Our research focuses on basic and applied research in machine learning. We develop approaches for anomaly detection, uncertainty estimation, out-of-distribution detection, few-shot learning, noisy label learning, as well as hardware-aware algorithms for model compression and efficient design of neural network architectures. Our applications include medical image analysis with problems such as cell counting and segmentation in histology images, organ segmentation from CT and MRI volumes, and multi-label medical image segmentation.
Research projects
Few-shot learning for medical images.
Cell Segmentation.
Multi-label image classification.
Subcontracting within the framework of the ÖGP NXT GEN AI METHODS – Generative methods for perception, prediction, and planning (NXTAIM)
(Third Party Funds Single)
Project leader: Vasileios Belagiannis Term: 1. January 2026 - 31. December 2026 Acronym: ÖGP NXT GEN AI METHODS Funding source: Industrie URL: https://nxtaim.de/en/home/
The aim of this project is to develop self-playing, multi-agent simulators based on GPUDrive. In this context, we will develop trajectory planning strategies with a focus on efficient training and optimisation. These strategies will then be tested in automated driving scenarios.
SUSTAINET-inNOvAte: Nachhaltige Technologien für fortschrittliche resiliente und energieeffiziente Netze - Reibungslose, sichere und widerstandsfähige Netze für die dynamische digitale Welt
(Third Party Funds Group – Sub project)
Overall project: Sustainable Technologies for Advanced Resilient and Energy-Efficient Networks - Frictionless, secure, and resilient communication networks for the dynamic digital world Project leader: Vasileios Belagiannis Term: 1. January 2025 - 31. December 2027 Acronym: SUSTAINET-inNOvAte Funding source: BMFTR / Verbundprojekt
The project deals with two aspects of generative modeling. First, generative models, especially fundamental models, have made a significant contribution in the areas of image, text, and audio. However, they have not yet been well researched for sequential and unstructured data, such as automotive data. Second, the latent spatial representation in generative models is not interpretable. However, this is linked to the predicted or generated output. This project aims to explore generative models for trajectory planning by integrating robustness measures.
Liu, Y., Chen, Y., Wang, H., Belagiannis, V., Reid, I., & Carneiro, G. (2025). ItTakesTwo: Leveraging Peer Representations for Semi-supervised LiDAR Semantic Segmentation. In Aleš Leonardis, Elisa Ricci, Stefan Roth, Olga Russakovsky, Torsten Sattler, Gül Varol (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 81-99). Milan, IT: Springer Science and Business Media Deutschland GmbH.
Briegleb, A., Haubner, T., Belagiannis, V., & Kellermann, W. (2023). Localizing Spatial Information in Neural Spatiospectral Filters. In IEEE (Eds.), Proceedings of the 2023 31st European Signal Processing Conference (EUSIPCO) (pp. 920-924). Helsinki, Finland.
Dawoud, Y., Bouazizi, A., Ernst, K., Carneiro, G., & Belagiannis, V. (2023). Knowing What to Label for Few Shot Microscopy Image Cell Segmentation. In Proceedings - 2023 IEEE Winter Conference on Applications of Computer Vision, WACV 2023 (pp. 3557-3566). Waikoloa, HI, USA: Institute of Electrical and Electronics Engineers Inc..
Dawoud, Y., Carneiro, G., & Belagiannis, V. (2023). SelectNAdapt: Support Set Selection for Few-Shot Domain Adaptation. In Proceedings - 2023 IEEE/CVF International Conference on Computer Vision Workshops, ICCVW 2023 (pp. 973-982). Paris, FRA: Institute of Electrical and Electronics Engineers Inc..
Holzbock, A., Hegde, A., Dietmayer, K., & Belagiannis, V. (2023). DATA-FREE BACKBONE FINE-TUNING FOR PRUNED NEURAL NETWORKS. In European Signal Processing Conference (pp. 1255-1259). Helsinki, FIN: European Signal Processing Conference, EUSIPCO.
Holzbock, A., Tsaregorodtsev, A., & Belagiannis, V. (2023). Pedestrian Environment Model for Automated Driving. In IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC (pp. 534-540). Bilbao, ES: Institute of Electrical and Electronics Engineers Inc..
Hornauer, J., & Belagiannis, V. (2023). Heatmap-based Out-of-Distribution Detection. In Proceedings - 2023 IEEE Winter Conference on Applications of Computer Vision, WACV 2023 (pp. 2602-2611). Waikoloa, HI, USA: Institute of Electrical and Electronics Engineers Inc..
Hornauer, J., Holzbock, A., & Belagiannis, V. (2023). Out-of-Distribution Detection for Monocular Depth Estimation. In Proceedings of the IEEE International Conference on Computer Vision (pp. 1911-1921). Paris, FR: Institute of Electrical and Electronics Engineers Inc..
Kern, N., Holzbock, A., Grebner, T., Belagiannis, V., Dietmayer, K., & Waldschmidt, C. (2022). A Ground Truth System for Radar Measurements of Humans. In 2022 German Microwave Conference, GeMiC 2022 (pp. 84-87). Ulm, DEU: Institute of Electrical and Electronics Engineers Inc..
Ülger, O., Wiederer, J., Ghafoorian, M., Belagiannis, V., & Mettes, P. (2022). Multi-Task Edge Prediction in Temporally-Dynamic Video Graphs. In BMVC 2022 - 33rd British Machine Vision Conference Proceedings. London, GBR: British Machine Vision Association, BMVA.
2021
Bouazizi, A., Kressel, U., & Belagiannis, V. (2021). Learning Temporal 3D Human Pose Estimation with Pseudo-Labels. In AVSS 2021 - 17th IEEE International Conference on Advanced Video and Signal-Based Surveillance. Virtual, Online, USA: Institute of Electrical and Electronics Engineers Inc..
Bouazizi, A., Wiederer, J., Kressel, U., & Belagiannis, V. (2021). Self-Supervised 3D Human Pose Estimation with Multiple-View Geometry. In Vitomir Struc, Marija Ivanovska (Eds.), Proceedings - 2021 16th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2021. Virtual, Jodhpur, IND: Institute of Electrical and Electronics Engineers Inc..
Casas, L., Klimmek, A., Navab, N., & Belagiannis, V. (2021). Adversarial signal denoising with encoder-decoder networks. In European Signal Processing Conference (pp. 1467-1471). Amsterdam, NLD: European Signal Processing Conference, EUSIPCO.
Conrad, J., Jiang, B., Kaesser, P., Ortmanns, M., & Belagiannis, V. (2021). Nonlinearity Modeling for Mixed-Signal Inference Accelerators in Training Frameworks. In 2021 28th IEEE International Conference on Electronics, Circuits, and Systems, ICECS 2021 - Proceedings. Dubai, ARE: Institute of Electrical and Electronics Engineers Inc..
Dawoud, Y., Hornauer, J., Carneiro, G., & Belagiannis, V. (2021). Few-Shot Microscopy Image Cell Segmentation. In Yuxiao Dong, Dunja Mladenic, Craig Saunders (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 139-154). Virtual, Online: Springer Science and Business Media Deutschland GmbH.
Engel, N., Belagiannis, V., & Dietmayer, K. (2021). Attention-based Vehicle Self-Localization with HD Feature Maps. In IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC (pp. 76-83). Indianapolis, IN, USA: Institute of Electrical and Electronics Engineers Inc..
Liu, F., Tian, Y., Cordeiro, F.R., Belagiannis, V., Reid, I., & Carneiro, G. (2021). Self-supervised Mean Teacher for Semi-supervised Chest X-Ray Classification. In Chunfeng Lian, Xiaohuan Cao, Islem Rekik, Xuanang Xu, Pingkun Yan (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 426-436). Virtual: Springer Science and Business Media Deutschland GmbH.
Sachdeva, R., Cordeiro, F.R., Reid, I., Carneiro, G., & Belagiannis, V. (2021). EvidentialMix: Learning with combined open-set and closed-set noisy labels. In Proceedings - 2021 IEEE Winter Conference on Applications of Computer Vision, WACV 2021 (pp. 3606-3614). Virtual, US: Institute of Electrical and Electronics Engineers Inc..
Schreiber, M., Glaeser, C., Dietmayer, K., & Belagiannis, V. (2021). Dynamic Occupancy Grid Mapping with Recurrent Neural Networks. In Proceedings - IEEE International Conference on Robotics and Automation (pp. 6717-6724). Xi'an, CHN: Institute of Electrical and Electronics Engineers Inc..
Wiederer, J., Bouazizi, A., Kressel, U., & Belagiannis, V. (2020). Traffic control gesture recognition for autonomous vehicles. In IEEE International Conference on Intelligent Robots and Systems (pp. 10676-10683). Las Vegas, NV, USA: Institute of Electrical and Electronics Engineers Inc..
Our research focuses on basic and applied research in machine learning. We develop approaches for anomaly detection, uncertainty estimation, out-of-distribution detection, few-shot learning, noisy label learning, as well as hardware-aware algorithms for model compression and efficient design of neural network architectures. Our applications include medical image analysis with problems such as cell counting and segmentation in histology images, organ segmentation from CT and MRI volumes, and multi-label medical image segmentation.
Research projects
Subcontracting within the framework of the ÖGP NXT GEN AI METHODS – Generative methods for perception, prediction, and planning (NXTAIM)
(Third Party Funds Single)
Term: 1. January 2026 - 31. December 2026
Acronym: ÖGP NXT GEN AI METHODS
Funding source: Industrie
URL: https://nxtaim.de/en/home/
The aim of this project is to develop self-playing, multi-agent simulators based on GPUDrive. In this context, we will develop trajectory planning strategies with a focus on efficient training and optimisation. These strategies will then be tested in automated driving scenarios.
Bavarian Advanced Resolution Radar
(Third Party Funds Group – Sub project)
Project leader: Vasileios Belagiannis
Term: 1. February 2025 - 31. January 2028
Acronym: BAVAR-RADAR
Funding source: Bayerisches Staatsministerium für Wirtschaft, Landesentwicklung und Energie (StMWi) (seit 2018)
SUSTAINET-inNOvAte: Nachhaltige Technologien für fortschrittliche resiliente und energieeffiziente Netze - Reibungslose, sichere und widerstandsfähige Netze für die dynamische digitale Welt
(Third Party Funds Group – Sub project)
Project leader: Vasileios Belagiannis
Term: 1. January 2025 - 31. December 2027
Acronym: SUSTAINET-inNOvAte
Funding source: BMFTR / Verbundprojekt
Subcontracting within the framework of the ÖGP NXT-AIM Generative Modeling
(Third Party Funds Single)
Term: 1. January 2024 - 31. December 2026
Acronym: Unterbeauftragung ÖGP NXT-AIM
Funding source: Industrie
URL: https://nxtaim.de/en/home/
The project deals with two aspects of generative modeling. First, generative models, especially fundamental models, have made a significant contribution in the areas of image, text, and audio. However, they have not yet been well researched for sequential and unstructured data, such as automotive data. Second, the latent spatial representation in generative models is not interpretable. However, this is linked to the predicted or generated output. This project aims to explore generative models for trajectory planning by integrating robustness measures.
2026
2025
2024
2023
2022
2021
2020
Related Research Fields:
Contact: