This article introduces a novel approach for finding a rigid transformation that coarsely aligns two 3D point clouds.\r\nThe algorithm performs an iterative comparison between 2D descriptors by using a purpose-designed similarity\r\nmeasure in order to find correspondences between two 3D point clouds sensed from different positions of a freeform\r\nobject. The descriptors (named with the acronym CIRCON) represent an ordered set of radial contours that\r\nare extracted around an interest-point within the point cloud. The search for correspondences is done iteratively,\r\nfollowing a cell distribution that allows the algorithm to converge toward a candidate point. Using a single\r\ncorrespondence an initial estimation of the Euclidean transformation is computed and later refined by means of a\r\nmultiresolution approach. This coarse alignment algorithm can be used for 3D modeling and object manipulation\r\ntasks such as ââ?¬Å?Bin Pickingââ?¬Â when free-form objects are partially occluded or present symmetries
Loading....