Face.com develops face recognition technology. Two of its main applications run on facebook – PhotoTagger, which automatically tags your facebook friends, and Photo Finder, which finds lost photos of friends.

After scanning billions of photos and tagging over 50 million users through Photo Tagger and Photo Finder on Facebook, the company is moving ahead with its goal of making face recognition approachable and available to all.

The new API,which is still in open alpha stage, lets external developers tap into Face.com’s face detection and face recognition through simple REST API calls. If you’re looking to build a photo tagging application, create personalized e-cards or campaigns, or any other sci-fi idea that comes to mind, you can now use Face.com’s platform for that purpose. Facebook users would eventually be able to identify images of themselves and friends as they browse any website.

Face.com introduced a few examples to show how the API can be used.

Tagger Widget – Using face.com’s Javascript Tagger Widget to quickly add a tagging experience

Poster Yourself – Using the API to automatically plug friend faces into posters and e-cards

CelebrityFindr – A full fledged mini site that uses face recognition to search and filter out Celebrity photos

David Sifry, developer and founder of Technorati, got an early look at The new API and said: “I’ve been impressed with Face.com’s API, and their plan for working closely with developers to build great applications that incorporate face detection and face recognition. Open platforms like this one will enable the creation of exciting new applications that we’ve never seen before at scale.”

In order to get started and start generating API calls, developers need to sign up. You can read more on Face.com’s new API.

Read more about Face.com on Thewadi’s company index.

Article originally posted on Holes In The Net (Hebrew) and translated by Yohay Elam.

We were curious to discover if this technology really works, and whether dark secrets will be instantly revealed. I can’t deny my doubts about the concept, but according to the guys at Profiler1, the Personology method that they are using has proved accurate according to a statistical survey made in the 60s, and used since then by the military and by relationship and business consultants.

A long piece in TheMarker (Hebrew) was made towards the event and explains the method. According to the article, the users should upload up to three good pictures of the person, and the system supplies a quick answer. The results are processed using a patent-protected technology and sometimes by an expert.

One of the founders, Itai Rauch was quoted saying:

Contrary to graphology that will eventually become extinct in an era when the keyboard is used, and contrary to most diagnosis methods in which the subject can easily search the web for a cheat sheet, our method cannot be gamed. The subject’s genes cannot be modified for the test. This method was revived after an American judge discovered a correlation between the facial features of the defendants and their felonies.

If you feel uncomfortable with his statement, you’re not alone. We sometimes get an initial impression of someone according to his facial features that turn out to be wrong after a deeper acquaintance. Add the fact that the anonymous American judge isn’t an academic authority. Perhaps the opposite is correct.

Did this judge see people with specific facial features as more guilty than others? Israeli researches showed that the chance of conviction differs between men,women, Jews, Arabs and also within Jews of different ethnic descent. This is closer to prejudice.

TheMarker’s test wasn’t that academic as well – they gave Profiler1 10 pictures of known figures from the hi-tech industry, and the results were too obvious – Were you aware that Yossi Vardi has a good sense for business? No need to read his face for that.

Ze’evi Bergman, the CEO of Nice, is a serious man that tends to put a lot on his shoulders, sometimes more than expected from him. His eye structure and eyebrows show that this is a charismatic person with excellent social skills. The texture of his skin and his hair show that this is a person capable of handling tough situations. His facial structure reflects charisma and leadership skills that might lead him to politics.

Testing the system with a known figure – the Prime Minister

I don’t know what you think of this technology so far, but astrology is what comes to my mind. I decided to test the technology by myself and uploaded an official picture of Israel’s Prime Minister – Benjamin Netanyahu. I was asked if this person underwent any plastic surgery, and I gave No for an answer. As far as I know, Netanyahu kept all his natural features.

After uploading the photo, I was asked to pay 10$ for the analysis. Profiler1 promises to supply an answer within 12 hours. This raised some suspicion, but it took merely 3 hours for the results to appear on my profile page.

I was surprised to see that Profiler1 forbids distributing its results in social media or anywhere else. This is quite aggravating for a paid service.

So, let’s see the results.

According to Profiler1, Netanyahu doesn’t tend to take impulsive decisions and usually measures his steps carefully. His past, serving twice as PM, shows the opposite.

Concerning pressure, Profiler1 draws a picture of a man that can take pressure, but sometimes cannot conceal his feelings. Even Netanyahu’s biggest fans will admit that he can easily be pressured. This applies to his time at Israel’s highest post and also when he wasn’t there.

Netanyahu is known as a suspicious man, sometimes obsessive about leaks from his office. According to Profiler1, he isn’t suspicious about information but it’s hard to deceive him. Their wording goes both ways.

And now for some correct results – Profiler1 identified Netanyahu as an articulate person that writes and speaks flawlessly, has a wide vocabulary and excellent persuasion skills. His biggest opponents won’t deny this. Also regarding charisma, Netanyahu is identified as a person with high charisma – a person that can drive others to action.

The results left me confused – if there’s a correlation between facial features and characteristics, but it isn’t precise, employers that will lean on these tests might make decisive decisions based on the results, especially if they strengthen their initial impression – sometimes prejudice. They might disregard a worker’s credentials. Such technology could also encourage plasic surgeries in order to give the candidate the perfect look.

This seems like a classic example of a technology that can be interpreted according to the initial feeling – tautology.

Read more on Profiler1 in TheWadi’s company index.

Translation: Yohay Elam

Earlier this month, at TechFest 2010, Microsoft demonstrated a neat prototype that was designed to solve this problem. Microsoft OneAlbum  is a project that uses face-recognition and event-matching technology to automatically retrieve photos that would be of interest to you from social  networks and your friends’ online photo albums—and add them to  yours.

OneAlbum uses face-recognition technology to search for photos  featuring you and your friends with whom you are often  photographed.

How does it work? OneAlbum scans faces that appear frequently on  your OneAlbum Live photo collection and then scours your social  networks (Facebook, Windows Live, Flickr etc) for more photos of those people. When it’s finished, you will see all the new pics in your  OneAlbum collection with links to the sites they were published on.

While the face-recognition algorithms were developed by Microsoft Research Asia, OneAlbum is the brainchild of researchers from Microsoft’s Israel Innovation Labs.

The technology redefines the concept of “my album” from referring to “photos I have taken” to “photos that interest me,” no matter where they might reside. “This,” said Eyal Krupka, principal research program manager for the Israel Innovation Labs, “includes photos of myself, my wife, and my kids, from the events I participate in, the places I like, and more. OneAlbum also goes further, by organizing ‘my album’—the photos from the same event are presented side-by-side, no matter where they come from.”

Krupka said that OneAlbum could be used to spark “after-party social interaction.” since it also looks for photos from specific events. Of course, Facebook already does this quite effectively. Here’s a video demonstration with Krupka.

OneAlbum resembles another technology developed in Israel, Face.com. Here’s our latest update on Face.com. While both technologies scour social networks after untagged photos, Face.com PhotoTagger is available only on Facebook. On the other hand, OneAlbum forces you to use OneAlbum, although it scours more social networks than Face.com. When examining the privacy aspects, it seems privacy advocates won’t be pleased from either one.

More in the company index: Face.com and Microsoft Israel R&D Center.