Put the paintbrush down – AI can restore artworks quicker and better
Artificial intelligence (AI) could spell the end of art restoration by humans after MIT showed that damaged paintings can be repaired in just a few hours.
Typically, conservators spend months or years researching and matching paints, colours and techniques to ensure the finished product is as close to the original as possible.
But Alex Kachkine, an engineering graduate at MIT, has shown it is possible to use AI to fill in the damaged areas digitally, then print the restored layers onto a thin film to attach on top of the painting.
It means the painting appears restored even though the original is still intact beneath.
The method was applied to a highly damaged 15th-century oil painting, and AI immediately identified 5,612 separate regions in need of repair, and filled in these regions using 57,314 different colours. The entire process, from start to finish, took just three and a half hours.
'There is a lot of damaged art in storage that might never be seen,' said Mr Kachkine. 'Hopefully with this new method, there's a chance we'll see more art, which I would be delighted by.'
In recent years, digital restoration tools have allowed conservators to create virtual representations of restored works.
AI algorithms can quickly sift through huge amounts of data about artists and time periods to generate a digitally restored version of a particular painting, in a way that closely resembles the correct style.
However, digital restorations are usually displayed virtually or printed as stand-alone works and until now there has never been a way to translate the digital restorations on to the original work.
The new technique involves first scanning the painting and then using software to create maps of the areas that require repair, before matching the correct colours.
The maps are then inkjet printed onto two sheets of a film. One film, which sits underneath, is printed in white so that the colours overlaid on the second sheet are not muddied by the colours of the original painting.
The layers are carefully aligned and overlaid by hand onto the original painting and adhered with a thin spray of varnish.
The printed films are made from materials that can be easily dissolved with conservation-grade solutions, in case conservators need to reveal the original, damaged work. The digital file of the mask can also be saved as a detailed record of what was restored.
For the painting that Mr Kachkine used, the method was able to fill in thousands of losses in just a few hours.
'A few years ago, I was restoring this baroque Italian painting with probably the same order magnitude of losses, and it took me nine months of part-time work,' he said. 'The more losses there are, the better this method is.'
The new technique may even help restore botched restorations such as Elias Garcia Martinez's Ecce Homo, which was restored so badly it was dubbed Monkey Christ, because of the Simian features of Jesus.
Mr Kachkine acknowledges that there were ethical issues to consider, in terms of whether a restored version was an appropriate representation of an artist's original style and intent.
He said the process should be carried out in consultation with conservators with knowledge of a painting's history and origins to check the AI programme was not going off-piste.
'Restoring a painting is fun, and it's great to sit down and infill things and have a nice evening,' he added 'But that's a very slow process.'
The research was published in Nature.
Broaden your horizons with award-winning British journalism. Try The Telegraph free for 1 month with unlimited access to our award-winning website, exclusive app, money-saving offers and more.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Yahoo
7 hours ago
- Yahoo
Video does not show astronaut exposing 'fake' life in space
"Former Astronaut Karen Nyberg Shows How NASA Fakes Space Flights.." a June 16, 2025 post on X claimed. It shares a video of a woman in front of a space station-like backdrop watching a chip bag float away from her, side-by-side with the same footage being filmed in front of a green screen. A person in a green body suit is manipulating the bag. The video garnered thousands of interactions in posts on X, Threads, Instagram and Facebook. The clip also circulated in posts making similar claims about Nyberg in French, Spanish and Italian. Nyberg is a retired NASA astronaut who completed two spaceflights during her career, including a 166-day stay on the ISS in 2013 (archived here). But she is not the woman in the green screen video. Comments on the post on X included links to previous debunks from USA Today and PolitiFact who named the woman in the video as Paige Windle. In the clip, a person off-camera is heard calling the woman Paige. Windle is the founder of a lifestyle management company and the wife of David Weiss, known online as "Flat Earth Dave," the host of "The Flat Earth Podcast." Contacted by AFP, Weiss confirmed Windle is the one on-camera. "This video never dies. It keeps coming back," Weiss said in a June 17 email. The video was originally posted on Weiss's YouTube channel as part of a series titled "Globebusters," but made no mention of Nyberg (archived here). "Someone took that clip and presented it as Karen Nyberg and it went viral a bunch of times and now it has started again," he said. He said he has repeatedly addressed the false use of the video, sharing with AFP the cover image of a YouTube video he posted in response to the false viral claims (archived here). Astronauts onboard the ISS experience microgravity, causing them and objects to float (archived here). At the altitude of the ISS, gravity is 90 percent of the total gravity one feels on Earth, but an absence of air resistance causes all objects in the ISS to fall at the same rate, producing a weightless appearance. The ISS stays afloat because it moves at a speed that matches the curve of the Earth, causing it to "fall around" the planet while staying at roughly the same altitude. The moon's orbit works in a similar way. NASA uses the ISS in part to study how extended time periods in microgravity and other conditions in space impact the human body as it prepares for future long-term missions in space. On her website, Nyberg features a video she recorded on the ISS where she worked on a quilt (archived here). Unlike in the video filmed in front of the green screen, Nyberg's hair and necklace float throughout the clip due to the microgravity conditions. AFP reached out to Nyberg's representative for comment, but a response was not forthcoming. AFP has previously debunked claims that ISS astronauts faked a video from the station.
Yahoo
7 hours ago
- Yahoo
Researchers Scanned the Brains of ChatGPT Users and Found Something Deeply Alarming
Scientists at the Massachusetts Institute of Technology have found some startling results in the brain scans of ChatGPT users, adding to the growing body of evidence suggesting that AI is having a serious — and barely-understood — impact on its users' cognition even as it explodes in popularity worldwide. In a new paper currently awaiting peer review, researchers from the school's storied Media Lab documented the vast differences between the brain activity of people who using ChatGPT to write versus those who did not. The research team recruited 54 adults between the ages of 18 and 39 and divided them into three groups: one that used ChatGPT to help them write essays, one that used Google search as their main writing aid, and one that didn't use AI tech. The study took place over four months, with each group tasked with writing one essay per month for the first three, while a smaller subset of the cohort either switched from not using ChatGPT to using it — or vice versa — in the fourth month. As they completed the essay tasks, the participants were hooked up to electroencephalogram (EEG) machines that recorded their brain activity. Here's where things get wild: the ChatGPT group not only "consistently underperformed at neural, linguistic, and behavioral levels," but also got lazier with each essay they wrote; the EEGs found "weaker neural connectivity and under-engagement of alpha and beta networks." The Google-assisted group, meanwhile, had "moderate" neural engagement, while the "brain-only" group exhibited the strongest cognitive metrics throughout. These findings about brain activity, while novel, aren't entirely surprising after prior studies and anecdotes about the many ways that AI chatbot use seems to be affecting people's brains and minds. Previous MIT research, for instance, found that ChatGPT "power users" were becoming dependent on the chatbot and experiencing "indicators of addiction" and "withdrawal symptoms" when they were cut off. And earlier this year Carnegie Mellon and Microsoft — which has invested billions to bankroll OpenAI, the maker of ChatGPT — found in a joint study that heavy chatbot use appears to almost atrophy critical thinking skills. A few months later, The Guardian found in an analysis of studies like that one that researchers are growing increasingly concerned that tech like ChatGPT is making us stupider, and a Wall Street Journal reporter even owned up to his cognitive skill loss from over-using chatbots. Beyond the neurological impacts, there are also lots of reasons to be concerned about how ChatGPT and other chatbots like it affects our mental health. As Futurism found in a recent investigation, many users are becoming obsessed with ChatGPT and developing paranoid delusions into which the chatbot is pushing them deeper. Some have even stopped taking their psychiatric medication because the chatbot told them to. "We know people use ChatGPT in a wide range of contexts, including deeply personal moments, and we take that responsibility seriously," OpenAI told us in response to that reporting. "We've built in safeguards to reduce the chance it reinforces harmful ideas, and continue working to better recognize and respond to sensitive situations." Add it all up, and the evidence is growing that AI is having profound and alarming effects on many users — but so far, we're seeing no evidence that corporations are slowing down in their attempts to injecting the tech into every part of of society. More on ChatGPT brain: Nation Cringes as Man Goes on TV to Declare That He's in Love With ChatGPT
Bloomberg
7 hours ago
- Bloomberg
Universities Win Order Voiding Agency's 15% Research Cost Cap
Brown and Cornell universities, the Massachusetts Institute of Technology and several other US schools won a federal court order striking down a National Science Foundation cap on indirect cost rates for government-funded research. Judge Indira Talwani struck down the cap on Friday, finding it 'arbitrary, capricious and contrary to the law,' granting summary judgment to the suing schools plus the Association of American Universities, and denying that relief to the government.
