CompactNSearch is a C++ library for parallel computation of neighboring points in a fixed radius in a three-dimensional point cloud. The implemented algorithm is a variant of the Compact Hashing approach proposed by Ihmsen et al. The neighborhood information can be efficiently updated when the points spatially move. Moreover, the library offers the possibility to reorder the points (and other array-stored per point information) according to a space-filling Z curve to improve the cache efficiency in later queries or accesses.

License: MIT