Leap Motion Hand Gesture Recognition Based on Deep Neural Network

This paper proposes a new gesture recognition system based on Deep Neural Network (DNN) and Leap Motion. The palm model is reconstructed to obtain the feature data, and then the feature data of all experimenters are obtained by the Leap Motion controller. The data are finally put into the DNN model for training to implement the recognition of specific hand gesture after being normalized. By testing with 30000 gesture frames of five volunteers, it is found that the recognition accuracy of the proposed scheme can reach 98%, which indicates that the scheme can complete the specific gesture recognition with a high average recognition rate.