Scientists from the Space Research Institute of the Russian Academy of Sciences (IKI RAS) and the Kirillo-Belozersky Museum-Reserve are actively investigating the use of neural networks to repair damaged works of art. Particular attention is paid to the frescoes of the Ferapontov Monastery, Cathedral of the Nativity of the Virgin, created by the master Dionysius the Wise in the 16th century. The restorers’ work restores images lost due to damage using the Stable Diffusion neural network, which can create images based on text descriptions.
The experiment was conducted on photos of murals taken in 2016, and the results show that the neural network successfully retouched both minor and significant losses. Not only the background and patterns were restored, but also the saints’ parts and their postures and gaze directions. Despite some “errors” in the reconstruction, experts believe that the quality is already impressive. It is then planned to eliminate the detected deficiencies and improve the results through additional training of the neural network on specially selected images.
The research represents the first steps in using artificial intelligence to restore works of art; The researchers acknowledge that it is not yet fully certain whether the restored image matches the original appearance of the artwork.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
