19 June 2009

Facebook: Picture perfect

Who's that on Facebook? Is it you?

It seems we'll soon be seeing ourselves with this new Friend Finder application, developed by an Israeli startup in Tel Aviv called Face.com. Read their blog here.

Since Facebook was founded, more than 15 billion pictures have been uploaded. Some 60% of all photos on the web are on that site's pages. But it can be hard to find yourself or others in that huge number. Friend Finder claims to change that.

According to TechCrunch:

If there is one feature on Facebook which delivers “social utility” magic even to the most average of users, it’s Photos. In fact the feature is so popular that by Facebook’s own account 1 billion photos are uploaded every month—a staggering number that makes it the largest photo site on the Web.

However, as with all good things, there are also drawbacks, and in this case discovery is high on the list. While Facebook makes it super easy to discover photos in which you were tagged, there is no chance that every one of those billion photos are tagged each month. And that leaves a big opportunity.

It was designed as a low-cost platform to meet two requirements: to be able to tag everyday photos taken at low resolution, bad lighting, with red-eye or sunglasses-obscured faces. It also had to be scalable. Face.com claims to be able to perform facial recognition on all of the one billion photos currently uploaded into Facebook every month using only a few machines.

Photos are not stored, only tagged, and privacy is a concern. The photo is analyzed and dumped, giving Facebook users a tagged photo on his or her page. It will only tag photos within Facebook and within Friend Finder. No one can see the photos unless they are also running it. It also follows users' Facebook privacy settings.

If you are auto-tagged in an image, the application notifies you via a Facebook alert. The user may approve the photo or untag it and hide it from other application users. If you untag your image, it is private - no one will know.

The company, in its first alpha test month, scanned about 400 million photos, identified some 700,000 people and users confirmed the identities of about 150,000 people, according to the company's CEO/co-founder Gil Hirsch, despite the poor quality photos (shadows, red-eye, etc.)

Read the company's blog here and go to the site to read more (see the Press tab).

No comments:

Post a Comment