As machine learning models become larger and more complex, they increasingly demand substantial computational resources and memory for training and deployment. Resource-efficient machine learning algorithms address these challenges by optimizing performance for systems with limited resources. With the increasing popularity of wearable devices such as AR glasses, deploying deep learning models on-device has become essential for enabling real-time and privacy-preserving intelligent applications. However, these wearable systems are often constrained in terms of compute, energy, and memory. Subspace-configurable networks (SCNs) provide a promising solution by enabling dynamic model reconfiguration to adapt to various input distributions (e.g., lighting, perspective, environment) without increasing model size.
In this project/thesis, students will explore and implement a novel SCN-based visual model deployed directly on the Brilliant AR Glasses. The goal is to demonstrate efficient on-device adaptation using SCNs, pushing the boundaries of intelligent AR applications under resource constraints.
Interested? Please contact us for more details!
Download as PDF
Student Target Groups:
- Students of ICE;
- Students of Computer Science;
- Students of Software Engineering.
Thesis Type:
- Master Thesis / Master Project
Goals and Tasks:
- Design or select a suitable visual task (e.g., classification, segmentation, gesture recognition) as a demonstration use case;
- Implement and evaluate SCN-based adaptation of the selected model under variable real-world conditions;
- Deploy the model on Brilliant AR Glasses using supported toolchains
- Measure runtime, adaptability, and resource usage on-device;
- Present the results and summarize the work in a written report.
Requirements:
- Familiarity with neural networks;
- Programming skills in Python and C++;
- Experience with deep learning frameworks (e.g., PyTorch or TensorFlow);
- (Optional) Knowledge of embedded Android development or interest in mobile hardware acceleration.
Used Tools & Equipment
- Brilliant AR Glasses developer kit;
- A computation cluster of TU Graz.
Start:
Contact: