The Representation and Use of a Visual Lexicon in Automated Graphics Generation

Michelle X. Zhou and Steven K. Feiner

Department of Computer Science
Columbia University 500 West 120th St.,
450 CS Building
New York, NY 10027
+1 212 939 7000
{zhou, feiner}


Most automated graphics generation systems employ either a constructive or a parametric graphics synthesis approach. Constructive graphics synthesis is a deductive approach that builds visual presentations from scratch by gluing together the most basic visual variables. Conversely, parametric graphics synthesis defines a set of parametrized visual models and interprets the information to be presented through instantiation of the selected model. To increase efficiency, we have combined parametric and constructive approaches in a system called IMPROVISE. In this paper, we focus on the parametric aspect of our approach. We present a comprehensive, general, and extensible formalism to represent a visual lexicon for use in automated graphics generation. A visual lexicon is a collection of parametrized primitive visual objects that serve as building blocks for constructing more complex visual presentations. We also illustrate how this representation can be effectively employed to aid the selection and instantiation of a visual lexical item in the graphics generation process. Examples are given from IMPROVISE to demonstrate the representation and use of this visual lexicon.


The research described in this overview is supported in part by DARPA Contract DAAL01-94-K-0119, the Columbia Univeristy Center for Advanced Technology in High Performance Computing and Communications in Healthcare (funded by the New York State Science and Technology Foundation), the Columbia Center for Telecommunications Research under NSF Grant ECD-88-11111, ONR Contract N00014-94-1-0564, and a gift from Digital Image Division.

Permission to make digital/hard copies of all or part of this material for personal or classroot use is granted without fee provided that the copies are not made or distributed for profit or commericial advantage, the copyright notice, the title of the publication and its date appear, and the notice is given that copyright is by permission of the IJCAI. To copy otherwise, to republish, to post on servers or to redistribute to list requires specific permissions and/or fee.

Copyright © 1997 ACM 0-89791-839-8/96/01 .. $3.50

Get a PostScript copy of the full version.