Gesture recognition based on deep learning

Abstract

Gesture recognition based on deep learning is a rapidly growing field of research and development that has the potential to revolutionize the way humans interact with computers and machines. Gesture recognition involves using algorithms and techniques to interpret human gestures, such as hand and body movements, facial expressions, and vocal intonations, to understand their meaning and intent. Gesture recognition based on deep learning has a wide range of potential applications in fields such as robotics, human-computer interaction, and healthcare. The aim of this dissertation is to design gesture recognition algorithm based on deep learning. Object detection based on convolutional neural network can be divided into one-stage object detection and two-stage object detection. Firstly, this dissertation investigates and introduces literature review, and then based on YOLOV3 and RESNET-50, this dissertation gives the models of two detection methods and the training and testing results of these two models on the dataset.