Thesis Efficient Deep Neural Networks for Bird's Eye View Perception (f/m/d)

Thesis Efficient Deep Neural Networks for Bird's Eye View Perception (f/m/d)

Job ID:  10510
Company:  Volkswagen AG
Location: 

Wolfsburg, DE, 38436

Department:  Research and Development
Career Level:  Interns
Working Model:  Full-time
Contract Type:  Fixed-term
Remote Working:  Up to 40%
Posting Date:  Dec 5, 2024

Thesis Efficient Deep Neural Networks for Bird's Eye View Perception (f/m/d)

 

 

Environment

As part of Volkswagen Group Innovation, the AI & Data Analytics subdivision deals with the methodology and concept development of digital services for all Volkswagen Group brands. The main topics are, on the one hand, the processing and analysis of vehicle data and, on the other hand, the integration of artificial intelligence (AI) into digitized and networked vehicles. With the help of AI, for example, aging information and misbehavior of vehicle components are detected, but also perception tasks of autonomous driving are solved. Within these areas of responsibility, we offer you the opportunity to participate in the research and development of intelligent algorithms in a young and interdisciplinary team and to help shape the mobility of tomorrow.

Possible tasks within this role

  • Designing and implementating of efficient Birds Eye View (BEV) based segmentation and planning models for autonomous driving.

  • Building robust deep learning architectures capable of processing high-resolution images, improving the prediction performance of BEV perception models.

  • Identifying and addressing bottlenecks in current BEV research, pushing forward the boundaries of BEV-based perception for autonomous applications.

  • Optimizing hyperparameters and network architectures using various optimization algorithms.

Qualification requirements

  • Currently enrolled in a Master’s program in Data Science, Computer Science, Computational Science, Mathematics, Physics or a related field.

  • Strong ability to comprehend, analyze, and present research papers within BEV and related domains.

  • Proficiency with PyTorch, CUDA, and other relevant deep learning frameworks.

  • Strong programming skills in Python and/or C#, with experience in CI/CD pipelines to streamline development.

  • Ability to dive in and work independently on a scientific topic.

  • Previous research experiences in the field of BEV perception and 3D Computer Vision is an advantage.

  • German and/or English level B2.

The following documents must be submitted with the application

  • Curriculum Vitae

  • Current certificate of enrolment

  • Current transcript of records

  • In case of a compulsory internship, a certificate from the university

  • Work permit for non-EU citizens


Contact person for this posting: Brigitte Adam-Huth

 

 

Environment

As part of Volkswagen Group Innovation, the AI & Data Analytics subdivision deals with the methodology and concept development of digital services for all Volkswagen Group brands. The main topics are, on the one hand, the processing and analysis of vehicle data and, on the other hand, the integration of artificial intelligence (AI) into digitized and networked vehicles. With the help of AI, for example, aging information and misbehavior of vehicle components are detected, but also perception tasks of autonomous driving are solved. Within these areas of responsibility, we offer you the opportunity to participate in the research and development of intelligent algorithms in a young and interdisciplinary team and to help shape the mobility of tomorrow.

Possible tasks within this role

  • Designing and implementating of efficient Birds Eye View (BEV) based segmentation and planning models for autonomous driving.

  • Building robust deep learning architectures capable of processing high-resolution images, improving the prediction performance of BEV perception models.

  • Identifying and addressing bottlenecks in current BEV research, pushing forward the boundaries of BEV-based perception for autonomous applications.

  • Optimizing hyperparameters and network architectures using various optimization algorithms.

Qualification requirements

  • Currently enrolled in a Master’s program in Data Science, Computer Science, Computational Science, Mathematics, Physics or a related field.

  • Strong ability to comprehend, analyze, and present research papers within BEV and related domains.

  • Proficiency with PyTorch, CUDA, and other relevant deep learning frameworks.

  • Strong programming skills in Python and/or C#, with experience in CI/CD pipelines to streamline development.

  • Ability to dive in and work independently on a scientific topic.

  • Previous research experiences in the field of BEV perception and 3D Computer Vision is an advantage.

  • German and/or English level B2.

The following documents must be submitted with the application

  • Curriculum Vitae

  • Current certificate of enrolment

  • Current transcript of records

  • In case of a compulsory internship, a certificate from the university

  • Work permit for non-EU citizens


Contact person for this posting: Brigitte Adam-Huth

Job ID:  10510
Company:  Volkswagen AG
Location: 

Wolfsburg, DE, 38436

Department:  Research and Development
Career Level:  Interns
Working Model:  Full-time
Contract Type:  Fixed-term
Remote Working:  Up to 40%
Posting Date:  Dec 5, 2024

Volkswagen - that's why

We want to revolutionize and shape the mobility of tomorrow sustainably. For this purpose, we are looking for talents who are willing to break completely new ground with us and are not afraid to make mistakes. We are looking for personalities who want to achieve something extraordinary. And talents who believe in the innovative power of big ideas. Are you in?

Go to our career website

Questions?

Your application | All you need to know
FAQ – Frequently Asked Questions
Mail: einstieg@volkswagen.de ✉️
Hotline: +49 5361 9 36363


The costs for calls depend on the your provider's tariff. Roaming charges may apply for calls from abroad.

Technical Issues?

Technical Recruiting Support ✉️