Would we be able to use KD-Trees in applications where we have in our space some regions where no points can be in those regions. For example in a 2 dimensional spaces, in some rectangular regions we would not be able to have points, thus take them into consideration when cutting with splitting planes.According to Wikipedia :
k-d trees are a special case of binary space partitioning trees.
In Binary Space Partitioning there are plenty of applications where such a case might arise, such as path planning in robotics, and I have found publications around that, but nothing along the lines of k-D Trees, being a special case of BSP.
- Would it even be possible to use a k-D Trees implementation for such purpose?
- What would the result of our algorithm be?
- What would a modified kd tree data structure be like to represent such a case?
