Restore Old Photos
Upload an old or damaged photo and let AI restore it. Removes scratches and fixes fading — then enhance further with colorization, face restoration, and upscaling.
Drop your old photo here or click to upload
Supports JPG, PNG, WebP (max 10MB)
Or try a sample image:
How AI Photo Restoration Works
Upload
Drop your old or damaged photo. JPG, PNG, or WebP up to 10MB.
AI Restores
Google Gemini analyzes damage patterns and reconstructs missing details.
Download
Compare before & after, then download in full resolution PNG.
What Our AI Restores
What Can AI Photo Restoration Fix?
Our AI is trained on millions of damaged and restored photo pairs, allowing it to recognize and repair a wide range of physical and digital degradation. Here is a closer look at each type of damage and how the restoration process handles it.
Scratches and Tears
Scratches are the most common form of old photo damage. They happen when prints are stacked without separators, stored loose in drawers, or handled without care over decades. Tears occur when photos are pulled from sticky albums or bent during moves. Our AI detects linear damage patterns across the image surface and fills in the missing information using surrounding pixel data. Fine surface scratches become invisible after a single pass, while deeper tears that remove the emulsion layer are reconstructed based on the context of adjacent areas. For photos with extensive tearing, running restoration in multiple passes produces the cleanest results.
Water Damage and Mold Stains
Water is one of the most destructive forces for photographs. Floods, basement leaks, and prolonged humidity cause paper to warp, colors to bleed, and mold to grow between emulsion layers. The AI identifies water stain boundaries and mold patterns as distinct damage types, removing discoloration while preserving the underlying image detail. Mild water damage restores almost perfectly. Severe cases where the image has partially dissolved require more reconstruction, but the result is still a dramatic improvement over the damaged original. For more detail, read our guide on fixing water-damaged photos.
Fading and Yellowing
Every photograph fades over time. Sunlight, heat, and chemical oxidation of the paper and emulsion turn vibrant colors into washed-out pastels and shift whites to yellow or brown. This damage is mathematically predictable, which makes it one of the easiest types for AI to correct. The model understands what natural skin tones, skies, and foliage should look like and adjusts the color channels to restore the original vibrancy. Yellowed photos often show the most dramatic before-and-after improvement of any damage type. For black-and-white photos that have yellowed, you can also add color with our colorization tool after restoring.
Dust and Scan Artifacts
When you digitize an old photo with a flatbed scanner, dust on the glass or on the print creates white or dark specks in the digital file. Low-quality scanners also introduce banding, moire patterns, and color noise that were never in the original photograph. The AI recognizes these digital artifacts as distinct from the actual photo content and removes them without affecting genuine image detail. Dust spots vanish entirely, banding is smoothed, and scanner noise is reduced to produce a clean, faithful representation of the original print.
Missing Areas and Heavy Degradation
Some photos have sections that are completely gone — corners torn off, chunks of emulsion flaked away, or areas dissolved by water. The AI handles these by analyzing what remains of the image and generating plausible content for the missing regions. Backgrounds, clothing patterns, and natural scenery reconstruct well because the model has strong priors for these subjects. Missing facial features benefit from following up with our dedicated face restoration tool, which specializes in reconstructing eyes, noses, and mouths with higher fidelity than general restoration alone.
AI Restoration vs Professional Restoration
Until recently, restoring a damaged photograph meant hiring a professional retoucher or learning Photoshop yourself. AI has changed the equation. Here is how the two approaches compare for everyday photo restoration work.
Professional Hand Restoration
- ×$25 - $150+ per photo depending on damage
- ×2 - 14 days turnaround time
- ×Quality varies by individual retoucher
- ×Requires finding and vetting a specialist
- ×Revisions cost extra time and money
- ×Better for museum-grade archival work
AI Photo Restoration
- Under $0.50 per photo with credit packs
- Results in under 30 seconds
- Consistent quality on every photo
- No appointments — upload and go
- Re-run as many times as you want
- Ideal for family photos and large collections
For most people restoring family photos, AI is the clear winner. The cost difference alone is significant: restoring a shoebox of 50 old photos with a professional retoucher could cost $1,250 or more, while AI restoration handles the same batch for under $25 and delivers results in minutes instead of weeks.
Professional hand restoration still has a place. If you own a historically significant photograph with large missing sections that require creative artistic reconstruction, or if you need archival-grade output for a museum exhibit, a skilled human retoucher can make nuanced decisions that AI cannot replicate. But for the other 99% of photo restoration work — the family reunions, the wedding portraits, the childhood snapshots — AI delivers faster, cheaper, and more consistent results. You can try 3 photos free and judge the quality for yourself before buying credits.
Best Practices for Photo Restoration
A few simple preparation steps can dramatically improve your restoration results. The AI works best when it has a high-quality, well-framed input image to analyze.
- 1
Scan at 300+ DPI for the best results. Higher resolution gives the AI more pixel data to work with, producing sharper restorations. For small prints or wallet photos, scan at 600 DPI.
- 2
Crop out album borders, page edges, and any non-photo content before uploading. This helps the AI focus entirely on the actual image and avoids artifacts from border areas.
- 3
Always restore damage before colorizing. Run the restoration tool first to remove scratches, stains, and fading, then use the colorization tool on the clean result. This two-step approach produces significantly better color accuracy.
- 4
Use face restoration after general restoration for portraits. The dedicated face model in our face restore tool recovers sharper eyes, noses, and mouths than general restoration alone.
- 5
Upscale as the final step if you plan to print. Running restoration and face restoration before upscaling avoids amplifying damage artifacts into the enlarged image.upscale tool
- 6
For severely damaged photos, restore in multiple passes. Run the first restoration to fix the worst damage, then upload the result for a second pass to clean up remaining imperfections.
3,100+ photos restored
Under 30 seconds
Zero data storage
Photo Restoration FAQ
Our AI restores scratches, tears, creases, stains, water damage, mold spots, yellowing, and fading. It works on both scanned physical photos and digital images.
Most photos are restored in 10-30 seconds. The AI processes your image in memory and returns the result directly to your browser.
No. You get 3 free restoration credits without signing up. Create an account only if you want to buy more credits or track your history.
We support JPEG, PNG, and WebP files up to 10MB. For best results, upload the highest resolution scan you have — 300+ DPI scans work best.
No. Your photo is processed in memory and returned directly to your browser. We never store, share, or access your images. All transfers are encrypted with 256-bit TLS.
Yes. Our AI handles extensive damage including heavy scratching, tears, water damage, and severe fading. Very badly damaged photos may benefit from restoring first, then using face restoration and colorization as separate steps.
You get 3 free restorations with no signup required. After that, credit packs start at $4.99 for 10 credits (about $0.50 per photo). Each restoration uses 1 credit, and credits never expire.
No. Photos are processed in memory on secure servers and are never permanently stored. Free users' photos are deleted immediately after processing. Pro users can optionally save restoration history, but this is opt-in and you can delete your history at any time.
Yes. Pro users can batch-process multiple photos in a single session. Each restoration uses 1 credit. Upload your photos one at a time or use batch mode to queue several at once.
Yes. Always restore damage first, then colorize. Restoring first removes scratches, stains, and fading so the colorization AI has clean source material to work with. Each step uses 1 credit.
Related Guides
How to Restore Old Photos at Home
A complete walkthrough for digitizing and restoring your family photo collection without professional help.
How to Fix Water-Damaged Photos
What to do when photos have been through floods, leaks, or humidity damage — and how AI can help recover them.
How to Remove Scratches from Old Photos
Step-by-step techniques for removing scratches, from light surface marks to deep emulsion gouges.
How to Restore Faded Photos
Why photos fade over time and the best techniques for bringing color and contrast back to washed-out prints.
Start Restoring Your Old Photos Today
3 free restorations, no signup required. Upload any old or damaged photo and see results in seconds.
Upload a Photo