Two quantitative measures of inlier distributions for precise fundamental matrix estimation

Seo, Jung-Kak; Hong, Hyun-Ki; Jho, Cheung-Woon; Choi, Min-Hyung

doi:10.1016/j.patrec.2004.01.014

Cited by 10 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, a large error increase was observed for the uncalibrated epipolar resampling methods, Hartley and Fusiello. This observation agrees well with previous investigations showing that the uncalibrated methods were more prone to tie-point errors in weak stereo geometry [ 10 , 11 ]. The uncalibrated methods also show larger image distortion.…”

Section: Results Of Performance Evaluationsupporting

confidence: 93%

“…Epipolar resampling has been studied extensively in the fields of computer vision and photogrammetry. In computer vision, epipolar resampling is achieved by a homographic transformation for sending epipoles of original images to infinity [ 5 , 6 , 7 , 8 , 9 , 10 , 11 ]. Depending on how the homography is estimated, epipolar resampling methods can be classified into two approaches: uncalibrated and calibrated cases.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of Computer Vision and Photogrammetric Approaches for Epipolar Resampling of Image Sequence

Kim

2016

Sensors

View full text Add to dashboard Cite

Epipolar resampling is the procedure of eliminating vertical disparity between stereo images. Due to its importance, many methods have been developed in the computer vision and photogrammetry field. However, we argue that epipolar resampling of image sequences, instead of a single pair, has not been studied thoroughly. In this paper, we compare epipolar resampling methods developed in both fields for handling image sequences. Firstly we briefly review the uncalibrated and calibrated epipolar resampling methods developed in computer vision and photogrammetric epipolar resampling methods. While it is well known that epipolar resampling methods developed in computer vision and in photogrammetry are mathematically identical, we also point out differences in parameter estimation between them. Secondly, we tested representative resampling methods in both fields and performed an analysis. We showed that for epipolar resampling of a single image pair all uncalibrated and photogrammetric methods tested could be used. More importantly, we also showed that, for image sequences, all methods tested, except the photogrammetric Bayesian method, showed significant variations in epipolar resampling performance. Our results indicate that the Bayesian method is favorable for epipolar resampling of image sequences.

show abstract

Section: Results Of Performance Evaluationsupporting

confidence: 93%

Section: Introductionmentioning

confidence: 99%

Comparison of Computer Vision and Photogrammetric Approaches for Epipolar Resampling of Image Sequence

Kim

2016

Sensors

View full text Add to dashboard Cite

show abstract

“…In particular, estimation of a fundamental matrix is dependent on the homogenous points, which should represent variation of the image owing to different viewpoints, and it is often suggested that evenly distributed homogenous points would always promise a correct solution [21]. To select well-distributed homogenous control points, a virtual grid is drawn in one image plane and the grid vertices will be initially chosen as candidate control points as shown in Fig.…”

Section: Rough 3d Reconstruction Of Homogenous Control Pointsmentioning

confidence: 99%

Pre-calibration-free 3D shape measurement method based on fringe projection

Zhong¹,

Li²,

Li³

et al. 2016

Opt. Express

View full text Add to dashboard Cite

This paper presents a pre-calibration-free 3D shape measurement method based on fringe projection. Unlike ordinary methods, it performs calibration and 3D shape measurement concurrently. The captured phase-coded fringe images are utilized to obtain homogenous control points from two camera viewpoints, and the rough 3D structure of these points can be retrieved. Further, a constrained non-linear least-squares optimization model is established to determine the in situ geometry of the optical components, and then, the 3D scene is reconstructed. This method provides an accurate 3D shape measurement capability even during disturbance of the optical geometry. Moreover, not requiring a preliminary calibration process makes the system ultra-flexible. The performance of this method was verified by experiments.

show abstract

“…15(a)). Therefore, consideration of the distribution/density of inliers can result in a more precise estimation of the fundamental matrix (Seo et al., 2004).…”

Section: Resultsmentioning

confidence: 99%

A new productive framework for point‐based matching of oblique aircraft and UAV‐based images

Baş

2021

The Photogrammetric Record

View full text Add to dashboard Cite

A novel productive framework for point‐based feature matching of oblique aircraft and UAV imagery is presented. The proposed framework makes use of the powerful AKAZE descriptor for feature extraction and an iterative scheme is developed to construct as many tentative matches as possible. During the iterations, cross checks, together with Lowe’s nearest‐next distance ratio test, are used to filter erroneous matches. In order to extract putative matches from the tentative matches, three robust approaches, including graph‐cut RANSAC, are evaluated along with the epipolar constraint enforced between the two datasets. The developed framework was validated using the ISPRS image orientation benchmark dataset and yielded successful results in terms of matching precision, even for some difficult cases. The results also outperformed the results of previously developed approaches in the same context.

show abstract

Two quantitative measures of inlier distributions for precise fundamental matrix estimation

Cited by 10 publications

References 9 publications

Comparison of Computer Vision and Photogrammetric Approaches for Epipolar Resampling of Image Sequence

Comparison of Computer Vision and Photogrammetric Approaches for Epipolar Resampling of Image Sequence

Pre-calibration-free 3D shape measurement method based on fringe projection

A new productive framework for point‐based matching of oblique aircraft and UAV‐based images

Contact Info

Product

Resources

About