Optipress - Machine Learning based image compression

Different images compressed with the same compression parameters may result in different perceptive quality. Moreover, the same image in different sizes may require different compression strategy to achieve maximum size gain without visible quality loss.

To overcome this, you can use our Optipress JPG compression algorithm.

Optipress finds out the best JPG compression parameters by analysing specific image features. A Machine Learning model determines the best compression strategy and quality is evaluated based on a simulation model of the Human Visual System.

By using Optipress, you can get the most of the JPG compression format.

Using Optipress

foptipress[1..3]

/width/400/foptipress3 /width/400/q90 /width/400/q80

Filter value File size
foptipress3 22 KB
q90 28 KB
q80 15 KB

You can combine different filters to achieve specific results. For example:

/width/500/fradius30.fgrey.webp-lossy-70.foptipress3/sample.li/flat2.jpg

This means that if the user’s browser supports WebP, such image compressed with quality 70 will be delivered. Otherwise, Optipress will find the maximum JPG compression without noticeable quality loss.