No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Control Points Based Semi-Dense Matching
Abstract
The common practice to get more matching points between two perspective images proceeds from coarse matching to dense matching. However, in some applications, for example, in the visualization of point-based D reconstruction results, one usually needs more matching points than those which could be obtained in coarse matching process to reveal sufficient structures, and does not need clouded matching points from dense matching process which is additionally of heavy computational load. In addition, it is sometimes desirable that the matching points could be evenly distributed across the whole image. The technique proposed in this note is a tentative step to this end. By adaptively adjusting the related parameters in the propagation stage, an appropriate number of matching points are obtained and distributed evenly. The extensive experiments validate our proposed new technique.
... (,dx 1 ,dx k ) describes the matching reliability of one point x 1 in an image and the corresponding point x k in another image. The matching reliability can be defined as (Zhu et al., 2005; Zhong and Zhang, 2002): ...
This paper compares several stereo image interest point detectors with respect to their repeatability and information content through experimental analysis. The Harris-Laplace detector gives better results than other detectors in areas of good texture; however, in areas of poor texture,
the Harris-Laplace detector may be not the best choice. A featurerelated filtering strategy is designed for the Harris-Laplace detector (as well as the standard Harris detector) to improve the repeatability and information content for imagery with both good and poor texture: (a) the local
information entropy is computed to describe the local feature of the image; and (b) the redundant interest points are filtered according to the interest strength and the local information entropy. After the filtering process, the repeatability and information content of the final interest
points are improved, and the mismatching then can be reduced. This conclusion is supported by experimental analysis with actual stereo images.
Camera calibration is an indispensable step in retrieving 3D metric information from 2D images. One classical self-calibration method is based on Kruppa equation derived from pairwise image correspondences. However, the calibration constraints derived from Kruppa equation are quadratic, which are computationally intensive and difficult to obtain initial values. In this paper, we propose a new initialization algorithm to estimate the unknown scalar in the equation, thus the camera parameters can be initialized linearly in a closed form and then optimized iteratively via the Kruppa constraints. We prove that the scalar can be uniquely recovered from the infinite homography and propose a practical method to estimate the homography from a physical or virtual plane located at far distance to the camera. Extensive experiments on synthetic and real images validate the effectiveness of the proposed method.
Kruppa equation based camera self-calibration is one of the classical problems in computer vision. Most state-of-the-art approaches
directly solve the quadratic constraints derived from Kruppa equations, which are computationally intensive and difficult
to obtain initial values. In this paper, we propose a new initialization algorithm by estimating the unknown scalar in the
equation, thus the camera parameters can be computed linearly in a closed form and then refined iteratively via global optimization
techniques. We prove that the scalar can be uniquely recovered from the infinite homography and propose a practical method
to estimate the homography from a physical or virtual plane located at a far distance to the camera. Extensive experiments
on synthetic and real images validate the effectiveness of the proposed method.
Purpose: Improve the accuracy and speed of the region-growth algorithm between two 2D images.Design/methodology/approach: The algorithm includes two parts: the selection of seeds points andpropagation. Some improvements are made in each one. For the first part, the best-first strategy is used to assurethe accuracy of seeds. The epipolar line constraint and continuity constraint reduce the double phase matchingcourse into single phase matching. For the second one, a dynamic and adaptive window is adopted instead ofthe large window.Findings: In the first section, the process of searching and the computational duties are decreased in largeextent. And in the second one, the adaptive window makes the searching course more efficient in time and space.It is really difficult to get the most suitable window to search for the points as soon as possible. If it can be easilygot, it will advance the efficiency of search. It is the future work.Practical implications: The method can be used in many different images, such as the structural images andthe facial images.Originality/value: The original value is the region-growth algorithm, and in this paper I made some bettermentsto advance the efficiency and accuracy.
ResearchGate has not been able to resolve any references for this publication.