Designing Interactive Visuals for Dance from Body Maps: Machine Learning and Composite Animation Approaches
Nuno N. Correia, Raul Masu, William Primett, Stephan Jürgens, Jochen Feitsch, Hugo Silva
Abstract
There is a growing interest in interactive visuals for dance performance. Recent research has identified potential in using interactive visuals to convey to the audience otherwise non-visible elements of performances. Informed by soma design, and with a co-design perspective, we aim to make apparent non-visible bodily aspects of dancers. We propose to design interactive visuals from body maps, following two approaches – Machine Learning and Composite Animation. We conducted a multi-stage study involving 12 dancers. We present and discuss the results of our evaluations, confirming that both prototypes were successful in addressing our aim, with some limitations. We discuss our two approaches, different uses and actors in different stages, tensions between research and dance creation, and potential applications. Our main contributions are the two approaches for designing interactive visuals from body maps and their analysis. These are materialized in two software systems released as open-source and in their design framework descriptions.