Dealing with degeneracy in essential matrix estimation

Decker, Peter; Paulus, Dietrich; Feldmann, Tobias

doi:10.1109/icip.2008.4712167

Cited by 7 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The only features closer than 10 meters from the camera, and thereby usable for the depth computation, were found from one side of the camera only. This resulted in a degenerated feature configuration and erroneous visual odometry solution [40]. The speed mean errors were 0.41 m/s for conventional visual odometry and 0.29 m/s for our solution and standard deviations 0.60 m/s and 0.43 m/s, respectively.…”

Section: A Resultsmentioning

confidence: 91%

Improving Computer Vision-Based Perception for Collaborative Indoor Navigation

et al. 2022

View full text Add to dashboard Cite

Section: A Resultsmentioning

confidence: 91%

Improving Computer Vision-Based Perception for Collaborative Indoor Navigation

et al. 2022

View full text Add to dashboard Cite

“…Most images in the database exhibit a strongly dominating quasi-planar surface. This induces the well-known coplanarity degeneracy in the computation of the essential matrix, as pointed out by Decker et al [9] and Irani et al [10]. Due to this degeneracy we propose our novel method, which adequately models this configuration and still computes a meaningful disparity map through SGM.…”

Section: Introductionmentioning

confidence: 91%

Detection and Segmentation of Quasi-Planar Surfaces Through Expectation Maximization Under a Planar Homography Constraint

Herbon

Schumann

Tönnies

et al. 2015

2015 12th Conference on Computer and Robot Vision

View full text Add to dashboard Cite

We propose a novel method capable of detecting and segmenting quasi-planar surfaces based on homography decomposition and Semi-Global-Matching without the need for extrinsic calibration or stereo rectification. Existing methods require coplanarity of all points on the dominant plane and are thus unsuited for unconstrained quasi-planar surfaces. In contrast to state of the art methods we can account for local depth variations. This is achieved by introducing a novel planar inter-image rectification technique, which enables us to perform block matching without epipolar rectification.Experiments are performed on two different databases. Firstly we quantitatively evaluate the general feasibility of our method on a database containing indoor and outdoor scenes with available ground truth data. Secondly we apply our method to the new publicly available HAWKwood database. Our experiments have shown that the true positive rate of our segmentation procedure exceeds 94.0% while the false positive rate is below 4.9%.

show abstract

“…In Ref. 16, the degeneracy is tested by successively testing motion models with decreasing degrees of freedom (DOF). If almost the same inlier support is obtained by the model with less DOF, then the data is degenerated for the previous.…”

Section: Detecting and Avoiding Planar Degeneraciesmentioning

confidence: 99%

Parallax beam: a vision-based motion estimation method robust to nearly planar scenes

et al. 2019

View full text Add to dashboard Cite

In computer vision, the epipolar geometry embeds the geometrical relationship between two views of a scene. This geometry is degenerated for planar scenes as they do not provide enough constraints to estimate it without ambiguity. Nearly planar scenes can provide the necessary constraints to resolve the ambiguity. But classic estimators such as the 5-point or 8-point algorithm combined with a random sampling strategy are likely to fail in this case because a large part of the scene is planar and it requires lots of trials to get a non-degenerated sample. However, the planar part can be associated with a homographic model and several links exist between the epipolar geometry and homographies. The epipolar geometry can indeed be recovered from at least two homographies or one homgraphy and two non-coplanar points. The latter fits a wider variety of scenes, as it is unsure to be able to find a second homography in the non-coplanar points. This method is called plane-and-parallax. The equivalence between the parallax and the epipolar lines allows to recover the epipole as their common intersection and so the epipolar geometry. Robust implementations of the method are rarely given and we encounter several limitations in our implementation. Noisy image features and outliers make the lines not to be concurrent in a common point. Also, off-plane features are unequally influenced by the noise level. We noticed that the bigger the parallax, the lesser the noise influence. We therefore propose a new model for the parallax, that takes into account the noise on the features location to cope with the previous limitations. We call our method the parallax beam. The method is validated on the KITTI vision benchmark and n synthetic scenes with strong planar degeneracy. The results show that the parallax beam improves the estimation of the camera motion in scene with planar degeneracy and remains usable when there is not any particular planar structure in the scene.

show abstract

Dealing with degeneracy in essential matrix estimation

Cited by 7 publications

References 6 publications

Improving Computer Vision-Based Perception for Collaborative Indoor Navigation

Improving Computer Vision-Based Perception for Collaborative Indoor Navigation

Detection and Segmentation of Quasi-Planar Surfaces Through Expectation Maximization Under a Planar Homography Constraint

Parallax beam: a vision-based motion estimation method robust to nearly planar scenes

Contact Info

Product

Resources

About