A Survey on Image Style Transfer Approaches Using Deep Learning
Changshen Zhao
Abstract
Abstract Image style transfer has been an important branch of computer vision and image processing. Inspired by the development of deep learning, applications of Convolutional Neural Networks (CNNs) in recomposing content and style of two separated images were proven to be effective by the recent works of image style transfer. This paper provides a survey of image style transfer, which focuses on methods which using deep learning. In addition, this paper also provides information of methods without deep learning, and seeks their points of innovation. In this paper, all references are published ranging from 2001 to 2017, which presents an overview of progress in this realm over the past two decades.
Topics & Concepts
Deep learningStyle (visual arts)RealmComputer scienceConvolutional neural networkTransfer of learningArtificial intelligenceImage (mathematics)Visual artsGeographyArtArchaeologyGenerative Adversarial Networks and Image SynthesisImage Enhancement TechniquesAdvanced Image Processing Techniques