Dance movement reconstruction via a 2D matrix model and Haute Monde Chimp Optimization

Current methods for recording dance movements are hampered by complicated recording procedures and inevitable gaps in the data caused by factors such as body structure or clothing. Restoring motion from impaired observations and the recreation of the entire motion sequence is difficult because of the inherent nonlinearity and kinematic characteristics of the movements. Prior methodologies have been shown to have potential for recording motion in a restricted temporal context. The standard multidimensional matrix processing paradigm has no theoretical guarantees for nonlinear motion information retrieval. Thus, Multidimensional Matrix-Calculation (MMC) model has been created to overcome this limitation to provide the restoration and rehabilitation of human complex and unpredictable movements. In addition, a new Haute Monde Chimp Optimization Algorithm (HMCOA) is developed to optimize the parameters of MMC, thus increasing its capability to reproduce accurate intricate dance motions. The paper is published with detailed experimental results of the model compared to other methods such as recurrent connection neural networks (RCNN), low-rank matrix fulfillment (LRMF), multimodal data and Kinect sensors (KISE). The 3D graphs indicate that MMC-HMCOA is better than other approaches in all of the measured performance measures and node configurations, with a 37% higher Performance Ratio (PR) than the best baseline, MMC, 3.3% higher Accuracy (Acc), 40% lower Error Rate (ER) and 2.1% higher AUC and 4.3% higher Recall improvements at 60 nodes.