An Exploration of Style Transfer Using Deep Neural Networks
Cameron Y Smith · UNM’s Digital Repository (University of New Mexico) · 2016
Convolutional Neural Networks and Graphics Processing Units have been at the core of a paradigm shift in computer vision research that some researchers have called `'the algorithmic perception revolution." This thesis presents the implementation and analysis of several techniques for performing artistic style transfer using a Convolutional Neural Network architecture trained for large-scale image recognition tasks. We present an implementation of an existing algorithm for artistic style transfer in images and video. The neural algorithm separates and recombines the style and content of arbitrary images. Additionally, we present an extension of the algorithm to perform weighted artistic style transfer.