 2024
PaperProceedings Cite this paperSupplementaryPosterCodeBibTeX:
@inproceedings{Kirillova_2024_BMVC,
title = {{Into the Fog: Evaluating Robustness of Multiple Object Tracking}},
author = {Kirillova, Nadezda** and Mirza, M. Jehanzeb and Bischof, Horst and Possegger, Horst},
booktitle = {In Proc. of the British Machine Vision Conference (BMVC)},
year = {2024}
}
-
A Comprehensive Crossroad Camera Dataset to Improve Traffic Safety of Mobility Aid Users
In Proc. of the British Machine Vision Conference (BMVC), 2023
Paper Cite this paperSupplementaryPosterVideoCodeDatasetBibTeX:
@inproceedings{mohr23a,
title = {{A Comprehensive Crossroad Camera Dataset to Improve Traffic Safety of Mobility Aid Users}},
author = {Mohr, Ludwig and Kirillova, Nadezda and Possegger, Horst and Bischof, Horst},
booktitle = {In Proc. of the British Machine Vision Conference (BMVC)},
year = {2023}
}
-
Identifying and Extracting Pedestrian Behavior in Critical Traffic Situations
In Proc. of the IEEE International Conference on Intelligent Transportation Systems (ITSC), 2023
Paper Cite this paperBibTeX:
@inproceedings{schachner23b,
title = {{Identifying and Extracting Pedestrian Behavior in Critical Traffic Situations}},
author = {Schachner, Martin and Schneider, Bernd and Wei{\ss}enbacher, Fabian and Kirillova, Nadezda and Possegger, Horst and Bischof, Horst and Klug, Corina},
booktitle = {In Proc. of the IEEE International Conference on Intelligent Transportation Systems (ITSC)},
year = {2023}
}
-
Observation-based Pedestrian Scenario Extraction for Virtual Testing
In Proc. of the International Technical Conference on the Enhanced Safety of Vehicles (ESV), 2023
Proceedings Cite this paperBibTeX:
@inproceedings{schachner23a,
title = {{Observation-based Pedestrian Scenario Extraction for Virtual Testing}},
author = {Schachner, Martin and Kirillova, Nadezda and Wei{\ss}enbacher, Fabian and Schneider, Bernd and Possegger, Horst and Bischof, Horst and Aichberger, Arno and Magosi, Zoltan Ferenc and Dobberstein, Jan and Lich, Thomas and Kirchengast, Martin and Hennecke, Marcus and Klug, Corina},
booktitle = {In Proc. of the International Technical Conference on the Enhanced Safety of Vehicles (ESV)},
year = {2023}
}
-
A Visual Surveillance System to Observe Realistic Road User Behavior for Improved Pedestrian and Cyclist Safety at Crossroads
In Proc. of the IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS), 2022
Paper Cite this paperBibTeX:
@inproceedings{kirillova22a,
title = {{A Visual Surveillance System to Observe Realistic Road User Behavior for Improved Pedestrian and Cyclist Safety at Crossroads}},
author = {Kirillova, Nadezda and Possegger, Horst and Bischof, Horst},
booktitle = {In Proc. of the IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS)},
year = {2022}
}