Osnabrück Vision AI (OSVIA) Lab

The OSVIA Lab researches methods that teach machines to see – from the basics of classical computer vision to modern deep learning approaches. Our goal is to develop algorithms that process visual data reliably, efficiently, and explainably. We work on problems ranging from the recognition and analysis of complex scenes to 3D reconstruction and robust machine learning. With our research, we bridge the gap between theory and application – for advances in science, industry, and society.

Universität Osnabrück: AG Computer Vision and Robust AI

Bergische Universität Wuppertal: AG Applied and Computational Mathematics

Recent Publications

More Publications

LFA Applied to CNNs: Efficient Singular Value Decomposition of Convolutional Mappings by Local Fourier Analysis

Van Betteray, A., Kahl, K., Rottmann, M. (2025)

Paper Project Page
FutrTrack: A Camera-LiDAR Fusion Transformer for 3D Multiple Object Tracking

Teye, M., Maoz, O., Rottmann, M. (2025)

Paper Project Page
Learning to Detect Label Errors by Making Them

Penquitt, S., Riedlinger, T., Heller, T., Reischl, R., Rottmann, M. (2025)

Paper GitHub Project Page

Datasets

More Datasets

Benchmark for Label Error Detection Methods in Object Detection

A Modular Framework and Benchmark for Object Detection Datasets. (2025)

Website
OoDIS

Anomaly Instance Segmentation Benchmark. (2024)

Website Project Page
SegmentMeIfYouCan

A Benchmark for Anomaly Segmentation. (2021)

Website

Projects

Ongoing and recent work with our research and industry partners.

Team

A multidisciplinary group committed to understanding and improving AI robustness in real-world scenarios.

Meet the Team

News

More News

Inaugural lecture by Prof. Rottmann and Prof. Mossakowski

Jan 20, 2026 UOS

On 20.01.2026, the Institute of Computer Science celebrated "50 years of computer science at Osnabrück University". Prior to this, Prof. Rottmann and Prof. Mossakowski presented their research areas in their inaugural lectures.

Workshop on Uncertainty Quantification at ControlExpert

Dec 12, 2025 Workshop

The OSVIA lab joined a scientific workshop hosted at ControlExpert together with the research team of Prof. Kira Maag from Heinrich Heine University Düsseldorf and colleagues from ControlExpert.

Two papers accepted at VISAPP 2026

Nov 6, 2025 VISAPP 2026 Paper

The OSVIA lab at the University of Osnabrück has two accepted papers at VISAPP 2026. The project pages and arXiv preprints are available.

DFG Funds Research on Reliable AI for Open-World Object Detection

Oct 10, 2025 Project DFG UOS

The German Research Foundation (DFG) has approved funding for a new research project led by Matthias Rottmann at Osnabrück University.

ICCV2025 Paper Acceptance

Jun 26, 2025 ICCV2025 Paper

Our work "TARS Traffic-Aware Radar Scene Flow Estimation" by Jialong Wu , Marco Braun, Dominic Spata and Matthias Rottmann (Aptiv, Bergische Universität Wuppertal) has been accepted for publication at ICCV2025.

Paper Project page

Team Event in Bochum

Jun 13, 2025 Team event

Our team headed underground at the German Mining Museum to see massive, century-old machinery in action and learn about 200 years of Ruhr region mining history—all while escaping the summer heat above.

Osnabrück Vision AI (OSVIA) Lab

Recent Publications

LFA Applied to CNNs: Efficient Singular Value Decomposition of Convolutional Mappings by Local Fourier Analysis

FutrTrack: A Camera-LiDAR Fusion Transformer for 3D Multiple Object Tracking

Learning to Detect Label Errors by Making Them

Datasets

Benchmark for Label Error Detection Methods in Object Detection

OoDIS

SegmentMeIfYouCan

Projects

Unified Uncertainty Estimation for Fine-tuned Open-Vocabulary Models in Image Classification and Object Detection

Project REFRAME

Project RELiABEL

Just Scan It 3D

Cooperation with QUT Center for Robotics

BMBF Junior Research Group UnrEAL