When you search for "work related" videos on the internet, the results are normally limited to the title or tag given to any uploaded clip. What if computers could recognize specific human interactions? Would they get embarrassed?
A new method developed by an Oxford University team including Alonso Patron-Perez, Dr. Ian Reid, Dr. Marcin Marszalek, and Professor Andrew Zisserman does exactly that: It teaches computers to recognize human interactions such as shaking hands, hugging, and kissing, all of which naturally lead to heavy petting.
"Human actions and activities are of central importance in video analysis," said Alonso Patron-Perez of Oxford University's Department of Engineering Science, who led the research. "This new work makes it possible to recognize two-person human interactions, such as hugs, kisses and hand-shakes, automatically. Once you can recognize these interactions the applications are numerous: for instance you could automatically search home videos and YouTube for kisses and handshakes or even fast forward CCTV to find incidents."
The new method, built on algorithms from computer vision and machine learning, uses visual cues to determine A: If humans are present in the scene and B: What they are doing.
So far the team has focused on four types of interactions: Handshakes, high-fives, hugs, and kisses, but you know it's only a matter of time before your computer starts sorting your pornography by sexual position.
I'm sure there are plenty of scientifically sound, non-pornography uses for this technology, but I'll be damned if I can think of any right now.
Mistletoe hugs, kisses spotted by computer (w/ Video) [Physorg.com]