“Text-to-Scene Conversion for Accident Visualization” by O’Kane, Carthy and Bertolotto

  • ©Maghnus O’Kane, Joe Carthy, and Michela Bertolotto



Entry Number: 063


    Text-to-Scene Conversion for Accident Visualization



    This document describes the creation of an automatic text-to-scene conversion system, AVis (Automatic Visualizer), for accident reports. Such reports vary from short text passages to long, complex documents describing the chain of events. The visualization of accidents is an important tool in Accident and Incident Analysis [Johnson 2002] For example, computer animations may be used as a training tool, for police and legal reconstruction purposes, or for aiding designers and engineers in analysing accidents and planning future development. Also, it can be difficult for non-expert readers to trace the increasingly complex arguments that are presented in accident reports. The automatic generation of 3D graphics from a text report can be useful for both expert and non-experts.


    1. Coyle, B., and Sproat, R. 2001. Wordseye: An automatic text-to-scene conversion system. Proceedings of the Siggraph Conference, Los Angeles.
    2. Johnson, C. 2002. Novel visualisation techniques for the presentation of accident reports. Tech. rep., Department of Computing Science, University of Glasgow, Glasgow, Scotland.
    3. Åkerberg, O., Svensson, H., Schulz, B., and Nugues, P. 2003. Carsim: An automatic 3d text-to-scene conversion system applied to road accident reports. In Research Notes and Demonstrations Conference Companion, 10th Conference of the European Chapter of the Association of Computational Linguistics, Association for Computational Linguistics, Budapest, Hungary, 191–194.



ACM Digital Library Publication:

Overview Page: