ArtBrain: An Explainable end-to-end Toolkit for Classification and Attribution of AI-Generated Art and Style

TL;DR

This paper introduces ArtBrain, an explainable toolkit utilizing a novel neural network to detect AI-generated art, attribute it to source models, and distinguish it from human-created artwork, supported by a large dataset and a web app.

Contribution

It presents a new dataset, a novel neural network model, and an accessible web-based tool for detecting and attributing AI-generated art, advancing the field of synthetic artwork authentication.

Findings

AttentionConvNeXt achieved an F1-Score of 0.869 in source attribution.

The model correctly identified AI-generated images with 99% accuracy.

Humans identified AI art with about 58% accuracy, less than the model.

Abstract

Recently, the quality of artworks generated using Artificial Intelligence (AI) has increased significantly, resulting in growing difficulties in detecting synthetic artworks. However, limited studies have been conducted on identifying the authenticity of synthetic artworks and their source. This paper introduces AI-ArtBench, a dataset featuring 185,015 artistic images across 10 art styles. It includes 125,015 AI-generated images and 60,000 pieces of human-created artwork. This paper also outlines a method to accurately detect AI-generated images and trace them to their source model. This work proposes a novel Convolutional Neural Network model based on the ConvNeXt model called AttentionConvNeXt. AttentionConvNeXt was implemented and trained to differentiate between the source of the artwork and its style with an F1-Score of 0.869. The accuracy of attribution to the generative model reaches 0.999. To combine the scientific contributions arising from this study, a web-based application named ArtBrain was developed to enable both technical and non-technical users to interact with the model. Finally, this study presents the results of an Artistic Turing Test conducted with 50 participants. The findings reveal that humans could identify AI-generated images with an accuracy of approximately 58%, while the model itself achieved a significantly higher accuracy of around 99%.