Artworks classification

In this work I will try to solve artist classification problem based on "Best Artworks Of All Time" dataset using TensorFlow 2.0.

Content

• artworks.ipynb: jupyter notebook containing all work
• artworks.pdf: copy in pdf format for instant view

Dataset description

After being challenged many times by my girlfriend about who is the best to guess the painter, I decided to use the power of machine learning to defeat her. I gathered a collection of artworks of the 50 most influential artists of all time. I added a dataset with basic information retrieved from wikipedia. I planned to create a convolutional neural network to recognise the artists looking the colors used and the geometric patterns inside the pictures.

Dataset content

This dataset contains three files:

• artists.csv: dataset of information for each artist
• images.zip: collection of images (full size), divided in folders and sequentially numbered
• resized.zip: same collection but images have been resized and extracted from folder structure

Use resized.zip allows you to download less data and process faster your model.

Inspiration

My goal is to create a model that learn to identify the artist analysing new pictures. I hope to learn new techniques or see some interesting usage of this data.