REMINI

An application for Smartphone that uses a neural network that improves image quality.

Remini works on the basis of a neural network (big data). The Remini server processes the photo and draws your image based on a large database of uploaded and analyzed photos.

The app has a paid and free rate. Paid is inexpensive, and if you have a lot of retouching, you can consider buying it. Plus, you will have additional ways to process photos.

But the free option is also very successful, because you can process up to 5 images per day for free. For most people, this is enough. The” fee ” for free processing is the viewing of ads in the app. 

The installation process is no different from installing other apps on your phone.

It is also easy to use: upload the desired image and in 10-30 seconds you will receive its processed version. Even very blurry images can be converted to a good result.

Experiment

At the same time, if you re-upload the processed time photo, you can further improve the quality of this photo.

1. For the experiment, I took the original photo and made it blurry. 

Agree, it is difficult to see the face from this photo.

If I had been sent such a photo for burning a year ago, I would have simply refused to burn it. Accordingly, I would have lost this order.

2. After that I uploaded the photo to the Remini app

Compared to a blurry photo-the result is amazing!

Already at this stage, you can save the image and process it in Photoshop. But I decided to go further.

3. For the experiment, I once again uploaded the image to Remini and got the following result

At first glance, there is no difference. But if you zoom in and take a closer look…

The eyes became clearer. The sharpness on the skin increased. Additional details appeared on the lips. The hair above the forehead is also better drawn.

4. Now let’s compare the original photo and our attempt to save the blurred image

And here we are faced with the fact that there are still differences. The eyes stand out the most. After all, Rimini is a neural network. It analyzes all the proportions, colors, shapes, and so on. And makes the image as averaged as possible with the base that is loaded into the neural network. But if you look closely, it seems that on the original photo and on the result, it seems that the result after processing looks more natural. There is a chance that the girl in the photo is actually wearing lenses, which is why her eyes are so big. 

Also, Rimini didn’t pull the sharpness of the hair to the right of the face. But you can try to fix it in Photoshop or make your hair with an imitation of oil paint. Otherwise, everything looks the same as in the original photo: the shape of the face, lips, nose, eyebrow lines, hands, and so on.

But let’s be fair – the neural network could not know this and took as a basis a blurry photo, where the size of the retina of the eye is very blurred.