This project is all about a Machine Learning image classification model able to discern between figurative and abstract artworks of my Grandpa's artworks. Its utility extends significantly for previous clients who have acquired artworks in the past and want to obtain specific details, such as the temporal range in which a particular painting was created.
The operational concept is this one: past clients can upload a photograph of their artwork. Subsequently, the uploaded image undergoes a Principal Component Analysis (PCA) dimensionality reduction function. Then a Logistic Regressor can determine whether the resultant data point resides above or below the hyperplane that delineates abstract from figurative artworks.
This process not only enriches the client's understanding of the artwork but also adds a sophisticated layer to the intersection of art and technology, where algorithms decipher the temporal and stylistic dimensions encapsulated within each masterpiece.
In the future, this model will be seamlessly integrated into the official website www.antoniopasciuti.com. Clients and art enthusiasts will have the opportunity to access this powerful tool directly on the website. By simply navigating to the website, users can leverage the model's capabilities.
For a more in-depth exploration of the technical details behind the construction of this Machine Learning model, you can refer to the comprehensive PDF presentation. This document provides a detailed overview of the step-by-step process involved in building and fine-tuning the model, offering a deeper dive into the underlying technical aspects involved.