Use of AI to Recreate and Repatriate Lost, Destroyed or Stolen Paint- ings: The 1785 Parisian Salon Case Study

Abstract

This study investigates the efficacy of artificial intelligence (AI) in the field of art- work restoration, focusing on lost, stolen, or destroyed artworks. Employing a dual approach that combines traditional manual restoration techniques with advanced generative AI tools, the research centers on a case study of the 1785 Parisian Salon. It specifically examines the recon- stitution of Antoine François Callet's painting, Achilles Dragging the Body of Hector, unveiled alongside Jacques-Louis David's Oath of the Horatii. The study utilizes Easy Diffusion and Stable Diffusion 2.1 technologies for inpainting and colorization processes. These AI tools are employed in concert with manual restoration practices to recreate the Callet painting. The methodology also includes the use of secondary visual materials, such as Pietro Martini's 1785 engraving of the Salon Carré, to inform the AI's trained dataset. The application of generative AI in this context significantly accelerates the restoration process. However, the study identi- fies a critical issue where successive AI-based inpainting iterations lead to a degradation in color fidelity and detail precision. This degradation is evidenced by the emergence of unin- tended artifacts and a loss of visual coherence in the restored images. While AI significantly expedites the artwork restoration process, its integration with manual techniques is crucial to mitigate the loss of artistic detail and color accuracy. The study's findings emphasize the need for a balanced approach that leverages the strengths of both AI and traditional restoration methods. This integrative strategy is essential for preserving the original artistic essence of artworks, contributing significantly to the fields of art restoration and digital humanities.