Facebook improves AI photo descriptions for the visually impaired

A new object detection system can also recognize where people are in the frame as well as the number of people in the scene. And while you’ll normally get a simple summary of a photo’s contents, you’ll have the choice of a detailed description that outlines the position, size and nature of objects.

This still offers an imperfect description — Facebook stressed that it still uses “may be” and errs on the side of excluding concepts it can’t solidly recognize. It’s a big leap for accessibility on the social network, though, and could ensure that more visually impaired people at least get the gist of photos in their feeds.