Last year, the facial recognition system picked out more than 1,000 cases that resulted in State Police investigations, officials say. And some of those people are guilty of nothing more than looking like someone else. Not all go through the long process that Gass says he endured, but each must visit the Registry with proof of their identity. “We send out 1,500 suspension letters every day,’’ said Registrar Rachel Kaprielian, who says the system has been a powerful weapon to fight identity fraud since it was installed in 2006 but that it is not without problems. “There are mistakes that can be made.’’ Neither the Registry nor State Police keep tabs on the number of people wrongly tagged by the system. But Gass estimates in his lawsuit that hundreds might have received revocation notices in error since the system was installed.
“Intel® Audience Impression Metrics Suite (Intel® AIM Suite) enables network operators, content providers, advertising agencies and service providers to deliver relevant messages and gauge the effectiveness of their networks and content like never before.
Intel AIM Suite detects the number of viewers, determines their gender and age group, and measures dwell time. This data can be used in real-time to tailor on-screen content based on the demographics of current viewers. For campaign planning and ROI measurement, AIM Suite data can be correlated with Proof-of-Play data to determine content-specific viewership metrics by demographics and time of day.”
In other words, there’s a camera in the signage, and it’s pointed at you. It analyses you for a group of basic characteristics, and then changes the adverts to match your presumed demographic. It also collects and collates this information, and feeds it back to the advertiser.
Where’s Waldo: Matching People in Images of Crowds
Given a community-contributed set of photos of a crowded public event, this paper addresses the problem of ﬁnding all images of each person in the scene. This problem is very challenging due to large changes in camera viewpoints, severe occlusions, low resolution and photos from tens or hundreds of different photographers. Despite these challenges, the problem is made tractable by exploiting a variety of visual and contextual cues – appearance, timestamps, camera pose and co-occurrence of people. This paper demonstrates an approach that integrates these cues to enable high quality person matching in community photo collections downloaded from Flickr.com
Citation: “Where’s Waldo: Matching People in Images of Crowds”, Rahul Garg, Deva Ramanan, Steven M. Seitz, Noah Snavely, Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2011, pp. 1793-1800.
And here’s a PDF of the paper itself. I’d love to see the Google Analytics results for this paper: Iran, China, Burma, Occupy Wall St…