Skip to main navigation Skip to search Skip to main content

Efficient large-scale photometric reconstruction using Divide-Recon-Fuse 3D Structure from Motion

  • Yueming Yang
  • , Ming Ching Chang
  • , Longyin Wen
  • , Peter Tu
  • , Honggang Qi
  • , Siwei Lyu
  • SUNY Albany
  • General Electric
  • University of Chinese Academy of Sciences

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

7 Scopus citations

Abstract

We propose an efficient framework for large-scale 3D reconstruction from a large set of photos following the Structure-from-Motion (SfM) paradigm with divide-conquer and fusion. Our main novelty is to ensure commonality from overlaps between image sets corresponding to their reconstructions, which facilitates effective stitching and fusion. Specifically, such commonality is ensured by selecting a set of duplicated images (which are termed anchor images) in adjacent image sets prior to the 3D reconstruction. The anchor images can assist accurate fusion of the 3D point clouds. We describe an efficient RANSAC scheme for pairwise stitching. Our method is intuitively scalable to large site reconstruction via subdivision and fusion following a graph construct. We further describe another RANSAC algorithm to improve loop closure in our anchor image approach. Experimental results on reconstructing a large portion of a university campus demonstrate the efficacy of our method.

Original languageEnglish
Title of host publication2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages180-186
Number of pages7
ISBN (Electronic)9781509038114
DOIs
StatePublished - Nov 7 2016
Event13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016 - Colorado Springs, United States
Duration: Aug 23 2016Aug 26 2016

Publication series

Name2016 13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016

Conference

Conference13th IEEE International Conference on Advanced Video and Signal Based Surveillance, AVSS 2016
Country/TerritoryUnited States
CityColorado Springs
Period08/23/1608/26/16

Fingerprint

Dive into the research topics of 'Efficient large-scale photometric reconstruction using Divide-Recon-Fuse 3D Structure from Motion'. Together they form a unique fingerprint.

Cite this