The technology can reliably identify concepts in categories including transportation (‘car,’boat,’ ‘airplane’), nature (‘snow,’ ‘ocean,’ ‘sunset’), sports (‘basketball court), and food (‘sushi’)
Visually impaired Facebook users can now use embedded object recognition technology to provide them with spoken text descriptions of photos.
Users of iOS devices are able to hear lists of items that are on shown in the photographs. Automatic alt text recognises objects using machine learning, which helps to build artificial intelligence by using algorithms to make predictions. The iPhone's VoiceOver feature to reads descriptions of the photos out loud to user.
While still in its early stages, the technology can reliably identify concepts in categories including transportation (‘car,'boat,' ‘airplane'), nature (‘snow,' ‘ocean,' ‘sunset'), sports (‘basketball court), and food (‘sushi'). It can also describe people (‘baby,' ‘smiling, ‘beard'), and identify selfies.
The company said the facility was part of its mission to make the world more open and connected, following on from research conducted in conjunction with Cornell University showed that although blind people were interested in visual content on social media, they often felt frustrated and even excluded or isolated because they couldn't fully participate in conversations centred on photos and videos.