All of us who have old photographs of our relatives usually keep them as a treasure, especially if they are people who are no longer with us. But in a very large percentage of cases, the older the photos, the more they are corrupted.
We all would like to have photos of our grandmother or great-grandmother (those of us who were lucky enough to meet her) in good condition, which seems almost impossible.
Well, the technology that some so often deplore can serve us in many good ways, and among them will be the ability to restore those photographs that mean so much to our memory, but, unfortunately, it shows that time has inexorably passed them by. . .
We will be able to restore faces that are less visible due to the loss of black and white definition, those that are damaged due to the fact that the paper already has wrinkles, and we will even be able to see details that we could never perceive.
To do this, we are going to use the GFP-GAN model, a website where we can eliminate as much as possible time lapse from old photos.
It is a service that helps us to digitize those very old photos that we all have at home, inherited from our parents, which we are not only going to save on our computer or in the cloud, but we will do it by greatly improving the quality.
But this will not only improve the photo in terms of downloadable parts or their color, but it will also improve the sharpness, which is very important in restoring the whole essence of what it means to the photo.
How GFP-GAN Works
GFP-GAN (Generative Facial Preview) was created by Xintao Wang, Yu Li, Honglun Zhang, and Ting Shan to integrate this new feature as a new architecture into the GAN model.
This allows you to restore human faces from photos damaged over time, and do it in a few seconds.
Get these amazing results since first convolutional neural network (U-Net) retouch a damaged photo to remove any parts that are worn or simply broken. This is done thanks to GAN stylea network capable of generating faces thanks to artificial intelligence.
This network is capable generate non-existent face parts due to the poor condition of the photo and thus create a new photo with all parts in good condition. Various filters and various parameters are passed to it until a result is obtained that is as close as possible to the original image.
Yes, it is true that in some cases where the facial image is severely damaged, it may happen that the face is slightly different from the original imagebecause when you try to cover up gaps or errors in the original photo, it turns out not quite right.
How can we fix our photos?
Once we know how this website can significantly improve photos, it’s time to find out what we need to do in order to be able to work with it and thus correct the pictures we need.
If you don’t have this type of image but want to see how it works, you can always find it online and follow the steps we’ll show you next to see the result. :
- We entered the GFP-GAN website.
- Once we’re inside, we’ll see you have a series of photos at the bottom to take a test and see how the software works.
- If we touch any of these images, they will be automatically processed and you can admire the result.
- If we already have a photo, we must touch the place where it says Select a file…
- The next thing to do is to go into our repository and select the one we want to significantly enhance.
- The moment we see that it has been loaded and indicate that it is 100% (they will put it in the same place where it put Choose file…), we press the blue button on the right where it says Restore photo.
- Then we will realize that just below the photo we uploaded appears on the left side of the image, so after a few seconds the improved photo due to processing appears on the right.
- If we want to download the result, just click on Download the restored image, button, which is located just below the already modified and improved version.
- Then we have to right click on the open image and then select Save image as.
From now on, you will save the enhanced image to the desired location on your device’s internal memory.
It’s a simple, quick and very effective way to bring old photos that we all have at home that are already badly damaged back to life.
Source: Computer Hoy
I am Bret Jackson, a professional journalist and author for Gadget Onus, where I specialize in writing about the gaming industry. With over 6 years of experience in my field, I have built up an extensive portfolio that ranges from reviews to interviews with top figures within the industry. My work has been featured on various news sites, providing readers with insightful analysis regarding the current state of gaming culture.