A general recursive linear method and unique solution pattern design for the perspective-n-point problem

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

55 Scopus Citations
View graph of relations

Author(s)

Detail(s)

Original languageEnglish
Pages (from-to)740-750
Journal / PublicationImage and Vision Computing
Volume26
Issue number6
Publication statusPublished - 2 Jun 2008

Abstract

In this paper, a new method for solving the perspective-n-point (PnP) problem is developed. With an auxiliary point, the linear method for the special case of four coplanar points is extended to find the coarse solutions for the general P3P problem. A recursive least square algorithm with a forgetting factor is introduced to find all the accurate solutions for the P3P problem. Then the algorithm is extended to the general PnP problem. The solution stability issues are investigated for the P3P, P4P and P5P problem, respectively. Furthermore, a pattern is designed to ensure unique solution for the PnP problem. Experiments are performed to verify the effectiveness of the proposed method. © 2007 Elsevier B.V. All rights reserved.

Research Area(s)

  • Pattern design, Perspective-n-point problem, Pose estimation, Recursive least square, Solution distribution, Solution stability, Three-dimensional sensing, Visual positioning