Dance Rehearsal Simulation and Optimization Algorithm Based on Virtual Reality Technology

Abstract

The use of virtual reality (VR) for industrial training helps minimize risks and costs by allowing more frequent and varied use of experiential learning activities, leading to active and improved learning. However, creating VR training experiences is costly and time consuming, requiring software development experts. In this manuscript, Attentive Evolutionary Generative Adversarial Network based Dance rehearsal based on virtual reality technology (DVR-AEGAN-COA)is proposed. The input data are collected from Real Time Data from various public dance dataset. Then the data are fed in to pre-processing using Geometric Interaction Augmented Graph Collaborative Filtering (GGCF) to remove noise and enhancing data. The pre-processed data are given into Adaptive Synchro Extracting Transform (ASET) for Feature extraction to align the dance movements, such features are tempo, beats, and rhythm. Then the extracted features are given into Attentive Evolutionary Generative Adversarial Network (AEGAN) for classification of attributes, styles and emotions. In general, the AEGAN does no express adapting optimization strategies to determine optimal parameters. To ensure accurate classification, the Coati Optimization Algorithm (COA) is introduced for optimizing AEGAN. The proposed DVR-AEGAN-COA is implemented in Python working platform and the analyzed performance metrics likes precision, accuracy, F-score, computational time, specificity and sensitivity. The proposed method attains higher accuracy25%, 29% and 25%, higher specificity 27.32%, 24.43%, 38.24% and higher recall 31.13%, 23.33% and 38.13% than the existing methods like enhanced Dance Rehearsal Virtual Reality through Artificial Neural Network (DVR-ANN), Dance Rehearsal Virtual Reality through Circle Search Algorithm (DVR-CSA) and Dance Rehearsal Virtual Reality through Skill Optimization Algorithm (DVR-SOA) respectively.