After all, biometrics and object recognition are closely related

Lithuania’s Neurotechnology releases SentiSight 3.2 for object recognition (Baltic Course) — Provider of biometric identification technologies releases SentiSight 3.2 for object recognition.

See: Biometrics, object recognition and search

IBM 5 in 5 2012

Five technologies coming in the next five years according to IBM research. 

From this year's list...

Computers that can "see"

 

Related:
Biometrics, object recognition and search
(Facial Recognition vs Human) & (Facial Recognition + Human)

Google, Image Recognition & Search

The Midnight Epiphany That Changed Like.com From An Over-Hyped Failure To A $100 Million Acquisition (Fast Company)

On April 25th, 2006, just after midnight, Shah was looking at Riya’s stats when he noticed that for every person uploading personal pictures and tagging them, 20 others were using Riya for search. He immediately emailed one of his cofounders, Gokturk, who replied within five minutes. “I wonder what they are searching for?” Gokturk wrote back. “Why are they using our site instead of Google or Yahoo Images?”

Read the whole thing, then see: Biometrics, object recognition and search, inspired by an earlier Google acquisition.
Key paragraph:

With this acquisition, I suspect Google doesn't so much have facial recognition for identity search as they have object recognition in mind. First, Google has been wary of face recognition in public search. Whether this is due to the technical challenges of an unbound face rec application or a respect for the privacy of their users, I'll leave it to the reader to judge. It is also a much different challenge to return the result "This is a human face" than it is to say "This is a human face and that face belongs to Guy Herbert." In most object search the first type of result will be the most desirable anyway. If you submit a photo of an insect to a search, you aren't asking about the insect's individual identity, you probably just want to know what type of insect it is."

The current case, well documented by Fast Company, seems to bear that out.

Bikini Detection algorithm raises the stakes in social media

Add bikinis to the list of objects recognizable by computer algorithms.

IPhone app that finds racy Facebook photos raises privacy worries (Los Angeles Times)

“This is a very touchy subject, of course,” Barto said. “Anything that’s readily available on Facebook, that’s what we can search. Those privacy tools on Facebook should be used to control the content that you want to be private.”

The app works in a similar way to the facial-recognition technology found in video chat programs and Facebook’s tag prompts. But instead of identifying faces, Badabing identifies the shape of a bikini. That means in addition to beach photos, the app may return pictures of a T-shirt with the outline of a swimsuit.

Object recognition is really starting to take off.

Related: Biometrics, object recognition and search

FBI Wants a Tattoo Interpretation System

What Does Your Tattoo Say About You? The FBI Wants to Know. (Nextgov)

The FBI is consulting local police and vendors about technology currently in use that can spot crooks and terrorists by interpreting the symbolism of their tattoos, according to government documents. The inquiry follows work already underway at the bureau and Homeland Security Department to add iris and facial recognition services to their respective fingerprint databases.

The FBI on Friday issued a request for information on existing databases “containing tattoo/symbol images, their possible meanings, gang affiliations, terrorist groups or other criminal organizations.”

What is described is more of a pattern recognition system but it is related to biometrics.

See: Biometrics, object recognition and search

Another Entertainment Use For Facial Recognition Tech

Phone app uses facial recognition software to determine ugliness

"It wasn't just made up, like, 'Hey this is what I think is beautiful,'" said co-creator Jo Overline. "It's published math behind how it's calculated." The app uses those facts, such as how far apart your eyes are and how big your nose is, to rank your beauty on a scale of 0 to 100; 100 is hot, 0 is not. And the app tells you why you rank a certain way. It took Gilbert native Overline and his partner about a year to develop Ugly Meter Pro, the $4.99 version that uses facial recognition software to determine how attractive you are.

This type of application is more like object recognition than facial recognition for ID management but the two are closely related. See also: Biometrics, object recognition and search.

In other recent news...

Does this Woman Have the Perfect Face? (Time)

Her face was judged the most symmetrical out of 8,000 analysed.

How Image Processing Will Change the World

Exploring the future of software that can "see" (The Economic TImes)

The linked article does an excellent job of outlining applications that flow from the increased ability of software to make sense of non-textual, optical/image-based inputs. Biometrics like facial recognition are just the beginning.

Rituparna Chatterjee, the article's author, places facial recognition in its proper context here:

While DNA and fingerprints remain the US Federal Bureau of Investigation's favoured means of identification, it's now also using facial recognition. "Image Processing is an important technology for investigation," says Bill Casey, program manager of the FBI's Biometric Center of Excellence.

"When hunting for a criminal, we look through our database of over nine million photos for possible investigative leads." Casey shares how his colleagues recently nabbed a double-homicide suspect by comparing his photo with the US Department of Motor Vehicles' local database of 30 million photos in the state of North Carolina.

She is also perceptive in her observations of how Google has been showing interest in image processing technology. It's an article well worth reading in its entirety.

See also:
Google debates face recognition technology
Biometrics, object recognition and search

BBC Video About Facial Recognition

Facial recognition marks the end of anonymity (BBC)
That's a bold statement but the video is interesting as a survey of what Carnegie Mellon University is up to in biometrics applications and platforms.

It's a good video. Click on over there and give it a look. It even discusses the privacy/anonymity issue in terms that have a lot in common with a something we posted last week.

But about the PittPatt Technology using face recognition to discover Social Security Numbers (this links to a text article describing the first app featured in the video linked above)...

I'm not buying it.

What Alessandro Acquisti and his team have succeeded at is using a face as a search term in a search engine focused on Facebook. This is probably why Google bought the technology. It is an interesting and important development in its own right and something discussed here: Biometrics, object recognition and search.

But, using a photo of a face to find a Facebook page has zero, nothing to do with Social Security Numbers. An individual's birthday and place of birth, however, has a lot to do with their Social Security Number. See: Google yanks request for kids' social security numbers:

It is open to children from kindergarten through the last year of high school and requires that a contestant provide a Parent Consent Form along with their submission. The original form asked for the child’s city of birth, date of birth, the last four digits of the child’s social security number, and complete contact info for the parents.

Now that's how you get Social Security Numbers! Social engineering and a wide net, not covert efforts focused on individuals.

The only way the PittPatt technology could be useful in stealing an identity by trying to deduce a person's SSN is:
♦ The identity thief has identified a particular individual whose identity she wants to steal.
♦ The identity thief knows no other information about the mark, but she has a photo.
♦ The mark has stored private information (either the SSN itself, or the details that make it somewhat predictable) online in a publicly viewable way together with their photo.

Essentially, the photo just replaces the name as the search term. I've never heard of identity thieves doing their research by typing random names into a search engine and hoping for the best. This isn't how identity theft works and no rationally profit-maximizing identity thief would ever do something like that. It is far easier to search for Facebook profiles that share birthdate information than to sit around on the sidewalk taking pictures of people in the hope that (a) they have a Facebook account, and (b) it contains instructions on how to steal their identity.

Granted, as this type of search becomes more common, and it will, it won't just be Facebook that gets searched. But at most, the PittPatt project allows users to do a search for people without knowing their name.

The real problem is the importance, predictability and unchangeability of Social Security Numbers, not facial recognition.

Biometrics assist in effort to save great apes

Special software helps to save species (Physorg.com)

With the aim of better protecting endangered species, game wardens are studying the behavior of surviving great apes in the wild. This is often painstaking work because it is difficult to distinguish between different individuals. A new software system will make things easier by analyzing the animals’ faces for individual identification.

There are other areas where biometric technologies are being applied in the life sciences. Leafsnap is an app that uses visual recognition software to help identify tree species from photographs of their leaves. StripeSpotter is a free open-source system with an algorithm that can identify animals in the wild and build biometric databases using photos of the different animals.

Biometric ID management technologies appear to be a small first step towards allowing computers to make sense of the world using inputs from nature rather than being limited to text inputs. Human ID management offers the big returns on investment that attract labor and capital but the technology has applicability to challenges far more diverse than human identity management.

Related thoughts:
Biometrics, object recognition and search

Biometrics, object recognition and search

One of the limitations of the internet search we all know and love is that it is bound to text. If you have a picture of something, you can't find out what it is with a search engine. If you've recorded a bird's song, there's no way to get that into the search field to identify the species of bird that created it.

Google obviously recognizes this limitation of its technology and the obvious benefits of extending its capabilities.

Two main types of technology have been developed to help get real-world (non-text) inputs into a form that computers are good at using: Optical Character Recognition and Biometrics.

Optical Character Recognition (OCR) involves taking an image of text and converting it to textual data. Banks use OCR to scan the numbers at the bottom of checks, document scanners use it to make better use of scanned documents, and more recently, it has been used with vehicle license plates in law enforcement. OCR was the logical first step along the path of using pictures as computer inputs because text is fairly easy to break into its constituent parts (characters) and there simply aren't that many different characters to identify. So the tech is pretty easy and the ROI (Return on Investment) of using computers to transcribe text instead of humans is very straightforward.

Biometrics are the second generation of automatic object recognition. Biometrics are far more complicated than OCR but far less complicated than a theoretical no-holds-barred image search engine. With biometrics the tech is more difficult but the ROI of using computers to help manage identity can be very substantial indeed.

Techniques developed for biometric identity management are also being applied to recognition tasks that do not deal with human identity management. Leafsnap is an app that uses visual recognition software to help identify tree species from photographs of their leaves. StripeSpotter is a free open-source system with an algorithm that can identify animals in the wild and build biometric databases using photos of the different animals. Here, the tech developed for high ROI applications is being applied to new, though lower value challenges.

Which brings us to Google's acquisition of PittPatt, a Pittsburgh pattern recognition company.

Google Buys PittPatt Facial Recognition Tech (PCMag.com)

In late March, Google denied plans for a dedicated facial-recognition app, although the company has said it could do so as far back as the launch of Google Goggles, which used object recognition to identify real-world objects.

With this acquisition, I suspect Google doesn't so much have facial recognition for identity search as they have object recognition in mind. First, Google has been wary of face recognition in public search. Whether this is due to the technical challenges of an unbound face rec application or a respect for the privacy of their users, I'll leave it to the reader to judge. It is also a much different challenge to return the result "This is a human face" than it is to say "This is a human face and that face belongs to Guy Herbert." In most object search the first type of result will be the most desirable anyway. If you submit a photo of an insect to a search, you aren't asking about the insect's individual identity, you probably just want to know what type of insect it is.

I'll bet Google's more interested in the object recognition capabilities of PittPatt than they are in facial recognition.

Biologists create barcode scanner for zebras

Free Open-source Zebra Biometrics (Wired.co.uk)

When researchers want to identify an animal, they need to take a photo of it, then select a similarly-sized rectangle on its body and the system will then analyse its constituent black-and-white pixels to assess which animal on the system it most closely resembles.

For obvious reasons, zebras are a great place to start to apply biometric techniques to wildlife. They come with their own bar codes. They live in well-lit, flat, uncluttered terrain. And there are lots of them.

A real promise of such systems, though, is their applicability to species and conditions that are considerably more challenging.