Figure 9: Creation of the picture-based actor on Story Maker
Users can not only add a new actor, but they can add new
objects as well. Whenever NLP passes to graphics a
word that is not associated with a MetaLex, the Sketch
Pad UI can open to allow the user to draw or import a
picture of the object, name it, and enter synonyms for it.
When the name is entered, NLP also proceeds to
dynamically add the same MetaLex for all the synonyms
it can find for that name.
5. Conclusion
Story Maker makes the animating of natural language
easy and fun. It is a powerful tool that allows natural
language to guide graphic presentation. Without NLP
technology, it would be very difficult for a user to have a
series of animated scenes generated automatically and
directly from a story that s/he enters on the computer in
unrestricted natural language. The novelty and the power
of the tool lie in the fact that the two technologies,
Graphics and NLP, coexist and collaborate in the same
application.
The current prototype has much room for growth.
Currently, NLP is not passing information about
attributes of objects (e.g., ‘a big/red/small/etc. car’) to the
graphics component. However, the linguistic analysis
contains such information, and much more. Similarly, it
only passes information about action verbs to the
graphics component. We plan to enable the tool to handle
sentences that don’t express an action, but a change of
state (e.g., the man became fat/become taller/etc.). NLP
already contains information on the nature of the verb; it
is just a matter of equipping the graphic component to
handle such information. In addition, we will work on
the translation of prepositions (e.g. on/under/besides) into
graphics (e.g., the man put the book under/on/besides the
table.).
5
Of course, understanding and representing the
semantics of spatial expressions in natural language are
very difficult problems. Initially, we will only be able to
pass along fairly simple spatial (and temporal)
information to the graphics component
From the point of view of the graphics component, we
would like to allow users to animate a stick figure (most
likely in 2D for easy use), or perhaps a robot, when a
particular action/behavior requested from users is not in
the library. The animation described by the user on the
5
See Winograd (1999) for related work.
stick figure would then be added to the behavior library
and applied to all actors in future stories. Another
interesting extension for Pen/Ink technology would be to
allow users to draw the actor body (in 2D) over a stick
figure template on which each body part can be easily
recognized by the graphics engine and animated. Of
course, no matter how a new action is added, the user
would be able to name the action, add his own synonyms
for that name, and have the NLP system automatically
extract synonyms as well.
With our plans to enable users to customize/create their
own graphics and name them, the tool will increase in
power. We also plan to integrate speech technology into
the tool so that users can tell and hear their stories while
seeing them on the tool. We see great potential in Story
Maker, not only for linking Graphics and NLP but also
for integrating technology from different fields into one
platform.
ACKNOWLEDGMENTS
We thank the people in the NLP group for their feedback
and support; especially, Lucy Vanderwende, Deborah
Coughlin, and Gary Kacmarcik.
References
Fellbaum, C. (ed.), 1998. WordNet: An Electronic
Lexical Database, Cambridge, MA, MIT Press.
Available at: http://www.cogsci.princeton.edu/~wn/.
Heidorn, G. E., 1988. Intelligence Writing Assistance. In
Dale R., Moisl H., and Somers H. (eds.), A Handbook
of Natural Language Processing: Techniques and
Applications for the Processing of Language as Text.
Marcel Dekker, New York, 181-207.
Liberman, H., Rosenzweig, E., and Singh, P. Aria (2001).
An Agent for Annotating and Retrieving Images.
IEEE Computer, July 2001, 57-61.
Lieberman, H., Liu, H., 2002. Adaptive Linking between
Text and Photo Using Common Sense Reasoning.
MIT Media Lab.
Winograd, T., 1999. A Procedural Model of Language
Understanding. In
D. Cummins and R. Cummins (eds.)
Minds, Brains, Computers: The Foundations of
Cognitive Science, Blackwell.