A patent application lodged by Google in July 2007 but recently made public seeks to patent a method where by robots (computers) can read and understand text in images and video, notes Duncan Riley at TechCrunch. That would be a big step forward in indexing visual media, since there would be no need to manually attach tags to such visual media.
Basically, the patent covers a method whereby any visible text in an image--a street sign, for example--can be automatically indexed. Obviously, as with any of the developing Web-based technologies, there are privacy issues. As someone who has to work with lots of images, and spends lots of time wading through images that a search suggests are appropriate, and aren't, this is really helpful.
No comments:
Post a Comment