Skip to main content

Old Photo Restorer API

spicyapi/photo-restorer-v1/edit

Old Photo Restorer takes one scanned photo and returns a cleaned PNG: scratches, dust specks, grain and blur removed, faces kept recognizable, tones unchanged.

From $0.048/image

Workbench

Input

Image editing
1/1

Drop files or media links, paste, or browse · 1 file · Max 10 MB · JPG · PNG · WEBP

  • 1
1 / 1 files
The scanned photo to repair: a JPEG, PNG or WebP upload, or a direct link to the file. One photo per request; the result is a PNG of about one megapixel in the same proportions. at least 1 characters
Safety checkerOff

SpicyAPI does not run a safety checker on this model. Your prompt and media go to the model as written, and the model itself decides what comes back.

Checking your account before enabling submission and pricing.

Output

Ready for a new run
Example result

Related models and next tasks

Missing the model you need?

OLD PHOTO RESTORER API

Scratches, dust and grain lifted from an old print in one call.

Key numbers

Test turnaround
~17 s
Time for our 640×960 sepia scan with scratches and dust.
Output size
~1 MP
640×960 came back at 832×1248; a 2048×1152 photo at 1360×768.
Prompts needed
0
Send the scan; there is nothing to describe or tune.

Before you spend

All modes

Restore photos you own or have permission to use; repairing a print does not change who holds the rights to it.

Image editing

One photo per request, returned as a PNG of about one megapixel in the source's proportions, so large scans come back smaller. Color is not added. Heavy damage across a face, missing pieces of the print and strong glare are rebuilt from what remains and may not match the original. Featureless areas such as an overcast sky can be smoothed out.

What it does

Old Photo Restorer before and after: a scratched, grainy sepia scan of a bearded man in a jacket on a city rooftop, and the restored version without scratches or dust.

Before and after

A scratched sepia portrait, cleaned in one pass

Our test scan had four scratches, seven dust spots, grain and blur. All of them were gone in the result, and the man's eyes, beard and jacket texture came back sharper while the framing stayed the same.

Up close

The face is repaired, not replaced

Close crop of the same face before and after restoration, with the dust spot and the grain removed.

At the same crop, the dust spot on his hair and the grain on his skin are gone, and his expression and features match the scan. Faces are where restoration matters most, so this is the place to check first.

Then add color

Pair it with Photo Colorizer

Restoration keeps the original tones. For black-and-white or sepia prints, send the restored PNG to Photo Colorizer; coloring a clean picture gives a more even result than coloring a scratched one.

Made with Old Photo Restorer

  • Old Photo Restorer before and after: a scratched, grainy sepia scan of a bearded man in a jacket on a city rooftop, and the restored version without scratches or dust.
  • Close crop of the same face before and after restoration, with the dust spot and the grain removed.
  • Old Photo Restorer output: a sharp, scratch-free sepia portrait of a bearded man in a dark jacket on a city rooftop.
  • A clean color photo of a man and a woman at a bar table, before and after Old Photo Restorer, with colors kept and the size reduced to about one megapixel.

Where it fits

  • Family albums and shoebox prints

    Clean up scans of grandparents' portraits, wedding photos and holiday snapshots before they go into a slideshow, a printed book or a family group chat.

  • Archives and local history projects

    Repair scanned collections where every print has a few scratches or dust specks, so volunteers spend their time on captions instead of retouching.

  • Memorial and anniversary pieces

    Prepare a single treasured photo for a frame, an order of service or a video tribute without booking a retoucher.

Model reference

Old Photo Restorer API reference

Old Photo Restorer API specifications

Endpoint
Image edit
Input
One photo (JPEG, PNG or WebP upload, or a direct link)
Output
PNG, about 1 megapixel, same aspect ratio
Repairs
Scratches, dust spots, grain, noise and softness
Color
Kept as sent; no colorization
Controls
None beyond the photo

How to call the Old Photo Restorer API

  1. Scan or photograph the print

    Flatten the print, avoid glare and keep the whole picture in frame. The tool repairs damage it can see; a reflection across a face is harder to undo than a scratch.

  2. Send image_url

    Upload the scan as JPEG, PNG or WebP, or pass a direct link to the file. That is the whole request.

  3. Check the faces, then add color

    Compare eyes and mouths with the original at full size. For a black-and-white print, pass the restored PNG to Photo Colorizer at /models/photo-colorizer-v1.

Old Photo Restorer API questions