Facebook uses artificial intelligence to address the needs of persons with vision disabilities

Similarly to other social media and web services’ providers, Facebook is now offering image description features for persons with vision disabilities. However, the solution offered by the organization is unique in the sense that it operates based on artificial intelligence functionalities.

According to Facebook, more than 2 billion photos shared through their platforms every day are not accessible for more than 39 million blind persons in the world and over 246 million with serious visual impairments. For this reason, they specify in their official site that: “As Facebook becomes an increasingly visual experience, we hope our new automatic alternative text technology will help the blind community experience Facebook the same way others enjoy it.”

 Facebook’s automatic alternative text (Alt text) or image descriptions are generated with object recognition technology that determine the main objects that appear in a picture. This list is read to the users when they watch images on their newsfeeds: “Image may contain people, smiling, outdoors”.

Today the feature is only available in some English speaking countries, nevertheless it is planned to be included in other languages and platforms soon.

To explore in more detail Facebook’s process for generating its image description feature, access their article: Under the hood: Building accessibility tools for the visually impaired on Facebook

 

Additional Sources: Facebook NewsroomResearch at Facebook