Abstract
In this paper, a robust fundamental matrix estimation method based on epipolar geometric error criterion is proposed. First, the method removes outliers into the computation of the fundamental matrix instead of taking it as an independent processing step. The potential error corresponding points are eliminated by iteration to achieve the stable estimation of the fundamental matrix. Then, the epipolar geometry error criterion is used to identify outliers and the estimation results of the fundamental matrix are obtained during each iteration. The iterative process can converge quickly. Even if a large number of matched outliers are present, the calculated values will soon become stable. Experiments have been carried out for synthetic and real image pairs, which show that the proposed method performs very well in terms of robustness to noises and outliers. Additionally it has a low computational cost and is convenient for use in practical applications.
Highlights
Two perspective projective images of a single rigid scene are related by the epipolar geometry [1], which can be described by a matrix called fundamental matrix (F-matrix)
A very fast and robust method to estimate the fundamental matrix from image pairs is proposed in this paper
The results show that the new approach achieves a good performance by comparing to several other robust methods
Summary
Two perspective projective images of a single rigid scene are related by the epipolar geometry [1], which can be described by a matrix called fundamental matrix (F-matrix). The fundamental matrix is independent of the scene structure, and can be computed from correspondences of imaged scene points alone, without requiring knowledge of the cameras’ internal parameters or relative pose [2]. Estimating the fundamental matrix is a basic step for a wide variety of vision-based applications [3], [4], such as 3D reconstruction [5], camera-pair calibration [5], [6], object matching & tracking [7] object recognition [8], etc. The estimated F-matrix is useful for recovering the relative motion of camera, guiding correspondences establishment in stereo matching, and so on. Developing a method for fundamental matrix estimation with high efficiency and robustness is of great significance
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
