Desarrollo de un modelo de reconocimiento de cinco gestos de la mano derecha usando redes neuronales transformers.

Abstract

Hand gestures represent a natural form of communication and device control. In the field of Hand Gesture Recognition (HGR), Electromyography (EMG) is used to detect the electrical impulses that muscles emit when a movement is generated. Currently, there are several HGR models that use EMG to predict hand gestures. However, most of these models have limited performance in practical applications. This study addresses this issue by using transformers to improve performance and mitigate ambiguity in HGR results. The architecture of our model is composed of a Convolutional Neural Network (CNN), a positional encoding layer and the transformer encoder. To obtain a generalizable model, the EMG-EPN-612 dataset was used. This dataset contains records of 612 individuals. The results were compared with previous research that used CNN, LSTM and transformers. The findings of this research reached a classification accuracy of 95.25 ±4.9% and a recognition accuracy of 91.73 ±8.19%.