It arrived in a shoebox — a black and white portrait from 1962, taken at a photography studio in Porto. My grandmother at 28, elegant in a fitted dress, smiling at the camera. But decades of poor storage had done their damage.
The photo was yellowed across the top third. A deep scratch ran diagonally from corner to corner. The bottom right had been torn and taped back, leaving a rough crease. One of the faces — a younger sibling — was almost completely lost to a water stain.
I uploaded the scan to MUSE AI’s photo restoration tool. Within minutes, the AI got to work — analysing the original image layer by layer, reconstructing what the photograph must have looked like when it first came out of the darkroom in 1962.
The yellowing lifted. The scratch disappeared entirely, with the AI intelligently filling in the texture and tonal values of the surrounding area. The torn corner was rebuilt from context. And the lost face — that was the moment I held my breath — came back. Soft, a little uncertain, but recognisably human.
It wasn’t magic. It was pattern recognition at an extraordinary scale, trained on millions of photographs from the same era. The AI understood what a 1960s studio portrait should look like. It corrected for the specific way that silver gelatin prints age. It knew where the shadows fell.
I printed it on canvas and gave it to my mother for her birthday. She didn’t say anything for a long time. Then: “That’s exactly how I remember her looking.”
That’s the thing about old photographs. They’re not just images. They’re memory anchors. When they degrade, the memories feel less certain. When you restore them, something clicks back into place.
The restoration took less than five minutes. The emotion lasted considerably longer.
Your family has photographs like this. Boxes of them, probably. They won’t wait forever.
✨ Try Photo Restoration for €9.99
Upload your damaged photo and get a restored version in minutes. No subscription required.
Start restoring →
