Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

A system for fast and scalable point cloud indexing using task parallelism

 
: Bormann, Pascal; Krämer, Michel

:
Volltext urn:nbn:de:0011-n-6154692 (4.2 MByte PDF)
MD5 Fingerprint: 31d278052676410a49b2c8d47f716c2a
Erstellt am: 27.11.2020


Berretti, Stefano (Ed.); Fellner, Dieter W. (Proceedings Production Ed.) ; European Association for Computer Graphics -EUROGRAPHICS-:
Italian Chapter Conference 2020 - Smart Tools and Apps in Computer Graphics : Online Event, November 12 - 13, 2020
Goslar: Eurographics Association, 2020
ISBN: 978-3-03868-124-3
S.153-162
Italian Chapter Conference "Smart Tools and Applications in Computer Graphics" (STAG) <2020, Online>
Englisch
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IGD ()
Lead Topic: Visual Computing as a Service; Research Line: Computer graphics (CG); point clouds; acceleration structures; parallel algorithms; spatial data

Abstract
We introduce a system for fast, scalable indexing of arbitrarily sized point clouds based on a task-parallel computation model. Points are sorted using Morton indices in order to efficiently distribute sets of related points onto multiple concurrent indexing tasks. To achieve a high degree of parallelism, a hybrid top-down, bottom-up processing strategy is used. Our system achieves a 2.3x to 9x speedup over existing point cloud indexing systems while retaining comparable visual quality of the resulting acceleration structures. It is also fully compatible with widely used data formats in the context of web-based point cloud visualization. We demonstrate the effectiveness of our system in two experiments, evaluating scalability and general performance while processing datasets of up to 52.5 billion points.

: http://publica.fraunhofer.de/dokumente/N-615469.html