In this paper, a method for the
automated registration of two partly-overlapped dense point clouds
corresponding to a given object or scene is addressed. The proposed approach
has a number of significant advantages, such as the consideration of both geometric
and colour data, the automated search for the best suited points for
registration (based on their local geometry and colour properties), and the
automated estimation of the overlapping region. The proposed approach has been
tested on real scanner data and compared with others, showing better
performance in terms of accuracy and speed.
Software Instructions
Binary File (Windows)
Source Files
Data source examples of the paper