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

Category

Computer Science, Software Engineering

Scope

1-2 Master thesis students or 2 Bachelor thesis students

Background

Face recognition in intercoms and access control systems requires reliable performance under varying imaging conditions while operating within limited computational and energy budgets. Recent studies have explored spectral convolution for efficient front-end processing directly in image sensors [1] and sub-byte quantization to reduce the computational requirements of mobile face-recognition networks [2]. However, the joint optimization of front-end processing parameters and face-recognition networks remains underexplored, particularly in terms of selecting task-relevant image information to balance recognition of reliability and computational efficiency.

Goal

The goal of this project is to develop a task-aware joint optimization framework for front-end processing and face recognition on resource-constrained edge devices. The front end will be optimized to preserve essential identity-discriminative information while reducing unnecessary image detail and processing overhead. In combination with a lightweight, quantized recognition network, the proposed approach will be evaluated in terms of recognition reliability under varying imaging conditions, computational complexity, and energy consumption, with implementation targeting FPGA-based edge systems. 

Who are you?

Master's or Bachelor student in Computer Science, Software Engineering, or a related field.

OK, I am interested! What do I do now?

You are valuable to us – how nice that you are interested in one of our proposals! There are a few things for you to keep in mind when applying.

• Applications are accepted in both Swedish and English, and you apply via the proposal advert.
• The announced thesis is open only to students affiliated with a Swedish University/College either directly or via an exchange program.
• When the thesis proposal states that it includes two students working together, we would like you to apply in pairs. In these cases, send one application each but make sure to clearly state in your application who your co-applicant is. If you have any questions regarding this, please do not hesitate to contact us.
• Please attach your CV and University/College grade summary.

Who to contact for any questions regarding the position!

Marie Åkesson on +46 272 1800 or marie.akesson@axis.com

References

1 Cui, Kaiyu, et al. "Spectral convolutional neural network chip for in-sensor edge computing of incoherent natural light." Nature Communications 16.1 (2025): 81.

2 Bunda, Sebastian, Luuk Spreeuwers, and Chris Zeinstra. "Sub-byte quantization of mobile face recognition convolutional neural networks." 2022 International Conference of the Biometrics Special Interest Group (BIOSIG). IEEE, 2022.