Technical Note: Quantifying music-dance synchrony during salsa dancing with a deep learning-based 2D pose estimator.
Journal:
Journal of biomechanics
PMID:
35907291
Abstract
Dance interventions hold promise for improving gait and balance in people with neurological conditions. It is possible that synchronization of movement to the music is one of the mechanisms through which dance bestows physical benefits. This technical note will describe a novel method using a deep learning-based 2D pose estimator: OpenPose, alongside beat analysis of music to quantify movement-music synchrony during salsa dancing. This method has four components: i) camera setup and recording, ii) tempo/downbeat analysis and waveform cleanup, iii) OpenPose estimation and data extraction, and iv) synchronization analysis. Four participants performed a solo basic salsa step continuously for 90 s to a salsa track while their movements and the music were recorded with a webcam. Two conditions were recorded for each participant: one in which they danced on the beat of the music and one where they did not. This data was then extracted from OpenPose and analyzed. Median asynchrony values highlighted differences between participants with and without dance training and between on- and off-beat conditions, indicating that this method may be an effective means to quantify a dancer's asynchrony while performing a basic salsa step.