AI Art Critic: Artistic Classification of Poster Images using Neural Networks

Artificial Intelligence (AI) techniques have successfully demonstrated reliable outcomes for assessing visual aesthetics. Throughout this work, we examine the visual art aesthetics classification problem, often known as automatically categorizing a picture as having a low or high aesthetic value in digital and hand-drawn posters, which is a complex challenge exceeding the domain of image recognition. We focus on configuring a system to judge digital art posters without any human intervention. A Convolutional Neural Network (CNN) model with a self-assembled and classified dataset is trained. The images of our dataset come from a variety of artists, preventing any bias. Our model was able to predict the classes with an accuracy percentage of 89%.