CarSim: An Automatic 3D Text-to-Scene Conversion System Applied to Road Accident Reports

CarSim is an automatic text-to-scene conversion system. It analyzes written descriptions of car accidents and synthesizes 3D scenes of them. The conversion process consists of two stages. An information extraction module creates a tabular description of the accident and a visual simulator generates and animates the scene. We implemented a first version of CarSim that considered a corpus of texts in French. We redesigned its linguistic modules and its interface and we applied it to texts in English from the National Transportation Safety Board in the United States. 1 Text -to-Scene Conversion Text-to-scene conversion consists in creating a 2D or 3D geometric description from a natural language text. The resulting scene can be static or animated. To be converted, the text must be appropriate in some sense, that is, contains explicit descriptions of objects and events for which we can form mental images. Animated 3D graphics have some advantages for the visualization of information. They can reproduce a real scene more accurately and render a sequence of events. Automatic text-to-scene conversion has been investigated in a few projects. NALIG (Adomi et al., 1984; Di Manzo et al., 1986) is an early system that was designed to recreate static 2D scenes from simple phrases in Italian. WordsEye (Coyne and Sproat, 2001) is a recent and ambitious example. It features a large database of 3D objects that can be animated. CogViSys (Nagel, 2001; Arens et al., 2002) is aimed a visualizing descriptions of simple car maneuvers at crossroads. All these systems use apparently invented narratives.