City:  Karlsruhe
Date:  Jan 19, 2023

Master thesis for self-supervised action recognition using contrastive learning

The Fraunhofer-Gesellschaft (www.fraunhofer.com) currently operates 76 institutes and research institutions throughout Germany and is the world’s leading applied research organization. Around 30 000 employees work with an annual research budget of 2.9 billion euros. 

Fraunhofer IOSB is doing basic research and exploring applications for optronic sensors, ranging from light production and detection to complex assistant and autonomous systems supporting humans in a wide variety of tasks. 

 

The team »Perceptual User Interfaces« of Fraunhofer IOSB offers you a master thesis for the conceptual design and development of a self-supervised machine learning approach in the area of 3D skeleton based action recognition. In this thesis, we will research state-of-the art approaches for label-free 3D skeleton based action recognition. With a self-supervised contrastive learning approach label-free training can be achieved. The aim of this approach is a better generalization to unseen data. This thesis provides an opportunity to implement and evaluate self-supervised learning approaches in the area of action recognition. Fraunhofer IOSB is working on action recognition for human-machine interaction (HMI) in vehicle interiors.

 

What you will do

  • Design and development of a self-supervised machine learning approach for action recognition based on 3D skeleton data
  • Research of state-of-the art approaches for self-supervised action recognition
  • Implementation of state-of-the-art approaches and further development
  • Training of the approach on publicly available datasets for 3D skeleton data-based action recognition
  • Evaluation of the approach in the area of action recognition in vehicle interiors

 

What you bring to the table

  • Enrollment at a college or university with a suitable field of study
  • Independent, transparent way of working, commitment and reliability
  • Prior knowledge in the areas of python and machine learning 
  • Interest in the field of algorithm development 
  • 6-month master thesis in full-time 

 

What you can expect

  • An exciting work opportunity in applied research
  • Intensive support from teammembers of our department
  • Ideal conditions for practical experience alongsiede your studies
  • Good public transportation connections

 

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 are given preference in the event of equal suitability. 

With its focus on developing key technologies that are vital for the future and enabling the commercial utilization of this work by business and industry, Fraunhofer plays a central role in the innovation process. As a pioneer and catalyst for groundbreaking developments and scientific excellence, Fraunhofer helps shape society now and in the future. 

Interested? Apply online now. We look forward to getting to know you!

 

If you have any questions please contact:

David Lerch

Phone: +49 721 6091-157

david.lerch@iosb.fraunhofer.de

Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB 

www.iosb.fraunhofer.de 

 

Requisition Number: 61435                Application Deadline:

 


Job Segment: Training, Learning, Research, Education, Human Resources