NOVEMBER 2, 2009--Typically, object recognition algorithms need to be "trained" using digital images in which objects have been outlined and labeled by hand. By looking at a million pictures of cars labeled "car," an algorithm can learn to recognize features shared by images of cars. The problem is that for every new class of objects -- trees, buildings, telephone poles -- the algorithm has to be trained all over again.
Antonio Torralba and his colleagues at MIT have developed an object recognition system that does not require any training. Nonetheless, it still identifies objects with 50% greater accuracy than the best prior algorithm. The system uses a modified version of a so-called motion estimation algorithm, a type of algorithm common in video processing. Since consecutive frames of video usually change very little, data compression schemes often store the unchanging aspects of a scene once, updating only the positions of moving objects. The motion estimation algorithm determines which objects have moved from one frame to the next. In a video, that's usually fairly easy to do: Most objects don't move very far in 1/30 of a second. Nor does the algorithm need to know what the object is; it just has to recognize, say, corners and edges, and how their appearance typically changes under different perspectives.
For more information, go to http://web.mit.edu.
--Posted by Conard Holton, Vision Systems Design, www.vision-systems.com
