Twitter now uses machine learning to bring smart auto-cropping for your photo posts

The social media giant, Twitter just announced a few feature on their website. Keeping in mind the photos in the Twitter post containing objects and scenery. These are present apart from humans or their animals. The company just come up with a new feature they called smart auto cropping. It uses the power of machine learning and image processing to frame the right parts of shared images. It then automatically crops the image to the right part. So that your followers can focus on the image and not the random background.

What does Twitter's new feature do?

According to the company, the new system will crop images with the help of a neural network. In this case, it uses a picture "saliency quality". It can then use that to help in determining attributes of the picture. Then the attributes are selected depending on which would normally attract attention. These attributes would be included in the pictures and cropped accordingly.

The company has also used data from eye-tracking studies. The data helped them to form various predictions for smart cropping. The data helped them to found that the puppies, faces, text, and regions of high contrast attracted the most attention from people. This data would help them to then crop the images to attract the most attention. Dubbed as smart auto-cropping, this can be helpful for those images where there may not be facial images.

Also Read: Xiaomi Redmi Note 5 to launch later next month, Here’s what we know so far

This new feature is being rolled out in Twitter’s app on iOS and Android. Right now the feature is only limited to photos. Later on, the feature will roll out to videos as well. Although, Twitter is expected to come out with a new feature to help users shoot images directly from the app. This will help them implement this feature to the videos easily.