About this role
## Company:
Qualcomm Auto Ltd Sweden Filial
## Job Area:
Interns Group, Interns Group > Interim Intern
General Summary:
The project is expected to provide recommendations on suitable foundation models, encoder architectures, and pretraining strategies for future ADAS perception systems, supported by experimental evaluation and quantitative analysis.
Background
Recent foundation models have demonstrated strong transfer learning capabilities across a wide range of computer vision tasks. In the automotive perception domain, it remains unclear which tasks benefit the most from foundation model pretraining and what model architectures are best suited for different perception tasks.
Scope
The goal of this project is to investigate, develop, and evaluate knowledge distillation methods for transferring knowledge from larger pretrained foundation models into compact ADAS perception models suitable for inference on vehicle hardware. The work will compare different teacher models, covering both image-domain and multi-view spatiotemporal approaches. The impact of distillation on downstream performance will be evaluated on tasks such as lane detection, traffic sign recognition, vehicle detection, and other road scene understanding applications.
Key research questions include:
· Which ADAS perception tasks benefit most from foundation model pretraining?
· How do different foundation model architectures compare in terms of performance and data efficiency?
· Does a foundation model pre-training based on 2D image task benefit a 3D application?
· How well do pretrained representations transfer across different perception tasks and datasets?
Self-supervised pretraining for object detection with few labels
The project is expected to yield recommendations on self-supervised pretraining strategies that reduce the dependence of 3D object detection on manual annotation, supported by an experimental evaluation of label efficiency and robustness.
Background
Training accurate 3D object detectors requires large annotated datasets that are costly to produce and affected by label noise. Self-supervised learning can exploit unannotated data and especially joint-embedding predictive architectures have the potential to unlock efficient pre-training for high-quality object detection.
Scope
The goal is to investigate self-supervised pretraining of bird's-eye-view (BEV) representations from LiDAR and camera data for object detection under limited annotation. The work comprises evaluating latent-space prediction on large unlabeled datasets and evaluating the pretrained encoders on 3D object detection at varying label fractions against training from scratch and image-space pretraining.
Key research questions include:
- Which self-supervised pretraining strategy produces the most label-efficient BEV representation for object detection?
- Is it more beneficial to learn representations in image space or in scene-centric BEV space?
- Does BEV-level self-supervised pretraining reduce sensitivity to annotation noise?
Required Skills
- Machine Learning and Deep Learning
- Computer Vision
- Python and PyTorch
- Interest in autonomous driving and ADAS technologies
Student Background
- 4th year student on a M.Sc. program in Computer Science and Engineering, Applied Physics and Electrical Engineering, Information Technology or related field — preferably with a specialization in machine learning and/or computer vision.
- Good, documented knowledge of deep learning techniques
- Good programming skills in Python
Take the chance to express your interest already now — we will contact you when we begin reviewing thesis applications during autumn 2026.
Employment Conditions
- Type: Fixed‑term employment
- Period: Spring 2027
- Location: Linköping
- Last application date: (to be added by HR)
*References to a particular number of years experience are for indicative purposes only. Applications from candidates with equivalent experience will be considered, provided that the candidate can demonstrate an ability to fulfill the principal duties of the role and possesses the required competencies.
Applicants : Qualcomm is an equal opportunity employer. If you are an individual with a disability and need an accommodation during the application/hiring process, rest assured that Qualcomm is committed to providing an accessible process. You may e-mail [email protected] or call Qualcomm's toll-free number found here . Upon request, Qualcomm will provide reasonable accommodations to support individuals with disabilities to be able participate in the hiring process. Qualcomm is also committed to making our workplace accessible for individuals with disabilities. (Keep in mind that this email address is used to provide reasonable accommodations for individuals with disabilities. We will not respond here to requests for updates on applications or resume inquiries).
Qualcomm expects its employees to abide by all applicable policies and procedures, including but not limited to security and other requirements regarding protection of Company confidential information and other confidential and/or proprietary information, to the extent those requirements are permissible under applicable law.
To all Staffing and Recruiting Agencies: Our Careers Site is only for individuals seeking a job at Qualcomm. Staffing and recruiting agencies and individuals being represented by an agency are not authorized to use this site or to submit profiles, applications or resumes, and any such submissions will be considered unsolicited. Qualcomm does not accept unsolicited resumes or applications from agencies. Please do not forward resumes to our jobs alias, Qualcomm employees or any other company location. Qualcomm is not responsible for any fees related to unsolicited resumes/applications.
If you would like more information about this role, please contact Qualcomm Careers .