Previous works that performed an automatic registration of multiple range views employed corresponding 3D matching points in order to compute relative
orientations.
However, it was often the case that false positives (corresponding 3D points) were detected that complied with the 3D mathematical model of the orientation calculated by using the Iterative Closest Point (ICP) algorithm.
In this work, I introduced a new method to detect correct 3D matches and to avoid the trap of generating false positives. The method employs firstly the removal of the false 2D matches. Subsequently, 3D matches are detected by using the one to one correspondences provided by the scanning device between the 3D points and their corresponding 2D image projections.
Due to its ability to detect the matching points in a fast manner, our method is suited for applications characterized by a 3D fast reconstruction demand. Experimental results
with real object demonstrate the effectiveness of our method.
Commenti