【24h】

Calculation of Global Optimal Fundamental Matrix

机译:计算全局最优基础矩阵

获取原文

摘要

In stereo vision, a fundamental matrix is a 3 × 3 matrix to characterize a relationship between two cameras. We need a fundamental matrix to reconstruct a 3D structure from two images which are taken by two cameras. If we do not know a fundamental matrix, we will calculate it from the correspondence between characteristic points of two images. However, since an image contains noise generally, parameter estimation is necessary. For this estimation, it has been shown that the maximum likelihood estimation is preferable to the least square method. For calculation of a fundamental matrix, the maximum likelihood estimation is formulated as a constrained nonlinear programming problem. In this research, we propose a method for finding a global optimal solution of this problem, that is, a global optimal fundamental matrix.
机译:在立体声视觉中,基本矩阵是3×3矩阵,以表征两个相机之间的关系。我们需要一个基本矩阵来重建由两个摄像机拍摄的两个图像的3D结构。如果我们不知道基本矩阵,我们将从两个图像的特征点之间的对应中计算它。然而,由于图像通常包含噪声,因此需要参数估计。对于该估计,已经表明,最大似然估计是最不正方的方法。为了计算基本矩阵,将最大似然估计作为约束的非线性编程问题。在这项研究中,我们提出了一种寻找该问题的全局最佳解决方案的方法,即全球最佳基本矩阵。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号