Web Based Neural Style Transfer

This is an example of using running Neural Style Transfer on the browser. This is possible through Tensorflow.JS (Machine Learning in JavaScript. The objective of the Neural Network is to create an image that takes the style of an artwork (Style Image) or photo and applies that to a photo of your choice(Content Image).The Nice thing about Tensorflow.JS is that the model is ran entirely on your browser! This means that your data and photos never leaves your computer and the model is ran entirely locally on your machine. There is a number of applications for this app, such as creating and testing new fashion designs and color schemes, quickly converting images to new styles and creating a distinct artstyle for a project. Credits for this projects goes to Reiichiro Nakano and Bensonruan tutorials. Also Credits to Authors of Arbituary Neural Style Transfer, Magenta Repository, Tensorflow.Js Library. This was really interesting to try out!