Article: Ethnic dance movement recognition based on motion capture sensor and machine learning Journal: International Journal of Information and Communication Technology (IJICT) 2024 Vol.25 No.8 pp.81 - 96 Abstract: To address the shortcomings of the existing folk dance movement recognition techniques in terms of accuracy, real-time and generalisation ability, this paper proposes an innovative research method, i.e., combining the motion capture technology with the machine learning algorithm 3D convolutional neural network. This paper describes in detail the various aspects of the research method, including the acquisition of motion data, preprocessing, feature extraction and selection, and the construction and training of machine learning models. Meanwhile, we adopt a variety of high-precision sensors to accurately capture the details of the dancer's movements, and conducts in-depth learning and analysis of the movement features by the 3D convolutional neural network in machine learning. Finally, the experimental results show that the method proposed significantly exceeds the traditional method in terms of the accuracy, robustness, and real-time performance of folk-dance movement recognition, which proves the effectiveness and superiority. Inderscience Publishers - linking academia, business and industry through research

Title: Ethnic dance movement recognition based on motion capture sensor and machine learning

Authors: Mengying Li

Addresses: School of Music, Suihua University, Suihua 152000, China

Abstract: To address the shortcomings of the existing folk dance movement recognition techniques in terms of accuracy, real-time and generalisation ability, this paper proposes an innovative research method, i.e., combining the motion capture technology with the machine learning algorithm 3D convolutional neural network. This paper describes in detail the various aspects of the research method, including the acquisition of motion data, preprocessing, feature extraction and selection, and the construction and training of machine learning models. Meanwhile, we adopt a variety of high-precision sensors to accurately capture the details of the dancer's movements, and conducts in-depth learning and analysis of the movement features by the 3D convolutional neural network in machine learning. Finally, the experimental results show that the method proposed significantly exceeds the traditional method in terms of the accuracy, robustness, and real-time performance of folk-dance movement recognition, which proves the effectiveness and superiority.

Keywords: motion capture; machine learning; folk dance; motion recognition; deep learning.

DOI: 10.1504/IJICT.2024.142296

International Journal of Information and Communication Technology, 2024 Vol.25 No.8, pp.81 - 96

Received: 30 Aug 2024
Accepted: 16 Sep 2024
Published online: 17 Oct 2024 *

Title: Ethnic dance movement recognition based on motion capture sensor and machine learning

Keep up-to-date