Supervisor(s): Dr Alina Bialkowski

Modern AI models consume a vast amount of resources, yet many valuable applications run in resource-constrained environments. Compression techniques help address this, but it’s poorly understood how they impact a model’s decision-making process. This project measures how compression (via knowledge distillation) shifts what image models pay attention to, using a novel evaluation approach.

Supervisor(s): Ethan Jones, Julia Drugova, Thilina Halloluwa, Mashhuda Glencross, Jason Weigel

Network planning aide utilising smart meter data to identify the correct low-voltage phase of customers. Data driven techniques provide visualisation of the combined electrical network topology. Intuitive user interface allows for result analysis and interpretation.

Supervisor(s): Zijian Wang

Wheat management strategies play a critical role in determining yield. Traditional management decisions often rely on labour-intensive expert inspections, which are expensive, subjective and difficult to scale. In this project, we explore vision-language models (VLMs) and agentic AI to enable scalable, data-driven management support.

Supervisor(s): Yanzhuo Yang, Julia Drugova, Thilina Halloluwa, Mashhuda Glencross, Jason Weigel

Rava is a one-of-a-kind telepresence device, which transports remote users into the room with immersive stereoscopic vision, crisp spatial audio and seamless device control from an intuitive web application.