Art restoration takes regular hands and a discerning eye. For hundreds of years, conservators have restored paintings by identifying areas needing repair, then mixing a precise shade to fill in a single area at a time. Often, a painting can have hundreds of tiny regions requiring individual attention. Restoring a single painting can take anywhere from just a few weeks to over a decade.
In recent times, digital restoration tools have opened a path to creating virtual representations of original, restored works. These tools apply techniques of computer vision, image recognition, and color matching, to generate a “digitally restored” version of a painting relatively quickly.
Still, there was no strategy to translate digital restorations directly onto an original work, until now. In a paper appearing today within the journal , Alex Kachkine, a mechanical engineering graduate student at MIT, presents a brand new method he’s developed to physically apply a digital restoration directly onto an original painting.
The restoration is printed on a really thin polymer film, in the shape of a mask that could be aligned and adhered to an original painting. It may even be easily removed. Kachkine says that a digital file of the mask could be stored and referred to by future conservators, to see exactly what changes were made to revive the unique painting.
“Because there’s a digital record of what mask was used, in 100 years, the subsequent time someone is working with this, they’ll have an especially clear understanding of what was done to the painting,” Kachkine says. “And that’s never really been possible in conservation before.”
As an indication, he applied the tactic to a highly damaged Fifteenth century oil painting. The strategy robotically identified 5,612 separate regions in need of repair, and filled in these regions using 57,314 different colours. The whole process, from start to complete, took 3.5 hours, which he estimates is about 66 times faster than traditional restoration methods.
Kachkine acknowledges that, as with all restoration project, there are ethical issues to contemplate, by way of whether a restored version is an appropriate representation of an artist’s original style and intent. Any application of his latest method, he says, must be done in consultation with conservators with knowledge of a painting’s history and origins.
“There’s plenty of damaged art in storage which may never be seen,” Kachkine says. “Hopefully with this latest method, there’s a likelihood we’ll see more art, which I could be delighted by.”
Digital connections
The brand new restoration process began as a side project. In 2021, as Kachkine made his strategy to MIT to start out his PhD program in mechanical engineering, he drove up the East Coast and made a degree to go to as many art galleries as he could along the way in which.
“I’ve been into art for a really very long time now, since I used to be a child,” says Kachkine, who restores paintings as a hobby, using traditional hand-painting techniques. As he toured galleries, he got here to appreciate that the art on the partitions is just a fraction of the works that galleries hold. Much of the art that galleries acquire is stored away since the works are aged or damaged, and take time to properly restore.
“Restoring a painting is fun, and it’s great to sit down down and infill things and have a pleasant evening,” Kachkine says. “But that’s a really slow process.”
As he has learned, digital tools can significantly speed up the restoration process. Researchers have developed artificial intelligence algorithms that quickly comb through huge amounts of knowledge. The algorithms learn connections inside this visual data, which they apply to generate a digitally restored version of a selected painting, in a way that closely resembles the sort of an artist or time period. Nevertheless, such digital restorations are frequently displayed virtually or printed as stand-alone works and can’t be directly applied to retouch original art.
“All this made me think: If we could just restore a painting digitally, and effect the outcomes physically, that might resolve plenty of pain points and disadvantages of a standard manual process,” Kachkine says.
“Align and restore”
For the brand new study, Kachkine developed a technique to physically apply a digital restoration onto an original painting, using a Fifteenth-century painting that he acquired when he first got here to MIT. His latest method involves first using traditional techniques to scrub a painting and take away any past restoration efforts.
“This painting is sort of 600 years old and has passed through conservation again and again,” he says. “On this case there was a good amount of overpainting, all of which needs to be cleaned off to see what’s actually there to start with.”
He scanned the cleaned painting, including the various regions where paint had faded or cracked. He then used existing artificial intelligence algorithms to research the scan and create a virtual version of what the painting likely looked like in its original state.
Then, Kachkine developed software that creates a map of regions on the unique painting that require infilling, together with the precise colours needed to match the digitally restored version. This map is then translated right into a physical, two-layer mask that’s printed onto thin polymer-based movies. The primary layer is printed in color, while the second layer is printed in the very same pattern, but in white.
“As a way to fully reproduce color, you wish each white and color ink to get the complete spectrum,” Kachkine explains. “If those two layers are misaligned, that’s very easy to see. So I also developed just a few computational tools, based on what we all know of human color perception, to find out how small of a region we will practically align and restore.”
Kachkine used high-fidelity industrial inkjets to print the mask’s two layers, which he fastidiously aligned and overlaid by hand onto the unique painting and adhered with a skinny spray of conventional varnish. The printed movies are created from materials that could be easily dissolved with conservation-grade solutions, in case conservators need to disclose the unique, damaged work. The digital file of the mask can be saved as an in depth record of what was restored.
For the painting that Kachkine used, the tactic was capable of fill in hundreds of losses in only just a few hours. “Just a few years ago, I used to be restoring this baroque Italian painting with probably the identical order magnitude of losses, and it took me nine months of part-time work,” he recalls. “The more losses there are, the higher this method is.”
He estimates that the brand new method could be orders of magnitude faster than traditional, hand-painted approaches. If the tactic is adopted widely, he emphasizes that conservators must be involved at every step in the method, to be certain that the ultimate work is consistent with an artist’s style and intent.
“It is going to take plenty of deliberation concerning the ethical challenges involved at every stage on this process to see how can this be applied in a way that’s most consistent with conservation principles,” he says. “We’re establishing a framework for developing further methods. As others work on this, we’ll find yourself with methods which are more precise.”
This work was supported, partly, by the John O. and Katherine A. Lutz Memorial Fund. The research was carried out, partly, through the use of apparatus and facilities at MIT.Nano, with additional support from the MIT Microsystems Technology Laboratories, the MIT Department of Mechanical Engineering, and the MIT Libraries.
