Xiaohui Yang

TL;DR: This paper presents a method for gesture recognition based on data glove that can achieve very high recognition accuracy on the data glove gestures.

TL;DR: Experimental results show that the proposed gaze zone estimation method has a high accuracy, which can be applied in human-computer interaction, and does not need the procedure of calibration.

Abstract: Driver's gaze has become an important indicator to analysis driving state. By estimating the gaze zone of drivers, we can further judge their fatigue state and even predict their driving intention in the next step. In this paper, we propose a four-channel gaze estimation model based on Convolutional Neural Network (CNN), which is used to estimate the gaze zones of the driver. In the proposed method, the images of the right eye, the left eye, the face, and the head are used as the input data of the multi-channel CNN. Then, the features of different channels are fused to estimate the gaze zone. Finally, we compared our method with several existing methods, and the experimental results show that the accuracy of our method is 96%.

TL;DR: The results prove that this system can guide users in completing their experiments, and can effectively reduce the user load during the interaction process and improve the efficiency.