Dance Emotion Characteristic Parameters Based on Deep Learning Model

Abstract

Abstract In this paper, the emotions of dancers are identified in combination with the integrated deep-learning model. Firstly, four initial value features with important emotional states are extracted from the time, frequency, and time-frequency domains, respectively. It was isolated using a deep belief network enhanced by neuro colloidal chains. Finally, the finite Boltzmann criterion integrates the features of higher abstractions and predicts the emotional states. The results of DEAP data show that the correlation between EEG channels can be discovered and applied by glial chains. The fused deep learning model combines EEG emotional features with temporal, frequency, and expressive qualities.