<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>With the development of digital technology, digital dance movement recognition and correction has become a hot research topic. This study focuses on a movement recognition system that incorporates human skeletal information and aims to improve dance movements’ accuracy and correction efficiency. This study develops a digital dance movement recognition and correction system that combines human lean information. This system can recognize and correct dance movements accurately by improving the spatio-temporal graph convolutional network (ST-GCN). The study employs a spatio-temporal graph representation of the human skeleton and spatial graph convolution technique, and enhances the system’s movement recognition capability through an adaptive graph convolution module. The experimental results show that the system achieves an average accuracy of 99.3% in dance movement recognition, 82% and 92% on the publicly available datasets UTKinect and MSRAction3D, respectively, and 95% on the dance movement dataset ETHDance constructed in this study. In addition, by introducing channel, spatial and temporal attention mechanisms, the system also shows high efficiency in dance movement correction. For the correction test of 10 basic dance movements, the correction accuracy is more than 95%, significantly higher than the 69.91% of the traditional method. This study improves the accuracy of dance movement recognition and provides adequate technical support for dance teaching and practicing.</p>
</abstract>

With the development of digital technology, digital dance movement recognition and correction has become a hot research topic. This study focuses on a movement recognition system that incorporates human skeletal information and aims to improve dance movements’ accuracy and correction efficiency. This study develops a digital dance movement recognition and correction system that combines human lean information. This system can recognize and correct dance movements accurately by improving the spatio-temporal graph convolutional network (ST-GCN). The study employs a spatio-temporal graph representation of the human skeleton and spatial graph convolution technique, and enhances the system’s movement recognition capability through an adaptive graph convolution module. The experimental results show that the system achieves an average accuracy of 99.3% in dance movement recognition, 82% and 92% on the publicly available datasets UTKinect and MSRAction3D, respectively, and 95% on the dance movement dataset ETHDance constructed in this study. In addition, by introducing channel, spatial and temporal attention mechanisms, the system also shows high efficiency in dance movement correction. For the correction test of 10 basic dance movements, the correction accuracy is more than 95%, significantly higher than the 69.91% of the traditional method. This study improves the accuracy of dance movement recognition and provides adequate technical support for dance teaching and practicing.

Digital Dance Movement Recognition and Correction System Combining Human Skeletal Information

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

With the development of digital technology, digital dance movement recognition and correction has become a hot research topic. This study focuses on a movement...

Digital Dance Movement Recognition and Correction System Combining Human Skeletal Information

Pubblicato online: 26 feb 2024

Pagine: -

Ricevuto: 17 gen 2024

Accettato: 27 gen 2024

DOI: https://doi.org/10.2478/amns-2024-0686

Parole chiave
Digital dance movement recognition, Human skeletal information, Spatio-temporal graph convolutional network, Movement correction

© 2024 Ke Li, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Digital Dance Movement Recognition and Correction System Combining Human Skeletal Information

Pubblicato online: 26 feb 2024

Pagine: -

Ricevuto: 17 gen 2024

Accettato: 27 gen 2024

DOI: https://doi.org/10.2478/amns-2024-0686

Parole chiaveDigital dance movement recognition, Human skeletal information, Spatio-temporal graph convolutional network, Movement correction

© 2024 Ke Li, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Parole chiave
Digital dance movement recognition, Human skeletal information, Spatio-temporal graph convolutional network, Movement correction