Analysis of Connectivity and Inheritance Path of Traditional Dance Elements Based on Graph Theory

Abstract

With the progress and development of the times, the excavation and protection of national traditional dance is the historical necessity of its development and inheritance. This paper carries out research on the analysis of traditional dance movements using digital technology and motion capture technology. The traditional dance digital inheritance technology path is proposed to decompose the traditional dance into four digital media: model, movement, audio and video, and determine the digital display and inheritance process of traditional dance activities in the diffusion dimension. Based on the segmentation algorithm of graph theory, the traditional dance action image is segmented, the optical flow map is used as the input of this stage, and the entropy value of each optical flow map is calculated so as to realize the extraction of action features. Finally, a dataset of six typical movement combinations of traditional dance was constructed to optimize the traditional dance movements of the trainers. The research results show that the traditional dance dataset is constructed reasonably, the recognition model has good performance for traditional dance classification, and the optimal classification accuracy rate is 0.9645 under the Euler Angle feature, which effectively records and preserves the traditional dance movements, and provides a new way for the connection and inheritance of folk dances.

Keywords

• digital inheritance
• graph theory
• segmentation algorithm
• optical flow graph
• movement features
• traditional dance
• 68t05