Research on Intelligent and Innovative Teaching Mode of Dance Discipline in Colleges and Universities under the Background of Informatization

In this paper, we first count the dance action motion information and compute different lengths of dance action videos through the threshold of segmentation as the input of a convolutional neural network. Combined with the two-layer level optimization function, the dance action trajectory values can be solved. The dance action trajectory values are compared with the standard action in terms of point distance, which suggests the location of the error for dance teaching assistance training. Finally, based on the three-dimensional data capture to establish a three-dimensional dance teaching model, the dance student’s various parts of the movement error results only between 0.25-2.21, the movement trajectory and the Kinect standard movement trajectory curve is highly fitted, and the dance amplitude, dance strength, dance consistency, dance standardization degree of performance is excellent. © 2023 Zhiwei Liu, published by Sciendo.