JAEarly / Online-Style-Interpolator

An online deep learning image style interpolator.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

University of Southampton

COMP6248 Reproducibility Challenge

A Learned Representation For Artistic Style

Cover Photo

OpenReview URL of Original Paper: https://openreview.net/forum?id=BJO-BuT1g&noteId=BJO-BuT1g
Medium Article: https://link.medium.com/UzJzB7Ah5X
Demo Video: https://youtu.be/h-UpR92l1DI

Team Members

Joseph Early (je5g15)
Jamie Sian (js17g15)
Alexander Stonard (ads1g15)

Overview

This is a reimplementation of A Learned Representation For Artistic Style (Dumoulin et al. 2016).
It also includes an extension that wraps the Style Interpolation feature as a web application.

Setup

Python 3.5.
All of the required libraries are in requirements.txt
To run the web application, run src/web/app.py and open http://127.0.0.1:5000/ (or whatever the console output says!).

About

An online deep learning image style interpolator.


Languages

Language:Python 92.1%Language:JavaScript 4.0%Language:HTML 2.7%Language:CSS 1.3%