A deep learning-based method for identifying differences in folk dance movements.

This article focuses on the development and application of a dance movement differential recognition system combining the Azure Kinect dynamic body sensor and a hidden Markov model (HMM) algorithm to support the preservation, teaching, and dissemination of Uyghur folk dance. The study involved capturing and analyzing key limb and trunk angles of typical Uyghur dance movements performed by professional dancers and beginners, revealing measurable differences that guided targeted training. After three months of systematic practice using feedback from the model, beginner dancers showed significant improvement in movement accuracy, and a majority expressed satisfaction with the system's effectiveness. The research highlights the potential of integrating deep learning and advanced motion capture technology to digitize and standardize folk dances, offering a scalable framework for cultural preservation and dance education. Future directions include exploring newer algorithms, additional sensing technologies, real-time feedback, and expanding to other traditional dances to enhance motion recognition and learning experiences.