Chair of Computer Vision

KIWI@SIWI is a project focused on detecting bison using machine learning methods. The goal is to use this technology to better detect the animals and thereby support herd management. The project is being carried out by the University of Siegen and NanoGiant GmbH. It has received funding through the NEXT.IN.NRW funding program. 

As part of the interdisciplinary L2S research group, which bridges electrical engineering and computer science, this project focuses on fundamental machine learning methods to simultaneously optimize sensor system parameters and network parameters for the semantic analysis of the recorded data. 

This research project investigates how computer vision (CV) can benefit from quantum hardware to overcome the limitations of classical resources. First, the focus is on using quantum annealing to efficiently solve CV problems that can be formulated as quadratic unconstrained optimization problems (known as QUBOs). Second, quantum machine learning is being used to investigate how 3D data can be encoded more compactly on gate-based systems. 

The WISENT pilot line at INCYTE offers an innovative infrastructure for chip and sensor development. Companies and startups can develop prototypes here without needing their own large-scale equipment. The use of novel materials (e.g., graphene), 3D-printed optics, and integrated AI hardware enables the creation of intelligent, high-performance sensors. The Computer Vision Group contributes to demonstrators in the field of data analysis using machine learning methods. 

This project explores fundamental approaches in machine learning that analyze data from wearable sensors without requiring time-consuming manual annotation. Through weakly supervised federated learning, local video data is used only on-site to automatically control the training of the wrist sensors. In particular, video data will be used to control the training of a neural network for analyzing data from wearable sensors. 

Launched in 2022, the Lamarr Institute combines cutting-edge AI research with practical applications. With annual funding of 20 million euros from the federal government and the state of North Rhine-Westphalia, it offers excellent conditions for researchers. Through the “Lamarr Fellow Network Program,” the state also specifically connects AI talent from across North Rhine-Westphalia. Michael Möller was named a Lamarr Fellow in 2023.