Comparative Analysis of Image Style Transfer Based on Deep Learning

In recent years, image style transfer has become one of the research fields in computer vision that has received extensive attention. With the development of deep learning, remarkable results have been achieved in image style transfer. This paper conducts a comprehensive comparative analysis of the existing research methods of image style transfer based on deep learning to promote the development of this field. The methods of image style transfer based on deep learning are classified, and two major categories of style transfer methods based on convolutional neural networks (CNNs) and generative adversarial networks (GANs) are systematically reviewed. The CNN methods based on image iteration and model iteration, as well as the supervised and unsupervised GAN methods, are respectively introduced. Moreover, the different characteristics and performance indicators of these four types of image style transfer methods are compared and analyzed, making important contributions to the future development and application of image style transfer in deep learning.