Amazon Web Services (AWS), Amazon’s software-as-a-service unit, introduced its Rekognition AI service toward the back end of 2016, along with Lex and Polly. Rekognition uses deep learning to analyse images, and recognise scenes, objects and faces – returning a series of attributes for each face. AWS have now taken this a step further with the introduction of an estimated age range attribute.
The new feature lets anyone upload their photos to let Rekognition have a go at guessing their age – while also displaying the same attributes as it did prior. The feature works much the same as Microsoft’s How-old.net, but shows an age range rather than an individual age estimate.
Jeff Barr, AWS’s chief evangelist, thinks the feature can be used to “power public safety applications, collect demographics, or to assemble a set of photos that span a desired time frame,” he said in a blog post.
However, “in general, Rekognition [sic] the actual age for each face will fall somewhere within the indicated range, but you should not count on it falling precisely in the middle,” Barr added.