Zdenek Kalal’s object tracking algorithm learns at the fly, more likely to make next 007 flick (video)

Microsoft’s own OneVision Video Recognizer might be novel, but when the people in Redmond are seriously trying to take things next-level, they ought to probably cast their gaze around the pond. Zdenek Kalal, a researcher on the University of Surrey, has just created what often is the most sophisticated vision system known to the civilian world. In essence, it takes the mundane task of tracking objects to a wholly new platform, enabling users to pick out an object at the fly and feature the algorithm immediately start tracking something new. Within seconds, it’s ready to maintain a lock although your object twists, turns, or leaves / returns. Furthermore, these “objects” can be used as air mice in case you force it to trace your digits, and in case you teach it what your staff seems like, you’ll have a completely automated security scanner which can recognize faces and grant / deny access in line with its database of white-listed individuals. Frankly, we’d rather you spot it for yourself than take heed to us extolling its virtues — vid’s after the break, per usual.

Source

This post is tagged: mundane task, security scanner, sophisticated vision, University of Surrey, vision system