L’IA per la generazione di immagini archeologiche

Giulia Ghiglia

«Archeologia e Calcolatori» 2025, Supplemento 13, 313-324; doi: 10.19282/acs.13.2025.22

Abstract

Artificial Intelligence (AI) has become an integral part of daily life, with applications expanding rapidly. This article focuses on AI’s ability to generate images for archeological purposes, delving into the technical aspects of image generation using deep learning and machine learning technologies. Midjourney and Playground AI, two popular web applications for image generation, are examined in detail, highlighting their functionality and interface differences. The study applies these tools to reconstruct scenarios from the Cistercian Abbey of San Galgano, assessing their accuracy and limitations. While AI-generated reconstructions offer rapid and cost-effective visualization, they often lack precision, especially in incorporating specific archaeological details. The article underscores the need for improved databases containing detailed architectural and archaeological elements to enhance AI-generated reconstructions’ accuracy and relevance.

Figures

Preview