A 3D segments based algorithm for heterogeneous data registration

International audienceCombining image and LiDAR draws increasing interest in surface reconstruction, city and building modeling for constructing 3D virtual reality models because of their complementary nature. However, to gain from this complementarity, these data sources must be precisely registered.The objective of this study is to propose a new primitive based registration algorithm that takes 3D segments as features in order to register heterogeneous data. The heterogeneity is both in data type (image and LiDAR) and acquisition platform (terrestrial and aerial).The basic idea of the proposed algorithm consists in defining a global robust distance between two segment sets and proposing a robust approach to minimize this distance based on RANSAC paradigm