The traditional educational approach is unable to foster dancing talent due to the substantial changes in social, economic, and cultural conditions. This research explores the design and implementation of a Deep Learning (DL)-based wearable-device-integrated dance teaching system for better comprehension and utilization of dance movements among students. The technology uses information from wearable sensors to analyze dancers’ movements and physiological reactions. It provides real-time feedback, resulting in a very dynamic learning environment. The application of the emotion-intelligence teaching mode theory with dance education shows the benefits derived from DL in wearables. It is observed that trainers based on the hybrid model, such as Dung Beetle Optimized Dynamic recurrent neural network (DBO-DRNN), enhance students’ overall learning performance by carrying out extensive data analysis, experimental research, and physical theory. Data are gathered in the form of dance motions, and physiological data are gathered by wearable sensors across multiple dance sessions; variables such as movement angles, speed, and heart rate are captured. Data preprocessing involves median filtering to remove the noise and smooth motion sequences by providing clean and accurate inputs for the DL model analysis. The system is tested in a series of dance teaching systems, with results indicating significant enhancements in movement accuracy, rhythm adherence, and student engagement in dance. The findings of the research indicate that integrating wearable technology and DL into dance teaching offers a talented way for effective personalized learning and emotional engagement.
Lu et al. (Tue,) studied this question.