Smart objects

From Wikipedia, the free encyclopedia
Jump to: navigation, search

A Smart Object is an object that can describe its own possible interactions. The concept was introduced by Marcelo Kallman and Daniel Thalmann [1][2] The opposite approach to Smart Objects are 'plain' objects that do not provide this information. The additional information provided by this concept enables far more general interaction schemes, and can greatly simplify the planner of an artificial intelligence agent.

Information that is provided by a Smart Object includes:

  • Object properties: physical properties and a text description
  • Interaction information: position of handles, buttons, grips, ...
  • Object behavior: different behaviors based on state variables
  • Agent behaviors: description of the behavior an agent should follow when using the object

Some versions of smart objects also include animation information in the object information, however this is not an efficient approach, as this may make objects innapropriately oversized.[3]


  1. ^ Marcelo Kallman, Daniel Thalmann: "Modeling Objects for Interaction Tasks", Proc. Eurographics Workshop on Animation and Simulation, Springer, 1998, pp.73-86.
  2. ^ In ca.2005 Lev Manovich described "smart objects" as "objects connected to the Net; objects that can sense their users and display 'smart' behaviour." Cf. Lev Manovich. The poetics of urban media surfaces. First Monday, Special Issue #4: Urban Screens: Discovering the potential of outdoor screens for urban society. [2005?]
  3. ^ Pieter Jorissen, Maarten Wijnants, Wim Lamotte: "Dynamic Interactions in Physically Realistic Collaborative Virtual Environments"

Further reading[edit]

External links[edit]

  • WorldCat publications about smart objects.

See also[edit]