Master thesis - Deep learning-based zero-shot image segmentation

The Fraunhofer-Gesellschaft (www.fraunhofer.com) is one of the world's leading organizations for application-oriented research. 75 institutes develop pioneering technologies for our economy and society – more precisely: 32 000 people from technology, science, administration and IT. They know: Anyone who comes to Fraunhofer wants to and can make a difference. For themselves, for us and for the markets of today and tomorrow.

Increasing the efficiency and safety of industrial processes and products with innovative technologies – that's what we work on at the Fraunhofer Institute for Physical Measurement Techniques IPM in Freiburg. Around 270 employees use their expertise and enthusiasm to research and develop measurement methods and systems for production control, object and shape detection, gas and process technology, and photonic systems. You can expect a friendly team with exciting research topics in an inspiring work environment.

In our business unit, “Object and Shape Detection,” we develop high-precision, multisensor systems for three-dimensional measurements and surveying. Deployed on road vehicles, trains, drones, or underwater ROVs, these systems use laser scanners, cameras, and additional sensors to capture the environment. For instance, we develop autonomous robots for automated building inspections. We post-process the acquired sensor data to generate actionable insights. In your master's thesis, you will focus on this critical stage. Specifically, you will research deep learning models for image segmentation to detect damage to concrete buildings. Since conventional models require large amounts of precisely labeled training data that are not always available, you will also research zero-shot deep learning architectures.

We are Fraunhofer IPM. We measure. We control. We optimize. To do this, we need: curiosity. Courage. Creativity. Vision. Cooperation. Communication. And you!

What we offer

Remuneration in accordance with the general works agreement on the employment of auxiliary staff
Work-life balance with flexible work schedules
Equal opportunities
Support from experts in the field
Individual career planning & entry opportunities
Modern working environment
Canteen at the institute with daily fresh food
Monitored (electric) car and bicycle parking spaces

Be part of change

You will research the current state of the art.
You will implement, train, and test deep learning model architectures.
You will develop approaches for the damage segmentation of concrete infrastructure images based on zero-shot models.

What you contribute

Personality: You are responsible and team-oriented. You complete your tasks reliably and consistently, fostering a respectful and open collaboration with your colleagues. You actively contribute ideas and help to create a positive and motivating work environment.
Education: You currently study computer science, embedded systems engineering, or a related field at a master’s level.
Working style: You think analytically and can solve problems independently.
Experience: Ideally, you have a background in software engineering or deep learning. You have strong Python programming skills and comprehensive knowledge of deep learning frameworks, such as PyTorch or TensorFlow, as well as computer vision. Knowledge of natural language processing (NLP) is an advantage, and proficiency in Microsoft Office is expected.
Languages: We work in international teams and therefore require proficiency in English at a minimum B2 level, both written and spoken. Knowledge of German is an advantage but not required.

We value and promote the diversity of our employees' skills and therefore welcome all applications – regardless of age, gender, nationality, ethnic and social origin, religion, ideology, disability, sexual orientation, and identity. Severely disabled persons with equal qualifications will be given preference.

With its focus on future-oriented key technologies and the exploitation of results in business and industry, the Fraunhofer Society plays a central role in the innovation process. As a guide and driving force for innovative developments and scientific excellence, it helps shape our society and our future.

Ready for a change? Then apply now and make a difference! Once we have received your online application, you will receive an automatic confirmation of receipt. We will then get back to you as soon as possible and let you know what happens next.

Do you have any questions? We are happy to help.

Saskia Sailer
personal@ipm.fraunhofer.de
+49 761 8857 307

Fraunhofer Institute for Physical Measurement Techniques IPM

www.ipm.fraunhofer.de

Requisition Number: 82626

Job Segment: Test Engineer, Testing, Embedded, Software Engineer, Computer Science, Engineering, Technology

Apply now »