There’s an old proverb “Looks are deceptive”. Now Microsoft is facing this challenge every now and then with its How Old Do I look bot. Microsoft has released a new website known as (how-old.net) How Old Do I look. This website does the toughest job in the world, that is guessing age. Yes, when you open the website, you will be able to pload your own photo and How Old Do I Look application will detect you age.
You can try out images of celebrities, sports persons or any famous personality. Old Old bot will try to detect the age of the person in the photo along with the gender. It can also detect ages from group photo and tags the faces with age and gender.
This is a fun app obviously and even though the app can extract the gender correctly but the precision for age detection is still nowhere near perfect. but what is the story behind this. How is this done? Here’s what Microsoft says:
We wanted to create an experience that was intelligent and fun could capture the attention of people globally, so we looked at the APIs available in the Azure Machine Learning Gallery. The gallery contains many finished intelligent services for Face, Speech, and Vision which are part of a new suite called Project Oxford from Bing and Microsoft Research. The Face API has a demo page that uses the API to detect and extract information about faces in a photograph. We found the ability of the face API to estimate age and gender to be particularly interesting and chose this aspect of it for our project. To make the experience more fun we used the face API alongside the Bing Search API from the Azure marketplace to create http://how-old.net.
However, Microsoft assures that they don’t store any photo or information from the app. As Microsoft is decking up for the Build 2015 Conference, it is giving a new shot to its upcoming Face Detection app by releasing the How Old Do I Look website. The face locating and gender detection feature is near perfect. Microsoft is still working hard on the age detection counterpart. Apart from being a tool for testing the Face Detection app, How Old website also helped in understanding the user demand. While Microsoft expected a fifty or a few more user contacts, but Within hours, over 210,000 images were submitted with 35000 users from all over the world. Interestingly, about 29k of them are from Turkey.
You can check what Microsoft thinks your age is from how-old.net