Bringing old photographs back to life
Three weeks after Cleanr launched, a woman named Sarah sent us a message. She'd used the restoration tool on a photograph of her grandmother from 1987, one she'd kept in a drawer for years because it was faded and worn. The restored version now sits in a frame on her shelf. That message is why we built this feature.
The problem we were trying to solve
When we started building Cleanr, we noticed something: people were juggling five different apps to do what ought to be one job. They'd use one tool for blemishes, another for sharpness, a third for colour work. But there was a gap nobody was filling well. Old photographs.
Not vintage-effect photos, which is a filter choice. I mean genuinely old prints. Pictures that had lived in boxes or albums for decades. Faded. Soft. Sometimes water-spotted or creased. Most existing tools either treated restoration as a single button that over-sharpened everything, or they required you to understand colour theory and spend twenty minutes tweaking sliders.
We wanted restoration to feel like discovery, not labour. You open an old picture, and the work happens. That's what we set out to build.
What happens under the surface
The restoration feature combines three separate processes that we run in sequence. First, the app performs JPEG cleanup using a median filter to remove compression artefacts that build up in old scans. Then it applies noise reduction to lift the grain and blur that comes with age. The third step is the sharpening pass, which uses our unblur engine to recover lost detail without making the image look harsh.
We don't use a single neural network that tries to do everything at once. Instead, each stage is designed to handle a specific problem that time introduces. A photograph from 1985 has different issues than a blurry portrait from last week, and treating them the same way just doesn't work.
The whole process takes about two to five seconds depending on the image size. You see the before and after side by side, and you can undo it if it's not right for that particular picture.
Why we didn't oversell it
During development, we had conversations about how to position this feature. Our product team wanted to call it something grand. "Bring memories back to life." "Time machine for your photos." All of it true in spirit, but we kept coming back to one question: will people trust it, or will they feel disappointed when it doesn't do the impossible?
A photograph from 1962 that was badly exposed will still be badly exposed. Restoration can recover some of what the original capture lost, but it cannot add information that was never there. We decided to be honest about that, even in marketing.
It's why the free tier includes one restoration per day, not unlimited. We wanted people to experience it on real photos and understand what it does. Not burn through it on everything and expect miracles. The patience that comes with limitation often teaches respect for the tool.
A feature that works quietly
Most of Cleanr's tools are things you see immediately. Remove an object, and the space is suddenly empty. Apply a sky preset, and the horizon shifts. But restoration is subtler. It's a compression artifact gone. A detail that was soft now just slightly sharper. The colour slightly truer.
When we tested it with real users, we noticed something. They didn't click it once and declare victory. They'd sit with an old image, apply restoration, compare, sometimes undo it, think about it, apply it again. That told us we'd gotten something right. The feature wasn't overselling itself. It was honest enough that people could make their own judgment.
Sarah's message is still in our Slack. Every few months someone from the team sends it again when we're debating whether a feature is too ambitious or needs to promise more. It reminds us: the goal isn't to make the best possible story about Cleanr. It's to make something useful enough that customers send notes without being asked.
How it fits into the larger toolkit
We see restoration as part of a broader philosophy: let one app do the things you currently need four or five to do. Most people don't just have one old photo that needs attention. They have a series of them. Grandmother's wedding pictures. Dad's scanned negatives. A box of prints from the nineties.
That's why restoration sits alongside object removal, background cleanup, and the JPEG artefact remover. Together, they handle most of the common photo problems. In the Plus tier and above, you get batch processing, which means you can restore ten or fifty photographs in one go without clicking through each one individually.
The Free tier gives you one restoration per day, which is genuinely enough if you're being thoughtful about it. If you're a family member helping an elderly relative digitise a collection, the Plus tier at three pounds ninety-nine a month lets you work through an entire album in an afternoon without worrying about daily limits.
When was the last time you looked at an old photograph you owned? Not a digital screenshot, but an actual print or a scan that's been sitting somewhere. What would it feel like to see it properly again?