Amazon Rekognition Update – Estimated Age Range for Faces
Amazon Rekognition is one of our artificial intelligence services. In addition to detecting objects, scenes, and faces in images, Rekognition can also search and compare faces. Behind the scenes, Rekognition uses deep neural network models to analyze billions of images daily (read Amazon Rekognition – Image Detection and Recognition Powered by Deep Learning to learn more).
Amazon Rekognition returns an array of attributes for each face that it locates in an image. Today we are adding a new attribute, an estimated age range. This value is expressed in years, and is returned as a pair of integers. The age ranges can overlap; the face of a 5 year old might have an estimated range of 4 to 6 but the face of a 6 year old might have an estimated range of 4 to 8. You can use this new attribute to power public safety applications, collect demographics, or to assemble a set of photos that span a desired time frame.
In order to have some fun with this new feature (I am writing this post on a Friday afternoon), I dug into my photo archives and asked Rekognition to estimate my age. Here are the results.
Let’s start at the beginning! I was probably about 2 years old here:
This picture was taken at my grandmother’s house in the spring of 1966:
I was 6 years old; Rekognition estimated that I was between 6 and 13:
My first official Amazon PR photo from 2003 when I was 43:
That’s a range of 17 years and my actual age was right in the middle.
And my most recent (late 2015) PR photo, age 55:
Again a fairly wide range, and I’m right in the middle of it! In general, Rekognition 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.
This feature is available now and you can start using it today.