Week 2# Identification of Artists and Movements from Paintings with Machine Learning
This week, we will elaborate on the classes that we have selected for classification, how we decided them and the process of collecting data.
An art movement is a tendency or style in art with a specific common philosophy or goal, followed by a group of artists during a specific period of time. -Wikipedia
For the classification of paintings, we focused on six different art movements: Cubism, Symbolism, Impressionism, Neoclassicism, Pop Art, Abstract and set of paintings from twelve different artists: Wassily Kandinsky, Jackson Pollock, Claude Monet, Pierre Auguste Renoir, Jean Auguste Dominique Ingres, Jacques-Louis David, Pablo Picasso, Fernand Leger, Roy Lichtenstein, Andy Warhol, Mikhail Vrubel, Odilon Redon.
For successful classifications, large number of training examples is required to train our model. So, in the process of choosing the classes, this fact is going be taken into consideration.
To collect our dataset, we will create a scraper and scrape the paintings with that scraper from wikiart.org. In total, we expect to gather close to 150 paintings per artist and a total of 1800 training examples.
Prior to building and developing our model, we aim to utilize Naive Bayes classifier to obtain the most effective features for our dataset. By this means, it is more likely for our model to achieve high accuracy under the restrictions.
Stay tuned.
