Biomechanical analysis of the effects of breathing techniques on dance performance and dancers’ physiological state

Abstract

Listen

Translate article

This study aims to investigate the effects of different breathing techniques on the physiological state and expressive force of modern dance dancers. Here, a motion recognition model based on a Three-Dimensional Convolutional Neural Network (3D CNN) and a Transformer network is proposed to recognize dancers’ movement performance under diverse breathing patterns. The study employs high-frequency motion sensors and physiological monitoring devices, combined with questionnaires and open datasets, to collect and analyze the dancers’ heart rate, respiratory rate, muscle activation rate, and other data. The results show that under deep breathing conditions, the dancers’ heart rate reaches 0.84, significantly higher than shallow breathing (0.46) and general breathing (0.61). Furthermore, the muscle activation rate is also remarkably increased to 0.95, better than general breathing (0.73) and shallow breathing (0.58). The model proposed in this study has excellent performance on motion recognition, with an accuracy of 96.89% at 0.5 dropout, remarkably exceeding other comparison models. The study concludes that deep breathing can markedly improve the dancer’s physiological activation and performance. Moreover, the proposed model can accurately identify the correlation between breathing patterns and dancers’ movements, providing scientific support for the application of breathing techniques in dance training in the future.