Intelligent Dance Teaching Support System based on 2D Convolutional Neural Networks

The development of artificial intelligence has transformed a number of industries, including the performing arts and education. This project investigates an AI-based intelligent dance teaching support system that combines 2D convolutional neural networks and few-shot learning. The technology is intended to assess and analyse dance moves while giving students immediate feedback. Few-shot learning makes the model extremely effective for dancing forms by allowing it to adjust to different dance styles with little training data. To provide accurate posture estimation and movement recognition, features are extracted from motion sequences using the 2D CNN architecture. The proposed method dramatically increases learning accuracy and efficiency in dance training, according to experimental results. Through intelligent feedback and adaptive learning methods, the study demonstrates how AI technologies may improve traditional dance education. The recommended approach achieves the maximum accuracy in the Intelligent Dance Teaching Support System, according to the performance metrics, with an accuracy score of 98.2%, a precision score of 98.12%, a recall score of 98.11%, and an F1 score of 98.12%. The study highlights how crucial this framework is for creating intelligent dance teaching systems and differentiating dancing styles.