Division of Informatics
Forrest Hill & 80 South Bridge

Research Paper #804
Title:	Simultaneous Registration of Multiple Range Views Satisfying Global Consistency Constraints for Use in Reverse Engineering
Authors:	Eggert,D; Fitzgibbon,AW; Fisher,RB
Date:	May 1996
Presented:	Accepted for the International Conference on Pattern Recognition (August, 1996)
Keywords:	3-d surface registration, iterative closest point algorithm
Abstract:	When reverse engineering a CAD model it is necessary to integrate information from several views of an object into a common reference frame. Given a rough initial alignment of local shape data, further refinement is achieved using an improved version of the recently popular Iterative Closest Point algorithm. For determining data correspondence, the slowly merging sets are considered as a whole, eliminating the need for a distance threshold as when only view pairs have been considered in previous efforts. Incremental alignment adjustments are computed simultaneously for all data sets, resulting in a more globally iptimal set of transformations. Individual motion updates are computed using force-based optimization, with the data sets considered as connected by groups of springs. Experiments on both 2-D and 3-D data sets indicate that convergence is possible even for very rough initial positionings, and that the final registration accuracy typically approaches less than one quarter of the interpoint sampling resolution of the views.
Download:	POSTSCRIPT COPY

Division of InformaticsForrest Hill & 80 South Bridge