Monday, June 11, 2012

Who gets credit? The Israeli-Spanish-Israeli Robocop World Cup Facial Recognition glasses


By way of introduction, I remember a "field trip" that I took when I was in the MBA program at Cal State Fullerton in the late 1980s. One day we visited two Toyota facilities in southern California - a manufacturing plant, and a corporate office. The one thing that I remember from that day were the differing stories that we got from the two locations. A manager at the manufacturing plant clearly stated that Toyota was an American company - a statement that makes sense, since Toyota was obviously manufacturing in the U.S. But by the time we got to corporate headquarters, the people there were saying that Toyota was "of course" a Japanese company.

In truth, Toyota is a Japanese company when it's in its interest, and an American company when that's in its interest. As a California employee of a Virginia-based subsidiary of a French company (with critical functions based in Texas), I know this all too well.

Which brings us to the Robocop glasses.

I thought that I had written about the World Cup Robocop glasses in the Empoprise-BI business blog, but it turns out that the only time I wrote about the World Cup in this blog, I talked about alcohol.

So let's go back to this story in Dvorak Uncensored from April 2011.

Brazilian police will use futuristic ‘Robocop-style’ glasses fitted with facial recognition equipment to identify and root out troublemakers at the 2014 World Cup.

Since I work in the biometrics industry, there were a couple of statements that caught my eye:

The system can compare biometric data at 46,000 points on a face....

The camera will generally be used to scan faces in crowds up to 50 metres (164ft) away but can be adjusted, if searching for a specific target, to recognise faces as far as 12 miles away.

I was wondering which company provided the technology for this, but I was unable to discover this at the time. The only thing that I found was in this Brazilian article:

Segundo o major Agostini, atualmente o equipamento é oferecido por um representante de uma empresa de Israel.

However, the name of the Israeli company itself was not identified.

I recently went to search for more information on this, and I ran across this item:

A small, young Spanish company, Ex-Sight, sells a leading edge technology for facial recognition that will be used by Brazilian police during the 2014 World Cup.

Well, even if I couldn't find out about the Israeli company, at least I could find out about the Spanish company that's working with them. So I went to Ex-Sight's website and actually found the hardware that the Brazilians were using, complete with the Robocop look.

The name of this Spanish product? xEye Yoav.


Now I took four years of Spanish in junior high and high school in Virginia (and I still have a connection with Virginia - my employer is headquartered there), and perhaps my Texan and Cuban teachers missed some vocabulary, but I don't recall learning the Spanish word "Yoav."

I dug deeper:

Ex-Sight.Com LTD
Etgar 2 Tirat Carmel 39032, Israel 39032
Tel: +972-777-841262
E-Mail: Sales@Ex-Sight.Com

World Wide Distributors:

175 175 - Ramal dos Menezes Street
São Paulo - SP São Paulo City,
Zip Code 02469-000 Brazil
Tel.: 55 -11- 2236 1422
Fax.: 55 -11- 2236 6789

Mr.Elazar Lozano Vidal
Ex-Sight España, C/ Ponent, 5 Alicante
Telf: 635654933 | Fax: 966873300
Mob: 628720760

So it became clear what happened. In order to score brownie points with Spanish business promoters, Ex-Sight positioned itself as a Spanish company. But when it works with investors, it probably positions itself as an Israeli company. And I'd be willing to bet that in order to get the World Cup business, it positioned itself as a Brazilian firm.

Whatever it takes.

P.S. Ex-Sight's blog is here.

P.P.S. For those who are biometrically inclined, here's a comment that I left on the Dvorak post back in the day:

Update – I am a member of the Biometrics group on Yahoo!, and I posed my question there. Dr. James Wayman posted a thorough response, and while I won’t attempt to reproduce the whole thing here, Dr. Wayman did say that one way to perform facial recognition calculations is via vector analysis in which a single vector may consist of the pixels from the image. Thus, it is correct to state that it is possible to use 46,000 “features” in facial recognition comparisons. Dr. Wayman referred to the definition of “feature” in ISO/IEC SC37 N3971, and although I haven’t reviewed that document, I trust Dr. Wayman on this one...

Scanning through all of the Dvorak comments, I was unable to find any reasonable explanation for the "12 miles away" claim.
blog comments powered by Disqus