Master Thesis Open-World 3D Anomaly Segmentation with Vision-Language Models (f/m/x)

TN Germany; München; Vor Ort; EUR 60.000 - 80.000; Vollzeit

Zusammenfassung

Join a leading company for an exciting Master thesis in open-world 3D anomaly segmentation. Focus on leveraging contrastive vision-language models to segment unknown objects in 3D. The project utilizes innovative multi-modal segmentation approaches and is supervised by a professor at the Technical University of Munich.

Leistungen

Digital offers & mobile working

Attractive remuneration

Apartment offers for students

Qualifikationen

• Experience with vision-language models is a plus but not mandatory.

Aufgaben

• Develop vision-language models for open-world perception.
• Implement pseudo-labeling techniques with segmentation models.
• Explore anomaly detection through contrastive feature representations.

Kenntnisse

Deep Learning

Computer Vision

Ausbildung

Enrollment at TUM

Jobbeschreibung

Master Thesis: Open-World 3D Anomaly Segmentation with Vision-Language Models (f/m/x), Munich

Client: BMW Group

Location: Munich, Germany

Job Category: Other

EU work permit required: Yes

Job Reference: 79edb011041d

Job Views: 1

Posted: 16.05.2025

Expiry Date: 30.06.2025

Job Description:

Join BMW Group for an exciting Master thesis in open-world 3D anomaly segmentation, focusing on leveraging contrastive vision-language models to understand and segment unknown objects in 3D environments. The project will utilize the Open World 3D dataset, which includes 3D bounding boxes and 2D image data, enabling innovative multi-modal segmentation approaches.

What awaits you?

• Develop vision-language models for open-world perception.
• Implement pseudo-labeling techniques with segmentation models.
• Explore anomaly detection through contrastive feature representations.
• Create methods to generate 3D semantic segmentation data from existing annotations.
• Evaluate open-vocabulary 2D segmentation as a basis for 3D segmentation.
• Compare various feature representations, including BEV-based heatmaps, for improved segmentation.

This thesis will be supervised by a professor at the Technical University of Munich (TUM). Please ensure your thesis is supervised by a university.

What should you bring along?

• Strong background in deep learning and computer vision.
• Enrollment at TUM is preferred.
• Experience with vision-language models, multi-view projection, or semantic segmentation is a plus but not mandatory.

If passionate about advancing open-world 3D anomaly segmentation and working with cutting-edge vision-language models, we look forward to your application!

What do we offer?

• Digital offers & mobile working.
• Attractive remuneration.
• Apartment offers for students (subject to availability & only Munich).
• Many other benefits — see our website for details.

Earliest starting date: from 19.05.2025

Duration: 6 months

Working hours: Full-time

If you have questions, please contact us via the provided enquiry form. We will respond by phone or email.

BMW Group values diversity and inclusion, believing they strengthen our teams. Equal opportunities and fair treatment are fundamental to our corporate policy. Our recruitment decisions are based on personality, experience, and skills.

Learn more about diversity at BMW Group on our website.