Publication

Abstract : Computer vision systems have embraced learning using networks in recent years. On the other hand, unsupervised learning with convolutional neural networks has received less attention. The proposed method will help to close the gap between convolutional networks’ performance and that of other machine learning algorithms. The goal of this paper is to use deep convolutional generative adversarial networks, a type of convolutional neural network, to create fake face images. The research demonstrating deep convolutional generative adversarial networks outperform generative adversarial networks is used in this paper. By training on picture datasets, the deep convolutional adversarial pair learns a series of representations in both the discriminator and generator leading to realistic face images. At the end of this paper, the results obtained by using deep convolutional generative adversarial networks are shown.