Language selection

Government of Canada / Gouvernement du Canada

Search

A methodology for creating large scale reference models with known uncertainty for evaluating imaging solution

From National Research Council Canada

DOIResolve DOI: https://doi.org/10.1109/3DV.2014.104
AuthorSearch for: 1; Search for: 2; Search for: 2; Search for: 2; Search for: 1; Search for: 1; Search for:
Affiliation
  1. National Research Council of Canada. Information and Communication Technologies
  2. National Research Council of Canada. Measurement Science and Standards
FormatText, Article
Conference2014 Workshop on 3D Computer Vision in the Built Environment, December 8-11, 2014, Tokyo, Japan
Abstract
We propose a methodology for acquiring reference models with known uncertainty of complex building-sized objects. Those can be used to quantitatively evaluate the performance of passive 3D reconstruction when working at large scale. The proposed methodology combines the use of a time-of-flight scanner, a laser tracker, spherical artifacts and contrast targets. To demonstrate the soundness of the proposed approach, we built a reference model composed of a 3D model of exterior walls and courtyards of a 130m X 55m X 20m Building. The expanded uncertainty of the 3D reference model and the spatial resolution were calculated.
Publication date
PublisherIEEE
In
  • 2014 2nd International Conference on 3D Vision: 137–144.
LanguageEnglish
Peer reviewedYes
NPARC number21274273
Export citationExport as RIS
Report a correctionReport a correction (opens in a new tab)
Record identifier71595b3b-94db-4c81-b74d-80c60a3d4576
Record created2015-03-09
Record modified2020-04-22
Date modified: