/MLH2021_AAST

Andrew's submission for Major League Hacking (MLH) 2021 New Year Hacking. Coded by Andrew Chen.

Primary LanguageJupyter Notebook

MLH2021_AAST

AAST- Andrew's Art Style Transfer

Intro

Art Style Transfer is a hot topic in recent years, accompanied with the prosperity of AI. Generally, two images are needed. One is the content image, and one is the style reference image. The output images should be a blend of the two images.

A popular method is deploying CNN (Convolution Neural Network). And that's what the operator wants to use in this project.

Methodology

The operator wants to build a project to analyze the features and transfer the art style from one to another.

Example

Content Image:

black-swan

Style Image:

MonetLotus

Blended:

BlendedSwan

Note:

This project referred much to the TensorFlow open source project. The link is here: https://github.com/tensorflow/models/blob/master/research/nst_blogpost/4_Neural_Style_Transfer_with_Eager_Execution.ipynb

The student also referred to this article on Medium: Neural Style Transfer: Creating Art with Deep Learning using tf.keras and eager execution, by TensorFlow.

Some bugs are not fixed completely, like how to cast uint8 into float without influencing loading images. I will keep trying to solve them.

Thank you for your time!