Microsoft 'How Old' Website Detects People's Age And Gender In Photos, Check Out How It Works!

Tags
Microsoft

Microsoft's new website guesses the age and gender of faces it detects in a photo you have uploaded.

One of Microsoft's internal teams decided to have fun by using the company's recently released tool for facial recognition to build a website that predicts people's age in photo -- and it's appropriately named "how-old.net"

The HuffingtonPost noted that the age-guessing website allows users to upload a photo and then either feel incredibly flattered or terribly offended as it guesses the gender and age.

The website is powered by Face APIs, a new tool that was launched at the Build Developer Conference this week. The program's page defines it as "a cloud-based API that provides the most advanced algorithms for face detection and recognition."

Recommended uses include verifying whether two faces in separate photos belong to the same person, or using one person's photo to find him or her in other photos, CBC noted.

Microsoft's Machine Learning Blog team emailed several hundred people urging them to try out the API, however, they expected only about 50 fifty people to use it. They logged more than 35,000 tests in only a few hours. Apparently, the team underestimated people's interest in age and appearance.

Almost 30,000 of their hits came from Turkey. "Apparently there were a bunch of tweets from Turkey mentioning this page," the team noted.

In addition, the page also includes a series of stock photos to test the API. The Microsoft team said they expected more people to use these photos rather than uploading their photos. On the contrary, they found that fewer than half of the tests were done with the stock photos.

Users online described getting a wide range of ages with multiple photos owing to differences in framing, lighting and resolution.

That said, it's still very much a work in progress.

Upload your photo and share with us how correctly the website is able to guess your age and gender in the comments sections below!

Join the Discussion

Latest News

Real Time Analytics