Honours and Masters project

Deep learning has achieved ground-breaking performance in many vision tasks in the recent years. The objective of this project is to apply state-of-the-art visual-language foundation models such as Qwen for medical image analysis and report generation.

Deep learning has achieved groundbreaking performance in many 2D vision tasks in recent years. With more and more 3D data available, such as that captured by LiDAR, the next research trend is to conduct advanced 3D perception and generation tasks. The objective of this project is to study the state-of-the-art 3D foundation models and apply them for tasks that require spatial intelligence.

For decades, the primary way people who are blind or have low vision (BLV) interact with computers has been through screen readers. These linearise an application or document interface, presenting content as synthetic speech or on a single-line braille display, which the user navigates using keyboard shortcuts or touch gestures on a smartphone. Screen readers have a major limitation: they cannot show graphics. Instead, they rely on alternative text descriptions, which are often incomplete or inadequate.

Vision-Language Models (VLMs) are increasingly capable of understanding visual scenes, interpreting natural-language instructions, and reasoning across visual and textual information. These capabilities create new opportunities for intelligent systems that can interact with users and understand their surroundings. However, most state-of-the-art VLMs are designed for cloud infrastructure or high-performance GPU systems, making their direct deployment on embedded platforms challenging because of memory, computation, latency, and energy constraints.

Embodied intelligence represents a rapidly emerging paradigm in robotics where intelligent behaviour arises through the integration of perception, decision-making, action, and interaction within a physical agent operating in the real world. Advances in autonomous robotics, computer vision, sensor fusion, and human-robot interaction have enabled mobile robotic systems to perform increasingly sophisticated tasks in dynamic environments.

Edge intelligence and Tiny Machine Learning (TinyML) have become key enablers for real-time, on-device intelligence in industrial IoT applications, such as predictive maintenance, anomaly detection, and process monitoring. TinyML allows inference to be performed locally, reducing latency, enhancing data privacy, and lowering energy consumption compared to cloud-based solutions.

Time series are an ever growing form of data, generated by numerous types of sensors and automated processes. However, machine learning and deep learning methods for analysing time series are much less advanced than for other forms of data.

Our research is revolutionising the analysis of time series data. But it is early days, and many more impactful challenges are yet to be overcome.

This project is funded by the Australian Research Council and will be conducted as part of a large world-leading research team.

Propositional satisfiability (SAT) is a well-known example of NP-complete problems. Although NP-completeness may be perceived as a drawback, it allows one to solve all the other problems in NP by reducing them to SAT and relying on the power of modern SAT solvers. This is confirmed by a wealth of successful examples of use cases for modern SAT solving, including generalisations and extensions of SAT as well as a wide variety of practical applications in artificial intelligence (AI).

Are you interested in applying your AI knowledge to music, especially classical music? You will explore how AI (and Deep Learning) can be used in music, such as using Generative AI to create Theme & Variations in classical music, analysing classical music structures (e.g., sonata form, theme & variation form, etc.), and identifying instruments (e.g., violin vs. viola).

Are you interested in programming maps, such as GoogleMaps or Open Street Maps? This project uses online maps extensively for visualising routes, and other objects of interest.