| 摘要 |
A method is described for controlling the level of detail displayed in a computer generated screen display of a complex structure wherein the structure is modeled as a root volume bounding all of the parts of the structure, with the indi-vidual parts comprising sub-objects bounded in sub-volumes. The method begins by selecting the root volume (202) and then enters the step of determining whether or not all sub-objects within the selected volume are turned off (204). If they are, the entire display is culled. If they are not, the system then computes the screen coordinates of the selected volume (208). The method pro-ceeds to the step of determining whether or not the selected volume is on the screen or off. If it is off the screen, the entire volume is culled. If it is on screen, however, a determination is made as to whether or not all sub-objects are turned on (214). If all sub-objects are turned on, the number of screen pixels for the selected volume is computed (216). This screen size is compared to a user defined threshold (218) and if it is smaller than the threshold, the object is drawn as a bounding volume (220). If the selected object is larger than the threshold, a determination is made (230) as to whether or not the object includes sub-volumes. If it does not, the detailed object is drawn (232). If it does include sub-volumes, the system passes each sub-volume (240) back to the step of determining whether or not all sub- objects are turned off (204) to thereby determine for the entire root volume and all sub-volumes the appropriate level of detail to be displayed. The color for any given volume within a level of detail is determined by algebraic summation of the color of all sub-volumes within a parent volume weighted by the ratio of the sub-volume's volume to the parent's volume raised to the 2/3 power. This provides a crude inverse square fall off of intensity of the sub-volumes color to the parent's volume with pleasing results.
|
