Cluster-based 3D reconstruction of aerial video

Scott M. Sawyer, Karl Ni, Nadya T. Bliss

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

Abstract

Large-scale 3D scene reconstruction using Structure from Motion (SfM) continues to be very computationally challenging despite much active research in the area. We propose an efficient, scalable processing chain designed for cluster computing and suitable for use on aerial video. The sparse bundle adjustment step, which is iterative and difficult to parallelize, is accomplished by partitioning the input image set, generating independent point clouds in parallel, and then fusing the clouds and combining duplicate points. We compare this processing chain to a leading parallel SfM implementation, which exploits fine-grained parallelism in various matrix operations and is not designed to scale beyond a multi-core workstation with GPU. We show our cluster-based approach offers significant improvement in scalability and runtime while producing comparable point cloud density and more accurate point location estimates.

Original languageEnglish (US)
Title of host publication2012 IEEE Conference on High Performance Extreme Computing, HPEC 2012
DOIs
StatePublished - Dec 1 2012
Externally publishedYes
Event2012 IEEE Conference on High Performance Extreme Computing, HPEC 2012 - Waltham, MA, United States
Duration: Sep 10 2012Sep 12 2012

Publication series

Name2012 IEEE Conference on High Performance Extreme Computing, HPEC 2012

Other

Other2012 IEEE Conference on High Performance Extreme Computing, HPEC 2012
Country/TerritoryUnited States
CityWaltham, MA
Period9/10/129/12/12

ASJC Scopus subject areas

  • Software

Fingerprint

Dive into the research topics of 'Cluster-based 3D reconstruction of aerial video'. Together they form a unique fingerprint.

Cite this