mConduct: A Multi-Sensor Interface for the Capture and Analysis of Conducting Gesture

The art of conducting has a long and well-established history, using physical gesture to convey musical intent and expressions. Conducting relies on visual communications to direct the ensemble as a coherent unit. The aim of this project is to capture and analyse the hand gestures of conducting in order to provide real-time, interactive multimodal feedback in a number of application contexts including visualisation. This paper presents the design and development of the interface involving hardware sensors and software analysis modules, and discusses the application of visualisation for conducting. The paper concludes with latest findings, future directions and the impact the research may have outside the realm of gesture communication application.