You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We need to fix the NNSearch to be fast even if there is no point close by.
At the moment we just return node 0 when there is nothing in the sorounding grid cells.
Look into adapting Sascha's octree code.
The text was updated successfully, but these errors were encountered:
You can apply some "flood filling algorithm" so that every cell has some content like done here (instead of using only one integer per cell you should keep your arraylist per cell)
But I prefer being more precise, consuming less memory and accepting that some queries do not always return with a node so I've created a quadtree alternative with which I was able to import world wide data but still it has a meaningful size (less than 10% of the graph)
We need to fix the NNSearch to be fast even if there is no point close by.
At the moment we just return node 0 when there is nothing in the sorounding grid cells.
Look into adapting Sascha's octree code.
The text was updated successfully, but these errors were encountered: