A multi-camera calibration method using a 3-axis frame and wand

Research output: Contribution to journalArticlepeer-review

24 Scopus citations

Abstract

Despite the recent advances in 3D camera technology, there is still a need to increase the accuracy of 3D reconstruction and a demand for a more user-friendly interface for camera calibration. To achieve this goal, the accuracy of the DLT method must be improved as well as the user convenience of the calibration method during motion capture using the wand. Therefore, in this study a new multi-camera calibration method for an optical motion capture system was developed. The proposed calibration procedure consists of two steps: (1) the calibration parameters are estimated using the DLT method from the three-axis calibration frame; and (2) the parameters estimated in the first step are improved iteratively through nonlinear optimization using a wand dance procedure. The objective function to be minimized is the 3D reconstruction error of the three-axis calibration frame and the difference between the actual distance and reconstructed distance of the marker at the wand. The proposed method was verified by comparing the RMS error and the mean difference of the proposed method, DLT method and wand method. In these experiments, the data obtained from a distance of 390mm between the markers and the wand was selected. In this analysis, no statistical significant difference between the proposed method and DLT method was observed, and the proposed method was more accurate than the other wand method. As a result, the proposed wand method holds great promise to increase the overall accuracy of the DLT algorithm and provide better user convenience.

Original languageEnglish
Pages (from-to)283-289
Number of pages7
JournalInternational Journal of Precision Engineering and Manufacturing
Volume13
Issue number2
DOIs
StatePublished - Feb 2012

Keywords

  • 3-axis wand
  • DLT
  • Motion capture
  • Multi-camera calibration
  • Wand calibration

Fingerprint

Dive into the research topics of 'A multi-camera calibration method using a 3-axis frame and wand'. Together they form a unique fingerprint.

Cite this