Indicative thesis title: What just happened? Enhancing video analysis for music therapy with Machine Learning.
Video analysis can be an effective tool for music therapy practitioners and researchers but is difficult and time-consuming. This research investigates how the workload might be reduced through the automated detection and analysis of events and client behaviours of interest, with a particular focus on the exploitation of Machine Learning technology.
The project will deliver a prototype software tool incorporating a library of models, each designed to address a specific type of event or behaviour such as the client looking at, approaching, or moving in synchrony with the therapist. The modular approach aims to facilitate post-PhD enhancement by the author and/or others.
Richard is a retired Chartered Engineer. In a career spanning nearly 40 years, he undertook a wide range of technical and leadership roles in domains including software development, reliability assessment and process modelling.
Despite many successful projects, numerous innovative software tools and several published papers, there remained a perception of unfulfilled potential. This, combined with a desire to use his acquired knowledge, in some small way, for the good of humanity, led to the aspiration to undertake a postgraduate research project on a topic which combines two long-held interests: machine learning and music.