Vision Plus

Algorithms to improve visual analysis of surface reconstructions.

Improvement of technology in scanning operations with laserscans and new devices for the creation of 3D datasets of surfaces leads to new challenges for visual exploration. In Vision Plus we started to improve our internal algorithms to handle scan data with much higher details than we used to have before.

We adapted state of the art algorithms like hardware tessellation to use the power of current 3D graphics technology to highly increase the performance for the exploration of pre-processedand textured ordered point clouds. We decreased the basic tessellation of the datasets to minimize the needed amount of graphics memory. The details were kept with displacement maps. This way we could increase the speed of the visual representation by a factor of about five without losing any essential information. In long terms, we could create a basis to explore much bigger laserscan datasets, which is possibly used in other projects of our group.