Correcting Old, Damaged Images with Generative Restore and AI - Cloudinary Product Update
Cloudinary is constantly adding more Generative AI features to its Programmable Media product. You can use the new Generative Restore transformations to correct old and damaged images. This AI-driven approach makes it easier to work with many types of images. These include user-generated content, vintage photographs, scanned images, and those damaged by over-compression.
Join Sam, Aditya and Paul from Cloudinary for this walkthrough of the Generative Restore features. You'll see how Cloudinary's Generative Restore can remove quality issues like compression artifacts, grainy or blurry pixels, and automatically enhance image quality by embedding a specific parameter in the URL. Users can apply these improvements programmatically at scale for hundreds or thousands of images.
Paul explains the technology behind Generative Restore, which involves using generative models to identify and correct blurred or damaged pixels in images. He demonstrates the effectiveness of this tool with before-and-after examples, highlighting how it can bring old, low-quality images back to life.
The walkthrough also emphasizes the versatility of Generative Restore when combined with other Cloudinary transformations like "e_improve," which enhances image lighting and quality, and the use of AI-based transformations in Cloudinary's SDKs.
You can also learn more about Cloudinary's Generative Restore from the following resources:
- Blog Post: https://cloudinary.com/blog/generativ...
- Documentation: https://cloudinary.com/documentation/...
Not a Cloudinary user yet? Sign up for your FREE account today! https://cloudinary.com/users/register...
Смотрите видео Correcting Old, Damaged Images with Generative Restore and AI - Cloudinary Product Update онлайн без регистрации, длительностью часов минут секунд в хорошем качестве. Это видео добавил пользователь Cloudinary 18 Октябрь 2023, не забудьте поделиться им ссылкой с друзьями и знакомыми, на нашем сайте его посмотрели 174 раз и оно понравилось 5 людям.