Artificial Intelligence Allows for the Restoration of Damaged Paintings in Just a Few Hours

Databricks придбає стартап Neon за $1 млрд для посилення позицій на ринку ШІ-агентів

An innovative artificial intelligence technology developed by Massachusetts Institute of Technology graduate student Alex Kachkin is opening a new era in the field of art restoration. This modern method allows for the restoration of damaged canvases in just a few hours, significantly reducing the time compared to traditional restoration methods.

This is reported by Business • Media

  • The technology is based on the analysis and digital restoration of damaged areas using AI.
  • Special printed masks that can be easily removed without damaging the original are used for restoration.
  • The new approach can even be applied to works that cannot be restored using classical methods.

How the Painting Restoration Technology Works Using AI

The essence of the method lies in the digital reconstruction of the canvas: artificial intelligence analyzes neighboring fragments of the painting and recreates the lost or damaged parts. The resulting digital mask is printed on a transparent polymer film, which is carefully placed over the original artwork. If necessary, this film can be easily removed without harming the original, ensuring the reversibility of the process.

Kachkin successfully tested his technology on a 15th-century canvas attributed to the Master of the Adoration of the Magi at the Prado. To do this, he digitized the painting, marked over 5,600 defects, and restored them using a software mask, employing more than 57,000 unique colors.

Ethical Questions and Future Development Prospects

In cases where details were missing — specifically, the image of a baby’s face — artificial intelligence selected appropriate elements from other works by the same author. As a result, AI was able to accurately reproduce the artist’s style and organically complement the composition. The outcome was a process that took only a few hours instead of the usual eight days of classical restoration.

Although the technology is currently suitable only for lacquered canvases, its potential is impressive. Kachkin believes that the new method will allow many works that have long been overlooked due to the complexity of restoration to return to exhibitions.

“It will require a lot of thought about the ethical issues that arise at every stage of this process to understand how it can be applied in a way that aligns with conservation principles. We are laying the groundwork for the development of further methods. As others work on this, we will arrive at more precise methods,” Kachkin noted.

With the emergence of such technologies, new questions arise for specialists, particularly regarding the ethics of concealing the original under a film or recreating lost fragments based on other works. At the same time, this innovative approach allows for the restoration of artistic values that were previously considered hopelessly lost.