This project will restore old face photos and make them higher quality using AI.
It uses an ML modal from the Applied Research Center called GFPGAN on Replicate to restore face photos. The application gives you the ability to upload any photo, which will send it through this ML Model using a Next.js API route, and return your restored photo.
After cloning the repo, go to Replicate to make an account and put your API key in a file called .env
.
Then, run the application in the command line and it will be available at http://localhost:3000
.
npm run dev
Deploy the example using Vercel:
- Migrate from react-uploader to S3 + filepond to have access to uploaded photos for the dynamic share page
- Implement a dynamic share page
- Use Vercel OG to dynamically generate an OG card that contains the old and new pics side by side
- Create a hash and store it in redis or postgres along with links to the old and new photos
- With this new hash, create a new sharable dynamic page that has the photos side by side
- Add toggle to be able to restore both face photos and other old photos using swinr model
- Improve the generation of the photo to extend beyond faces; maybe run it through a general model before
- Add a carousel of good examples to the index page