Abstract
We present an autotuning approach applied to exhaustive performance engineering of the EM-ICP algorithm for the point set registration problem with a known reference. We were able to achieve progressively higher performance levels through a variety of code transformations and an automated procedure of generating a large number of implementation variants. Furthermore, we managed to exploit code patterns that are not common when only attempting manual optimization but which yielded in our tests better performance for the chosen registration algorithm. Finally, we also show how we maintained high levels of the performance rate in a portable fashion across a wide range of hardware platforms including multicore, manycore coprocessors, and accelerators. Each of these hardware classes is much different from the others and, consequently, cannot reliably be mastered by a single developer in a short time required to deliver a close-to-optimal implementation. We assert in our concluding remarks that our methodology as well as the presented tools provide a valid automation system for software optimization tasks on modern HPC hardware.
Highlights
Many aspects of computer vision commonly uses algorithms for registration of point sets in three-dimensions (3D)
This is necessary in order to produce unambiguous descriptions of atomic-scale structures from large data sets originating in Atomic Probe Tomography (APT) [15, 19] and multimodal electron microscopy (EM) [14, 24]
Such data is in many ways similar, in its basic form, to visualization tasks but the registration of the points will be followed by derivation of physics, chemistry, or material science profiles that inform the scientists of emergent properties of the analyzed samples
Summary
Many aspects of computer vision commonly uses algorithms for registration of point sets in three-dimensions (3D). The incoming instrument data is in the form of sets of atomic coordinates (x, y, z) in three-dimensional (3D) space accompanied by identification of the atom type Such data is in many ways similar, in its basic form, to visualization tasks but the registration of the points will be followed by derivation of physics, chemistry, or material science profiles that inform the scientists of emergent properties of the analyzed samples. This serves as a motivation for fast and accurate implementations of the registration algorithms that are the subject of this paper
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
