This project is developed to support aculei.xyz mission.
This repository contains the code used to build a dataset from animal photographs taken automatically by hunter cameras located in the forests of Umbria. Building the dataset aims to collect a large amount of data so that it can feed artificial intelligence techniques in order to group and potentially classify the photographs. Two approaches are explored: the first involves unsupervised learning based on clustering algorithms, while the second relies on using a technique called zero-shot image classification in order to classify photographs according to the animal portrayed. In recent years, the discipline known as computer vision has made huge strides, thanks to this branch of computer science it is possible to manipulate complex data such as photographs by extracting knowledge from them. The information obtained from photographs, relationships and hidden patterns that are identified by artificial intelligence are used to generate an interactive experience within a photographic archive called aculei. Aculei project is the brainchild of a photographer from Milan, Italy, who deployed several automated hunter cameras in the woods surrounding the house where he grew up in Umbria. The desire to publish the photographs and make the experience on the archive ai-driven are the motivations that led him to conceive the aculei project.