Design and Implementation of Digital Dance Teaching Platform Based on Kinect

This paper combines Kinect human movement recognition, movement comparison technology, and dance teaching to design a digital dance teaching platform based on Kinect. In the human action recognition algorithm, a way is suggested to show joint point angles using fixed axes. This is done by combining the Hidden Markov Model with Hausdorff distance pose recognition. And in the human movement comparison algorithm, a movement comparison analysis method based on the dynamic time regularization algorithm is introduced, and an optimization method based on the improved longest common subsequence algorithm is proposed on this basis. In the application practice of the digital dance teaching platform, the experimental class and the control class were set up with the 2023 dance major students of a sports college in Wuhan City, Hubei Province, as the research object. In terms of dance performance, the average scores of the experimental class in learning, choreography combinations, and theory were 89.23, 83.28, and 84.27, which were higher than the scores of the control class, with P < 0.05, all of which showed significant differences. In terms of independent learning ability and course satisfaction, the mean values of the experimental class were higher than those of the control class and showed significant overall differences (P<0.05). The utility of the Kinect-supported dance teaching platform has been fully demonstrated.