Scaling point set registration in 3D across thread counts on multicore and hardware accelerator platforms through autotuning for large scale analysis of scientific point clouds

In this article, we present an autotuning approach applied to systematic performance engineering of the EM-ICP (Expectation-Maximization Iterative Closest Point) algorithm for the point set registration problem. We show how we were able to exceed the performance achieved by the reference code through multiple dependence transformations and automated procedure of generating and evaluating numerous implementation variants. Furthermore, we also managed to exploit code transformations that are not that common during manual optimization but yielded better performance in our tests for the EM-ICP algorithm. Finally, we maintained high levels of performance rate in a portable fashion across a wide range of HPC hardware platforms including multicore, many-core, and GPU-based accelerators. More importantly, the results indicate consistently high performance level and ability to move the task of data analysis through point-set registration to any modern compute platform without the concern of inferior asymptotic efficiency.

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

[2]  Pritish Narayanan,et al.  Deep Learning with Limited Numerical Precision , 2015, ICML.

[3]  Sergei V. Kalinin,et al.  Direct observation of ferroelectric field effect and vacancy-controlled screening at the BiFeO3/LaxSr1-xMnO3 interface. , 2014, Nature materials.

[4]  H. Chui,et al.  A feature registration framework using mixture models , 2000, Proceedings IEEE Workshop on Mathematical Methods in Biomedical Image Analysis. MMBIA-2000 (Cat. No.PR00737).

[5]  Richard G. Forbes,et al.  Atom-Probe Tomography: The Local Electrode Atom Probe , 2000 .

[6]  Jack J. Dongarra,et al.  Experiences in autotuning matrix multiplication for energy minimization on GPUs , 2015, Concurr. Comput. Pract. Exp..

[7]  Anand Rangarajan,et al.  A new point matching algorithm for non-rigid registration , 2003, Comput. Vis. Image Underst..

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

[9]  Anand Rangarajan,et al.  A new algorithm for non-rigid point matching , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[10]  Roger W. Hockney,et al.  F1/2: a Parameter to Characterize Memory and Communication Bottlenecks , 1989, Parallel Comput..

[11]  Karl Sohlberg,et al.  Origin of anomalous Pt-Pt distances in the Pt/alumina catalytic system. , 2004, Chemphyschem : a European journal of chemical physics and physical chemistry.

[12]  Jack J. Dongarra,et al.  Search Space Generation and Pruning System for Autotuners , 2016, 2016 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW).

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

[14]  Zhengyou Zhang,et al.  Iterative point matching for registration of free-form curves and surfaces , 1994, International Journal of Computer Vision.

[15]  Jack J. Dongarra,et al.  Towards numerical benchmark for half-precision floating point arithmetic , 2017, 2017 IEEE High Performance Extreme Computing Conference (HPEC).

[16]  Miguel Á. Carreira-Perpiñán,et al.  Non-rigid point set registration: Coherent Point Drift , 2006, NIPS.

[17]  Marc Levoy,et al.  Efficient variants of the ICP algorithm , 2001, Proceedings Third International Conference on 3-D Digital Imaging and Modeling.

[18]  R. Kondor,et al.  On representing chemical environments , 2012, 1209.3140.

[19]  Andrew W. Fitzgibbon.  Robust registration of 2D and 3D point sets , 2003, Image Vis. Comput..

[20]  James S. Duncan,et al.  A Robust Point Matching Algorithm for Autoradiograph Alignment , 1996, VBC.

[21]  Hongdong Li,et al.  The 3D-3D Registration Problem Revisited , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[22]  Radu Bogdan Rusu,et al.  3D is here: Point Cloud Library (PCL) , 2011, 2011 IEEE International Conference on Robotics and Automation.