Date: March 2005
Creator: Borman, Andy; Mihalcea, Rada, 1974- & Tarau, Paul
Description: In this paper, the authors introduce PicNet, a Web-based system for augmenting semantic resources with illustrative images using volunteer contributions over the Web. PicNet seeks to build rich knowledge-bases that encode word/image associations, to the end of combining the advantages and power of both visual and linguistic representations as means of defining world concepts. In this paper, the authors address some of the issues encountered in identifying prototypical illustrations for various concepts, as well as issues related to the construction of such pictorial knowledge-bases with the help of Web users.
Contributing Partner: UNT College of Engineering