In this project, we built two deep learning algorithms on a dataset of Artworks for the most famous artists, which are CNN and GAN. The goal is to generate new images of Artworks that look as realistic as possible. Also, predict the artwork belongs to which artist.
We divided the project into two sections, Artworks Generation and Artworks Prediction.
The dataset is called Best Artworks of All Time, and contains a collection of paintings of the 50 most influential artists of all time.
You can find and download the dataset from here
Note: This dataset is more complex than the number datasets (like MNIST or SVHN) that we usually work with, and so, it needs deeper networks and training for a longer time to get good results.
- Generate new artworks.
- Predict the artwork belongs to which artist.