You will conduct research on multi-object tracking and safety in autonomous driving, implement advanced algorithms, and collaborate with experts to enhance perception systems.
Anforderungen
- •Enrolled student in computer science
- •Strong background in computer vision
- •Proficient in Python, PyTorch, Git, OpenCV
- •Experience with C++ is beneficial
- •Excellent problem-solving skills
- •Strong analytical and communication skills
- •Collaborative attitude
- •Familiarity with diverse tracking architectures
Deine Aufgaben
- •Conduct in-depth research on multi-object tracking.
- •Summarize findings on safety-critical applications.
- •Implement and evaluate advanced perception algorithms.
- •Train tracking systems using modular and end-to-end methods.
- •Enhance bounding box accuracy with new techniques.
- •Analyze algorithm performance with large-scale datasets.
- •Collaborate with ADAS/AD perception experts.
Deine Vorteile
Remote work options
Duration of 3 to 6 months
35-hour work week
Original Beschreibung
# Internship / Master Thesis - Optimizing AD perception using Egocentric based matching metric (f/m/d)
Job ID:
13834
Company:
CARIAD SE
Location:
Ingolstadt, DE, 85053
München, DE, 80807
Berlin, DE, 10178
Department:
IT Software Development
Career Level:
Students
Working Model:
Full-time
Contract Type:
Fixed-term
Remote Working:
By agreement
Posting Date:
May 10, 2025
## YOUR TEAM
To support our "ADAS/AD Predevelopment" team, we are currently looking for an intern or thesis student. Our department works on software for automated driving in urban environments. Within this department, we are a team of ambitious and highly motivated experts in the field of environment perception for self-driving vehicles.
## WHAT YOU WILL DO
* Conduct an in-depth study and summarization of existing research related to multi-object tracking, safety-critical applications, and situation awareness in autonomous driving
* Implement, train, and rigorously evaluate advanced perception algorithms for both modular (e.g., tracking-by-detection) and end-to-end multi-object tracking systems
* Research and apply cutting-edge techniques, such as training tracking pipelines with new egocentric matching criteria integrated as loss functions, to directly enhance bounding box accuracy and robustness, particularly for partially occluded objects
* Systematically analyze algorithm performance using large-scale datasets and contribute improvements to our AD perception stack
* Collaborate closely with ADAS/AD perception experts on state-of-the-art autonomous driving challenges
## WHO YOU ARE
* Enrolled student in computer science, electrical engineering, robotics, or related field with a strong focus on computer vision and machine learning (please specify expected date of your graduation or end of enrollment)
* Strong background in computer vision (esp. multi-object tracking, object detection) and deep learning
* Proficient in Python, PyTorch, Git and OpenCV. Experience with C++ is beneficial
* Excellent problem-solving skills demonstrated through projects or prior experience; possess a research-oriented mindset
* Strong analytical and communication skills (English required); collaborative attitude
* Familiarity with diverse tracking architectures or advanced evaluation techniques is advantageous
## NICE TO KNOW
* Remote work options within Germany
* Duration: 3 to 6 months
* 35-hour week
At CARIAD, we embrace individuality and diversity because we believe our differences make us stronger. We actively seek to build teams with a variety of backgrounds, perspectives, and experiences. Our goal is to create an environment where everyone feels valued and empowered to contribute.