interesting announcment out of ibm alphaworks - Automatic indexing of multimedia content http://minformax.wordpress.com/2008/12/16/featureextr/ "Tackling the problem of dealing with large amounts of image and video data, this new technology on alphaWorks called Feature Extraction Service annotates multimedia content. It applies machine learning techniques across multiple visual features to automatically index visual content. These semantic annotations can then be used by applications for search, filtering, classification and other content management."