Hotpot Design
Your Design Assistant
Start new thread

AI Picture Restoration - Free way to remove scratches, sharpen colors, and fix photos

byā€¢
Restore and repair pictures with AI. Hotpot builds on the latest research to automatically remove scratches, sharpen colors, and enhance faces, transforming damaged photos into cherished memories. Repair both color photos and black & white images for free.

Add a comment

Replies

Best
Maker
Hi Product Hunters! We're excited to launch AI Picture Restoration. This service is free and automatically removes scratches, sharpens colors, and enhances faces. The deep learning model repairs both color pictures and black & white ones. We built on top of extraordinary open-source research from the Microsoft team led by Ziyu Wan and Bo Zhang. Our changes included performance, quality, and image enhancements that render the academic model more fit for production use. While AI cannot match professionals, we have seen many people use this to restore damaged photos and revive cherished memories. The main adopters are consumers who can't afford professionally restored family pictures. In a surprising twist, professionals also use Hotpot to accelerate their workflow and serve more clients. If you repair photos with Hotpot, please share them: attach the before/after versions and background story. for how to create image comments. We're also launching our AI Picture Colorizer soon. for our PH launch. Can anyone recognize who's in the first pair of pictures? :)
Steve Jobs!
Maker
Correct! :) Maybe I'll add a harder one in the comments.
Maker
Here's a much harder one. Can anyone guess the second person in the picture? If it's too easy, I can also increase the trivia difficulty to headache level. :)
this effect is amaizing! Great job!
based on the background which puts the date in 1983 or 1982 and the height difference of the second person, Iā€™d guess that the second person is Mike Markkula an early angle investor in Apple.
I was just talking to someone about there should be a product like this. Mind blow AI technology is being used. Have a bunch of family photos from the 60s that I want to get restored.
Maker
Thanks Asad! Please also sign up for the AI Picture Colorizer, which we will launch soon as well. How did the family photos turn out that you tried restoring so far?
just got all the photos I want to restore - going to do it after work today. Super excited TBH. šŸ˜‡
Maker
Awesome! If you have questions, feel free to message/email me. If any pictures convert poorly, please share them. We're constantly improving the model, and could use more training data.
bless! Appreciate it šŸš€
surprised that the images can not only be made softer, but also sharper. Respect.
Maker
Thanks very much, Walter!
Wow, I need to try this. I love technology and this "magic"
Maker
Thanks, Piotr! Hope we can help. If some images convert poorly, please share the results with api *at* hotpot dot ai.
Very cool stuff helping a lot with digitizing old family photos :)
would you recommend that to business solution?
Going to try in out. Good luck and congratulations!
Maker
Thanks, Molly! Would love to hear how the pictures turn out.
In high school they taught us how to restore old photos like this using Photoshop. Of course, it took several more hours than this. šŸ˜…
Maker
Thanks, Ryan. We're not as good as professionals yet, but we're working hard to close the gap. Were you trying to restore family photos or historic images?
I like this! I had blurry pictures and it was able to clearly define lines around eyes and the smile. My grandmother and mother look beautiful.
Maker
Thanks, Cathleen. I'm glad we were able to help. :)
I am running an Instagram project (.ai) dedicated to AI-powered black and white colorizations. I use available models like DeOldify or Instance-aware Colorisation. Would love to try/test your model, too. Ping me if interested! Peace!
Maker
Hi there! Please message me separately. We're exploring different ways to open-source some of our models so happy to chat more.
nice idea
Maker
Thanks Julia! Did you get a chance to try it yet?
12
Next