Thesis – Task-Aware Joint Optimization for Efficient Edge Face Recognition

Face recognition for intercoms and access control must remain reliable across changing imaging conditions despite tight computing and energy constraints. This thesis investigates how image front-end processing and recognition networks can be optimized together to retain identity-relevant information while avoiding unnecessary processing. The aim is to develop and evaluate an efficient recognition framework for resource-constrained edge devices, targeting FPGA-based systems.

The assignment

• Develop a task-aware framework that jointly optimizes front-end image processing and a face-recognition network.
• Tune front-end processing to retain information needed to distinguish identities while reducing unnecessary image detail and processing overhead.
• Combine the optimized front end with a lightweight, quantized recognition network.
• Evaluate recognition reliability across imaging conditions, computational complexity and energy consumption, with implementation targeting FPGA-based edge hardware.

Who we are looking for

• Bachelor’s or master’s student in computer science, software engineering or a related discipline.
• Enrolled at a Swedish university or university college, either directly or through an exchange programme.

What we offer

• An open, friendly and collaborative working culture at Axis Communications.
• An environment that encourages new ideas and ongoing development of skills and working methods.

Practical details

Location: Lund. Suitable for one or two master’s students, or two bachelor’s students. Apply through the thesis advertisement by 27 November 2026; applications may be written in Swedish or English. Include your CV and university grade summary. For a two-student project, apply as a pair but submit separate applications, each naming the other applicant. Questions: Marie Åkesson, +46 272 1800, marie.akesson@axis.com.