Title: Artificial intelligence in classifying and creating art: a survey

Authors: Andrej Šimić; Marina Bagić Babac

Addresses: Faculty of Electrical Engineering and Computing, University of Zagreb, Unska 3, HR-10000, Zagreb, Croatia ' Faculty of Electrical Engineering and Computing, University of Zagreb, Unska 3, HR-10000, Zagreb, Croatia

Abstract: The survey investigates research papers on AI-based art classification and creation, analysing their methodologies and outcomes. Categorising the papers by focus areas, such as genre, style, and artist classification, the study assesses the techniques and accuracy achieved. Additionally, it explores art creation techniques, including style transfer, GAN-based image creation, and text-to-image synthesis. This survey stands out by offering a comprehensive overview of AI-based art research, highlighting the originality of approaches and methodologies used in these papers. It showcases the innovative use of AI algorithms and data processing techniques in the art domain. The findings reveal two primary applications of AI in art: automated art analysis for categorisation and the generation of new artwork. AI aids art historians and collectors in identifying artworks based on artist, style, or genre. Moreover, it assists artists in the creative process by inspiring them and generating novel ideas through art creation.

Keywords: artificial intelligence; art classification; art generation.

DOI: 10.1504/IJSPR.2024.137964

International Journal of Student Project Reporting, 2024 Vol.2 No.1, pp.59 - 89

Received: 30 Jul 2023
Accepted: 06 Nov 2023
Published online: 12 Apr 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article