Smart glasses can provide users with immediate information about the objects and environments around them. This project explores how wearable cameras, multimodal AI and online information retrieval can be combined to help users look at an object, ask a question and receive a useful response through a smart-glasses interface.
The student will develop a proof-of-concept platform that recognises objects captured by a wearable camera, understands spoken questions, retrieves relevant information from online sources, and presents concise answers through a simple wearable user interface. Example applications may include identifying everyday objects, finding product or location information, providing instructions, and assisting users with visual impairments. The resulting platform will provide a foundation for future smart-glasses applications such as navigation, translation and personalised assistance.
Students should have basic Python programming skills and an interest in computer vision, multimodal AI or human–computer interaction. Experience with machine learning, APIs or UI development is desirable.
School
Computer Science and Engineering
Research Area
Artificial intelligence | Computer vision | Multimodal AI | Human-Computer interaction | Wearable computing
Suitable for recognition of Work Integrated Learning (industrial training)?
No
- Research environment
- Expected outcomes
- Supervisory team
- Reference material/links
The student will work with researchers in artificial intelligence, computer vision and wearable computing. The project will provide access to smart-glasses development hardware, multimodal AI models, web-search services and relevant computing resources. The student will receive regular supervision and gain hands-on experience in platform design, prototype development and experimental evaluation.
- A modular architecture connecting wearable camera input, multimodal AI, web search and user interaction
- A proof-of-concept platform for object recognition, spoken questions and online information retrieval
- A simple wearable user interface, demonstration scenarios, and an evaluation of accuracy, response time and usability
- A software prototype, technical documentation, research report and project poster
