An Improved Gesture Recognition Model Based on Mini-Xception

Abstract

With the rapid development of artificial intelligence technology, gestures have become the mainstream in the field of human-computer interaction because of their simplicity, easy understanding and non-contact. Compared with the early data gloves, the vision-based non-contact gesture recognition interaction method has obvious advantages. However, the variability of the gesture itself, the complexity of the background and the influence of different lighting conditions have impacted the accuracy of gesture recognition. With the rapid development of deep learning technology, gesture recognition has achieved amazing results in accuracy. To improve gesture recognition accuracy in complex background, this paper optimizes the mini-xception network model of lightweight convolutional neural network algorithm, introduce the transfer learner and integrate YOLOV4-tiny into the mini-xception model, generate a new network model YT_ mini-Xception. After experimental verification, by YT_ Mini-xception network model, the average accuracy of 0-9 gesture recognition on complex background data set is 96.64%, the average recognition time was 39.8 milliseconds, and the expected goal was achieved.