Dynamic Gesture Recognition Using Inertial Sensors-based Data Gloves

Gesture recognition systems provide a natural interface of interaction between humans and computational systems. This study proposes a system for dynamic gesture recognition using inertial sensors-based data gloves. The proposed data gloves that consist of thirty-six inertial measurement units capture the motion of two arms and hands. The multimodal dataset included the sign language information of data gloves and skeletons is built. Then the convolutional neural network structure for sign language recognition named SLRNet is designed. It mainly consists of a convolutional layer, a batch normalization layer, and a fully connected layer. Finally dynamical gesture recognition experiments are implemented to prove the effectiveness of the proposed method.