Carsim: A System To Visualize Written Road Accident Reports As Animated 3D Scenes

This paper describes a system to create animated 3D scenes of car accidents from reports written in Swedish. The system has been developed using news reports of varying size and complexity. The text-to-scene conversion process consists of two stages. An information extraction module creates a structured representation of the accident and a visual simulator generates and animates the scene. We first describe the overall structure of the textto-scene conversion and the structure of the representation. We then explain the information extraction and visualization modules. We show snapshots of the car animation output and we conclude with the results we obtained. 1 Text-to-Scene Conversion As noted by Michel Denis, language and images are two different representation modes whose cooperation is needed in many forms of cognitive operations. The description of physical events, mathematical theorems, or structures of any kind using language is sometimes difficult to understand. Images and graphics can then help understand ideas or situations and realize their complexity. They have an indisputable capacity to represent and to communicate knowledge and are an effective means to represent and explain things, see (Kosslyn, 1983; Tufte, 1997; Denis, 1991). Narratives of a car accidents, for instance, often make use of space descriptions, movements, and directions that are sometimes difficult to grasp for most readers. We believe that forming consistent mental images are necessary to understand them properly. However, some people have difficulties in imagining situations and may need visual aids predesigned by professional analysts. In this paper, we will describe Carsim, a text-toscene converter that automates the generation of images from texts. 2 Related Work The conversion of natural language texts into graphics has been investigated in a few projects. NALIG (Adorni et al., 1984; Manzo et al., 1986) is an early example of them that was aimed at recreating static 2D scenes. One of its major goals was to study relationships between space and prepositions. NALIG considered simple phrases in Italian of the type subject, preposition, object that in spite of their simplicity can have ambiguous interpretations. From what is described in the papers, NALIG has not been extended to process sentences and even less to texts. WordsEye (Coyne and Sproat, 2001) is an impressive system that recreates 3D animated scenes from short descriptions. The number of 3D objects WordsEye uses – 12,000 – gives an idea of its ambition. WordsEye integrates resources such as the Collins’ dependency parser and the WordNet lexical database. The narratives cited as examples resemble imaginary fairy tales and WordsEye does not seem to address real world stories. CogViSys is a last example that started with the idea of generating texts from a sequence of video images. The authors found that it could also be useful to reverse the process and generate synthetic video sequences from texts. The logic engine behind the text-to-scene converter (Arens et al., 2002) is based on the Discourse Representation Theory. The system is limited to the visualization of single vehicle maneuvers at an intersection as the one described in this two-sentence narrative: A car came from Kriegstrasse. It turned left at the intersection. The authors give no further details on the text corpus and no precise description of the results.