The project is to develop a framework for applying machine learning to identify mechanical and electrical building services in photos of existing buildings. This would involve identifying appropriate approaches based on academic literature, collecting sample images, developing a working approach to train the machine learning algorithms, and writing code in C/C++/OpenCL to run the trained algorithms.
EZAMETRIC is Perth-based start-up founded to help designers survey sites efficiently in the Australian and European construction industries. The student’s work will be integrated as enhancements to EZAMETRIC’s early software products, which are due to launch in early 2023.
This CEED project would be most suited to a student wanting to make a difference in an innovative start-up business environment. Students valuing active listening, creativity and a willingness to learn would be ideal candidates to run the project.
Note 1: Building services refers to the HVAC (heating, ventilation and air conditioning), plumbing, electrical, telecommunications and fire suppressant systems in buildings
Note 2: The student must be able to code in C (or C++) to apply (prior experience in OpenCL is not required)