As Austria's largest research and technology organisation for applied research, we are dedicated to make substantial contributions to solving the major challenges of our time, climate change and digitalisation. To achieve our goals, we rely on our specific research, development and technology competencies, which are the basis of our commitment to excellence in all areas. With our open culture of innovation and our motivated, international teams, we are working to position AIT as Austria's leading research institution at the highest international level and to make a positive contribution to the economy and society.

Our Center for Vision, Automation & Control located in Vienna invites applications for a master’s thesis position. At the Center for Vision, Automation & Control our research unit “Assistive & Autonomous Systems” closely works with industrial partners in order to develop safe and reliable technology components for assistance systems, that can be used in various areas of applications, for e.g.: Railway / Public Transport, Construction Industry, Logistics & Transportation, Agriculture & Forestry as well as in Aviation.

Image descriptors are a fundamental technology in computer vision, driving applications such as 3D reconstruction, SLAM, visual localisation, and more. Recent advances in Deep Learning (DL) have led to significant improvements over traditional hand-crafted methods, offering enhanced robustness and adaptability. However, these advancements come with notable challenges. Deep learning-based descriptors demand high computational resources, limiting their suitability for deployment on resource-constrained edge devices. Moreover, recent research has highlighted privacy vulnerabilities in both traditional and DL-based descriptors.

• You closely work with our researchers and engineers, who focus on the development and deployment of assistive and autonomous systems, particularly in the area of robust and accurate localisation and positioning.
• Under the guidance of our experts, you contribute to the development of cutting-edge methodologies in deep learning-based image descriptors, balancing efficiency, and security on resource-constrained edge devices.
• You investigate state-of-the-art deep learning advancements in descriptor optimisation and privacy-aware techniques, identifying gaps and proposing innovative solutions.
• You apply your findings to practical challenges, designing and implementing a proof-of-concept demonstration to validate your methods.

• You design, implement, and optimise deep learning models to extract robust and privacy-preserving image descriptors efficiently on resource-constrained devices.
• You benchmark and evaluate model performance, improving robustness, efficiency, and privacy-preserving capabilities across varied scenarios.
• You exchange insights with interdisciplinary and international experts in machine learning and computer vision, contributing to a dynamic and collaborative environment.
• You document your results, with the potential to present your work in conference proceedings or publish it in peer-reviewed journals.