Back to Custom Key Chains
Enlarging your photo with good artificial intelligence makes all the difference in the world for final print quality. Most printing companies use classical methods build into image editing tools like bicubic resampling. We use AI single image super resolution. The best way to understand the difference is to look at the image below.
The results seem almost too good to be true. That is how much better artificial intelligence is at enlarging images.
Why is AI so much Better?
With classical image enlargement methods, the image is stretched and the pixels that are created as a result of the enlargement are colored according to a gradient curve or standard math formula.
With AI image enlargement, the pixels are smartly painted. The algorithm is able to smartly paint the pixel because it has been trained on many thousands of small and large versions of images. Over time, it learns how to take a small picture and make it a big picture much better than classic methods.
Do All Images Get AI Image Enlargement?
Not every image gets the AI image enlargement. That is because not every image needs it. If you have a large starting image, that is over 2000 pixels on each side, most of the time the image quality is going to be near perfect at print size, and the AI is not needed.
Where AI works best is mid-range images, in the 1000 pixel per side range. It really sharpens those up and most of the time the image quality is increased as the photo gets larger.
It also does an amazing job in taking very small pictures and turning them into printable enlargements.
Return to Custom Key Chains