The code is an implementation of Iterative Closest Point Registration algorithm. This algorithm is used to find the rigid transformation between Sensor Point Cloud and Model Point Cloud.
It also contains two datasets. One of armadillo and another of car. Armadillo dataset has no noise. But Car dataset has a guassian noise.
icp_with_options.cpp contains the main file to handle point clouds and call ICP method. ICP/ICP.h is the class for Iterative Closest Point Algorithm. KDTree/kd_tree.h is the class for KD-Trees. It has insert & kd_search_targets methods to insert elements in a KD-Tree and search a nearest neighbour for given target respectively.
mkdir build
make