When using the triangulation, we observe strong geometric distortions for some data sets. This is particularly the case with official LOD2 datasets. These distortions can usually be avoided by skipping the triangulation.
The following graphic gives an example of these distortions
Experiments have shown that a translation of the CityGML data set can lead to an improvement of the results. However this is not the case for all data sets with this issue.
An examination of the CityGML data sets has shown that the error occurs with data sets in which the coordinates only have 2-3 decimal places.
When using the triangulation, we observe strong geometric distortions for some data sets. This is particularly the case with official LOD2 datasets. These distortions can usually be avoided by skipping the triangulation. The following graphic gives an example of these distortions
Experiments have shown that a translation of the CityGML data set can lead to an improvement of the results. However this is not the case for all data sets with this issue. An examination of the CityGML data sets has shown that the error occurs with data sets in which the coordinates only have 2-3 decimal places.