Research and Implementation of Specific Action Generation Model in Dance Video Based on Deep Learning Technology

In recent years, with the development and popularization of deep learning, artificial neural networks have been successfully applied to the creation of dance movements. The important advantage of using deep learning to create dances is that they can extract advanced features from raw data (sounds, motion log data, etc.). The purpose of this article is to study the specific action generation model in dance videos based on deep learning technology. This paper designs a dance creation algorithm based on deep learning to extract the correspondence between voice and action features. In view of the dance scene, this article analyzes the generation of the dance movements of the characters on the dance videos, and analyzes the joint angles with large differences in some intervals, so as to provide a reference for the learning and improvement of movements. Experimental research shows that the ConvNet-based authentication method proposed in this paper is 2.7 times faster than the traditional LSTM-based authentication method.