Accurate Overlap Area Detection Using a Histogram and Multiple Closest Points

In this paper, we propose a novel ICP variant that uses a histogram in conjunction with multiple closest points to detect the overlap area between range images being registered. Tentative correspondences sharing similar distances are normally all within, or all outside, the overlap area. Thus, the overlap area can be detected in a bin by bin batch manner using a histogram. Using multiple closest points is likely to enlarge the distance difference for tentative correspondences in the histogram, and pull together the images being registered, facilitating the overlap area detection. Our experimental results based on real range images show that the performance of our proposed algorithm enhances the state of the art.

[1]  Radu Horaud,et al.  Hand Motion from 3D Point Trajectories and a Smooth Surface Model , 2004, ECCV.

[2]  Kim L. Boyer,et al.  Precision range image registration using a robust surface interpenetration measure and enhanced genetic algorithms , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Kari Pulli,et al.  Multiview registration for large data sets , 1999, Second International Conference on 3-D Digital Imaging and Modeling (Cat. No.PR00062).

[4]  Yonghuai Liu,et al.  Improving ICP with easy implementation for free-form surface matching , 2004, Pattern Recognit..

[5]  Ran Liu,et al.  Outlier Robust ICP for Minimizing Fractional RMSD , 2006, Sixth International Conference on 3-D Digital Imaging and Modeling (3DIM 2007).

[6]  Umberto Castellani,et al.  Sparse points matching by combining 3D mesh saliency with statistical descriptors , 2008, Comput. Graph. Forum.

[7]  Yonghuai Liu,et al.  Replicator Dynamics in the Iterative Process for Accurate Range Image Matching , 2009, International Journal of Computer Vision.

[8]  Oscar Cordón,et al.  A scatter search-based technique for pair-wise 3D range image registration in forensic anthropology , 2006, Soft Comput..

[9]  Xavier Pennec,et al.  Multi-scale EM-ICP: A Fast and Robust Approach for Surface Registration , 2002, ECCV.

[10]  Jiří Matas,et al.  Computer Vision - ECCV 2004 , 2004, Lecture Notes in Computer Science.

[11]  Sang Wook Lee,et al.  Maximum-Likelihood Registration of Range Images with Missing Data , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Eric Mjolsness,et al.  New Algorithms for 2D and 3D Point Matching: Pose Estimation and Correspondence , 1998, NIPS.

[13]  Takeshi Oishi,et al.  Flying Laser Range Sensor for Large-Scale Site-Modeling and Its Applications in Bayon Digital Archival Project , 2008, International Journal of Computer Vision.

[14]  Anikó Ekárt,et al.  Pre-registration of arbitrarily oriented 3D surfaces using a genetic algorithm , 2006, Pattern Recognit. Lett..

[15]  Mads Nielsen,et al.  Computer Vision — ECCV 2002 , 2002, Lecture Notes in Computer Science.

[16]  Yonghuai Liu,et al.  Constraints for closest point finding , 2008, Pattern Recognit. Lett..

[17]  Anand Rangarajan,et al.  Groupwise point pattern registration using a novel CDF-based Jensen-Shannon Divergence , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[18]  Takeshi Masuda,et al.  Log-polar height maps for multiple range image registration , 2009, Comput. Vis. Image Underst..

[19]  Kwang-Ho Bae,et al.  Evaluation of the Convergence Region of an Automated Registration Method for 3D Laser Scanner Point Clouds , 2009, Sensors.

[20]  H. Quynh Dinh,et al.  Multi-Resolution Spin-Images , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[21]  Wu Shin-Ting,et al.  Using simplified meshes for crude registration of two partially overlapping range images , 2007 .

[22]  D. W. Scott.  On optimal and data based histograms , 1979 .

[23]  Paul J. Besl,et al.  A Method for Registration of 3-D Shapes , 1992, IEEE Trans. Pattern Anal. Mach. Intell..