Dorffner, Lionel 
  
Besides this pure visualization it is often desirable to have possibilities to react on user inputs. These inputs can be 
mouse-clicks within or without the VRML world or just the movement of the virtual user walking through the VRML 
scene. By using special objects - called nodes - VRML provides different methods to react on user commands. In 
connection with an information system the most important nodes are "Anchor", "Inline", "Sensor" and "Script". 
[Dorffner L., 1999] 
Anchor: Similar to HTML VRML also allows access to other files using an Anchor-node which serves as 
hyperlink. Each object of a VRML scene can be used as Anchor. By clicking on such an object the scene is 
replaced by the selected Web address (HTML file, other VRML world, ...). In a 3D information system a mouse- 
click on a door, for instance, may change to a photo-model of the room behind this door. The hyperlink may 
concurrently also open a HTML page showing additional information about a selected painting or wall mosaic. 
Inline: The Inline-node allows the inclusion of another VRML file stored anywhere on the Web into the current 
VRML scene. So very large and complex photo-models can be split into smaller parts leading to a higher 
performance during visualization and simplifying the problems of updating the model. Besides, photo-models 
generated at different time-stamps can be visualized simultaneously for comparing variable states of an object. 
Sensor: Sensors are special VRML-nodes which generate events when the user interacts with them. This can 
happen by a mouse-click or by moving the cursor over an object of the VRML scene. This event is then passed to 
other nodes (for instance JavaScripts) which then launch some action (starting of an animation, playing of some 
audio- or video-clips). Events can also be generated just by the movements of the virtual user within the scene. In 
reaction on such movements different levels of object resolutions can be loaded, depending on the distance and 
visibility of parts of the model with respect to the current position of the virtual user. 
Script: Script nodes allow the world creator to define arbitrary behaviors, defined in any supported scripting 
language. The VRML 2.0 specification defines Script node bindings for the Java and JavaScript language. Scripts 
are used in animations where position and orientation of the camera have to be interpolated depending on 
predefined time-stamps. When measuring coordinates, angles or distances on the photo-model Scripts are used to 
perform the type conversions that become necessary to display the measured values on the screen (Figure 7). 
  
Photomodell analysieren - Netscai = {Of X 
File Edt View Go Window Help S 
« Location: file: ///D|/Id/Projekte/HagiaS ofia/VRML-Modelle/vrml-T extur/start.htrn x] XE What's Related = 
  
  
  
  
Batch Metscape — Fr 
Modell analysieren — 
  
  
  
Figure 7. Detail of the dome showing a modeled structural fissure (also visible in Figure 4) and the measured 
coordinates at the cursor position as well as a measured distance marked at the bri ght line. 
  
178 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B5. Amsterdam 2000. 
  
  
  
Up t 
it 1s 
Inte: 
emb 
requ 
VR! 
beu 
Dor 
Dui 
"Fo 
Ka; 
Kra 
Lai 
UN 
Zei