Description

In this master’s thesis, you will explore the exciting intersection of Generative AI and Automotive AI by working with advanced diffusion models. Your work will involve generating high-fidelity synthetic images and analyzing their effects on computer vision tasks, while contributing to innovative driver monitoring systems.

Requirements

• •Enrollment in a master’s program in Artificial Intelligence, Machine Learning, Computer Vision, Computer Science, Robotics, Mathematics, or a related field
• •Strong Python programming skills
• •Solid understanding of deep learning and computer vision fundamentals
• •Experience with PyTorch or similar deep learning frameworks
• •Prior exposure to diffusion models, GANs, or synthetic data generation
• •Experience with pose estimation, activity recognition, or multimodal models
• •Familiarity with CLIP, ControlNet, or image quality evaluation metrics
• •Experience using Git, Docker, and GPU-based training
• •Previous research projects, internships, or open-source contributions are a plus
• •Independent, structured, and goal-oriented
• •Strong analytical thinkers with good problem-solving skills
• •Curious and creative when exploring new ideas
• •Good communicators who enjoy working in a team
• •Motivated to deliver high-quality and reproducible research
• •Proof of enrollment in a master’s program
• •A current transcript of grades
• •An excerpt from your study regulations confirming master’s thesis requirements
• •Valid residence permit and work permit if not from the EU

Tasks

• •Implement diffusion-based image generation models
• •Experiment with image generation techniques
• •Develop an evaluation pipeline for synthetic in-cabin image datasets
• •Investigate metric-driven feedback loops for image quality improvement
• •Analyze synthetic data's impact on computer vision tasks
• •Summarize findings in a master’s thesis and potential research paper or patent

Tools & Technologies

Benefits