Automated Face Recognition

Face recognition is one of the next big violations of privacy.

Photos

Large databases of tagged face metrics are build with the goal of commercial use in future. The Facebook software tagged 30-40 billion fotos and every month 2.5 billion fotos are added. Face.com offers an API for their face recognition software and database. It is hosted in the cloud. You can use it for 5.000 fotos per hour for free. Today it is possible to identify 65 million individuals with this database.

A world leading company is Cognitec (Saxonia, Germany). In most cases Cognitec is working for gouverments. For the Mexican gouverment the passport photos of all the 80 million inhabitants were analyzed and the face metrics are provided in a database. 20 federeal US-states are using the service for millions of driver's licence fotos. The largest project of Cognitec is a secret database for US intelligence services. This database can identify 100 million individuals.

Videos

VideoSurf will publish a software for face recognition in videos this year. The software can scan a video and creates an index of identified individuals with timestamps. It is easy to jump to the position, where a person turns up in the video. The face metrics database of VideoSurf contains 60,000 well known people.

Smartphone software

All steps of automated face recognition can be done by a smartphone. Large databases can connect via Internet. Neven Vision Inc. (acquired by Google) developed such an face recognition app for photos taken by your smartphone. Google did not release the app at the moment. The software Viewdle (presented at the CES in Las Vegas this year) does not need a photo. It can display the names of identified individuals and additional informations from Facebook und Twitter in the preview screen of a camera.

CV Dazzle

Computer Vision Dazzle is a project by Adam Harvey. It combines makeup and hair styling (or other modifications) with face-detection thwarting designs. The goal of CV Dazzle is to break apart the gestalt of a face, or object, and make it undetectable to computer vision algorithms, in particular face detection.

Dazzle
How to hide from machines shows some more stylings.

This entry was posted by cane on Saturday, July 2. 2011 at 16:17. and is filed under Speaker's Corner. You can leave a response, or trackback from your own blog.

Trackbacks

Trackback specific URI for this entry
  1. Improvements in Face Recognition

    If a human will see two photos with unknown persons he will recognize the same person right 97.53 percent of the time. Facebook's new AI research group developed a software which was right 97.25 percent on the same challenge, regardless of variations in l

    Weblog: JonDonym News Center
    Tracked: Mar 20, 18:21

Comments

Display comments as (Linear | Threaded)

  1. Andrei D. says:
    #1 2011-07-02 22:56 (Reply)

    Face recognition has been around and used by governments for years. Databases are shared among the governments and can track people from ATM machines, satellites, to traffic cameras and many more. Interpol, FBI, NASA and other high level agencies have access to the databases, as well as continuously developing software. Privacy is a thing of the 70's :-(


Add Comment


Enclosing asterisks marks text as bold (*word*), underscore are made via _word_.
Standard emoticons like :-) and ;-) are converted to images.
E-Mail addresses will not be displayed and will only be used for E-Mail notifications.