How-Old.net started as an experiment with Microsoft’s newly released face detection APIs. Corom Thompson and Santosh Balasubramanian, Engineers in Information Management and Machine Learning at Microsoft, explained in a blog post that they expected 50 users for this test but have seen tens of thousands of people flock to the site during this week’s Microsoft Build Developer Conference.
“We were shocked. Within a few hours, over 35,000 users had hit the page from all over the world (about 29k of them from Turkey, as it turned out — apparently there were a bunch of tweets from Turkey mentioning this page),” they wrote. “What a great example of people having fun thanks to the power of [machine learning].”
Machine Learning Apps Made Easy
What may be more interesting is that it only took a couple of developers one day to create How-Old.net and it’s fairly accurate. The developers created a pipeline from the How-Old.net Web page to Microsoft’s machine learning APIs to real-time streaming analytics and real-time business intelligence. Altogether, this truly demonstrates the power of Azure services, they said.
“We wanted to create an experience that was intelligent and fun [and] could capture the attention of people globally, so we looked at the APIs available in the Azure Machine Learning Gallery,” Thompson and Balasubramanian said. “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.”
The Machine Learning Gallery is home to intelligent service from the beta Project Oxford, which debuted at the Build conference earlier this week. Project Oxford helps developers create smarter apps that can do things like recognize faces and interpret natural language even if the app developers are not experts in those fields.
“If you are an app developer, you could just take the API capabilities and not worry about the machine learning aspect,” said Vijay Vokkaarne, a principal group program manager with Bing, whose team is working on the speech aspect of Project Oxford.
Beyond Age Guessing
We asked Rob Enderle, principal analyst at the Enderle Group, for his thoughts on the viral app. He told us this is clearly a blend of facial recognition, analytics and learning machines. With that in mind, what applications does it have beyond guessing our ages?
“We are likely to see more blends like this to help locate things on the Web, identify or analyze automatically problems at home, with our devices and cars, and at work, and give advice on what to wear to an event or to identify physical areas of concern early like skin cancer,” Enderle said. “We are moving into a time when ever more intelligent machines will be increasingly helpful in a proactive way and this is one of the showcases to that progress.”