Visual planning: a practical approach to automated presentation design

  • Authors:
  • Michelle X. Zhou

  • Affiliations:
  • IBM T. J. Watson Research Center, Hawthorne, NY

  • Venue:
  • IJCAI'99 Proceedings of the 16th international joint conference on Artifical intelligence - Volume 1
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

Based on a set of design principles, automated visual presentation systems promise to simplify an application programmer's design tasks by automatically constructing appropriate visual explanations for different information. However, these automated presentation systems must be equipped with a powerful inference approach to suit practical applications. Here, we present a planning-based, practical inference approach that can design a series of connected visual presentations in interactive environments. Our emphasis here is on a set of important visual planning features and how they facilitate visual design. This set of features includes a knowledge-rich representation of visual planning variables and constraints, a novel object-decomposition model that can be used with action decomposition to simplify the visual synthesis process, and practical temporal and spatial reasoning capabilities to facilitate coherent visual design and presentation. In addition, we have implemented our visual planning approach in a visual planner called PREVISE, as part of our automated presentation testbed system. A set of examples is also given to illustrate the necessity and utility of our visual planning approach.