sciencetech-blog

comments

Microsoft has revealed a new website that claims to be able to predict someone's age and sex from just a picture.

Called  How old do I look, it allows people to age and sex any image found on Bing, Microsoft's search engine, or upload their own.

It even allows users to search for celebrities - and see what Microsoft thinks their real ages are. 

The site was able to accurately guess Kim Kardashian's age as 34, but thought 37 year old Kanye West was actually 38 - while a woman in the background of the image was deemed to be just 6 years old.

'This may be hard to believe but it took a couple of developers just a day to put this whole solution together, starting with the pipeline from the web page to the Machine Learning APIs to the real time streaming analytics and real time B,' said Microsoft. 

It designed to site to show off the capabilities of its cloud server software. 

It revealed the site today at its annual Build developers conference in San Francisco. 

'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,' it said.

The system works by analysing 27 points on the face.

Called face landmarks, they are a series of specifically detailed points on a face; typically points of face components like the pupils, canthus or nose. 

'These attributes are predicted by using statistical algorithms and may not always be 100% precise. 

The site incorrectly guessed guesses 33 year old Chloe Kardashian is actually 44

'However, they are still helpful when you want to classify faces by these attributes.

'Face landmarks are a series of specifically detailed points on a face; typically points of face components like the pupils, canthus or nose.

Face landmarks are optional attributes that can be analyzed during face detection. 

By default, there are 27 predefined landmark points.  

HOW IT WORKS 

The system works by analysing 27 points on the face (left)

Called face landmarks, they are a series of specifically detailed points on a face; typically points of face components like the pupils, canthus or nose. 

'These attributes are predicted by using statistical algorithms and may not always be 100% precise. 

'However, they are still helpful when you want to classify faces by these attributes.Face landmarks are a series of specifically detailed points on a face; typically points of face components like the pupils, canthus or nose. 

The Face API has a demo page that uses the API to detect and extract information about faces in a photograph, and the team decide to work with this to create a site they hoped would go viral.

'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.'

The site can even identify groups of poeple

The site allows people to age and sex any image found on Bing, Microsoft's search engine, or upload their own.