Publications

A serious problem in image classification is that a trained model might perform well for input data that originates from the same distribution as the data available for model training but performs much worse for out-of-distribution (OOD) samples. This paper proposes a first-of-its-kind OOD detection architecture named OODformer that leverages the contextualization capabilities of the transformer.

The quality of semantic image segmentation models can be affected by external factors such as weather or daytime. Those factors can lead to safety-critical mistakes. In this work, we propose a systematic approach to detect and alleviate such weaknesses of semantic segmentation models.

Automated cars have to take correct decisions in complex situations. For this, the understanding of a vehicle system’s own capabilities and the environmental context is crucial. We introduce our approach of enhancing the system awareness of vehicles to handle changes gracefully, while optimizing the overall performance.

Together with the Huawei Research Center Munich, Fraunhofer IKS has written a white paper on Artificial Intelligence (AI) for Intelligent Transport Systems (ITS). It summarizes current and future challenges of introducing AI into Intelligent Traffic Systems in a trustworthy manner. Here, special focus is laid on Smart Cities.

As part of the ADA Lovelace Center for Analytics, Data and Applications,  Fraunhofer IKS has developed a technical white paper on machine learning methods for reliable perception of autonomous systems. It reviews, develops and evaluates new methods for quantifying uncertainty in deep neural networks.

In collaboration with Magazino GmbH, Frauhofer IKS presents a white paper on the obstacles and potentials of “DevOps for developing cyber-physical systems”:

• What is the difference between DevOps and traditional software development?
• What are the challenges in the DevOps process for cyber-physical systems?
• How can DevOps processes be implemented?