유아의 발레체험을 위한 인공지능 튜터 시스템 기획

Abstract

This study proposes a non-face-to-face ballet experience system for toddlers according to the changing educational environment. The system proposed in this study is to encourage toddlers to naturally take in ballet through avatars by recognizing their poses based on machine learning, which in turn providing similar ballet movements demonstrated by artificial intelligence tutors. In order to plan this, the purpose of the study was to look at non-face to-face cases of early childhood dance education and to suggest directions and detailed procedures for developing an artificial intelligence tutoring system for youngsters. As a result, related research data emphasized the datafication of dance movements, hyper-connected, superintelligence, surreal ballet practical education blended with ICT technology, and the importance of rhythmic robot and infant interaction. Existing cases of non-face-to face ballet education were studied, which were found to include augmented reality-based creative ballet education and AI-based classic ballet education. Based on the results of related studies and case studies, the running process of this system consisted of first, avatar generation, second, interaction with artificial intelligence tutor, third, pose recognition of kids, fourth, ballet movement demonstration and dancing, and fifth, video watching. In addition, a guide for implementing the service was presented. Through the system proposed in this study, it will be possible to provide practical guidance on the development of dance experience and educational system related to this in the educational industry in the future, and further contribute to the sound growth of toddlers and expansion of ballet art.