Locally Optimized RANSAC

Chum, Ondřej; Matas, Jiřı́; Kittler, Josef

doi:10.1007/978-3-540-45243-0_31

Cited by 654 publications

(494 citation statements)

References 12 publications

Supporting

Mentioning

469

Contrasting

Unclassified

Order By: Relevance

“…In our simulation, the real error (6) is divided by a factor of 3 when considering datasets of n = 12 correspondences instead of n = 8. This behaviour was reported before in [3], where the normalized eight-point algorithm was compared to its variant for 7 correspondences, which takes as solution F = F 1 +αF 2 , for α such that det(F 1 + αF 2 ) = 0 [7].…”

Section: The Rank-two Constraintsupporting

confidence: 64%

“…This is a requirement of the RANSAC [5] algorithm and variants. It can also be estimated using non-minimal but relatively small sets of data in the local random sampling method of Chum et al [3]. Finally, it is estimated using all the available inlier correspondences, yielding an initial solution for an overall error optimization, known as bundle adjustment [15]; even in this context, the size of the available dataset can be low due to typical problems in feature detection and matching.…”

Section: Overviewmentioning

confidence: 99%

See 1 more Smart Citation

Singular Vector Methods for Fundamental Matrix Computation

Espuny

Monasse

2014

Image and Video Technology

View full text Add to dashboard Cite

Abstract. The normalized eight-point algorithm is broadly used for the computation of the fundamental matrix between two images given a set of correspondences. However, it performs poorly for low-size datasets due to the way in which the rank-two constraint is imposed on the fundamental matrix. We propose two new algorithms to enforce the rank-two constraint on the fundamental matrix in closed form. The first one restricts the projection on the manifold of fundamental matrices along the most favorable direction with respect to algebraic error. Its complexity is akin to the classical seven point algorithm. The second algorithm relaxes the search to the best plane with respect to the algebraic error. The minimization of this error amounts to finding the intersection of two bivariate cubic polynomial curves. These methods are based on the minimization of the algebraic error and perform equally well for large datasets. However, we show through synthetic and real experiments that the proposed algorithms compare favorably with the normalized eight-point algorithm for low-size datasets.

show abstract

Section: The Rank-two Constraintsupporting

confidence: 64%

Section: Overviewmentioning

confidence: 99%

Singular Vector Methods for Fundamental Matrix Computation

Espuny

Monasse

2014

Image and Video Technology

View full text Add to dashboard Cite

show abstract

“…Next, we estimate the extrinsic pose (i.e., camera rotation and translation) with the P3P algorithm of Kneip et al (2011), yielding four estimates of the extrinsic pose. Poses that pass the T 1,1 test (Matas and Chum 2002) on a fourth sampled correspondence are improved further with a local optimization (Chum et al 2003), where we employ 10 inner RANSAC iterations on the inlier set. In each inner iteration, we sample 6 correspondences from the inlier set and use the non-minimal EPnP pose solver of Lepetit et al (2009) to calculate an improved extrinsic pose.…”

Section: Pose Estimationmentioning

confidence: 99%

Automatic Registration of Images to Untextured Geometry Using Average Shading Gradients

Plötz

Roth

2017

Int J Comput Vis

View full text Add to dashboard Cite

Many existing approaches for image-to-geometry registration assume that either a textured 3D model or a good initial guess of the 3D pose is available to bootstrap the registration process. In this paper we consider the registration of photographs to 3D models even when no texture information is available. This is very challenging as we cannot rely on texture gradients, and even shading gradients are hard to estimate since the lighting conditions are unknown. To that end, we propose average shading gradients, a rendering technique that estimates the average gradient magnitude over all lighting directions under Lambertian shading. We use this gradient representation as the building block of a registration pipeline based on matching sparse features. To cope with inevitable false matches due to the missing texture information and to increase robustness, the pose of the 3D model is estimated in two stages. Coarse pose hypotheses are first obtained from a single correct match each, subsequently refined using SIFT flow, and finally verified. We apply our algorithm to registering images of real-world objects to untextured 3D meshes of limited accuracy. Moreover, we show that registration can be performed even for paintings despite lacking photo-realism.

show abstract

“…Some of these strategies [2,3,4] aim to optimize the process of model verification, while others [5,6,7] seek to modify the sampling process in order to preferentially generate more useful hypotheses. While these efforts have shown considerable promise, none of them are directly applicable in situations where real-time performance is essential.…”

Section: Introductionmentioning

confidence: 99%

“…First, we provide a comparative analysis of the important RANSAC algorithms that have been proposed over the past few years [2,3,4,5,7,8,11]. We classify the various approaches based on the aspect of RANSAC that they seek to optimize, and evaluate their performance on synthetic and real data.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Analysis of RANSAC Techniques Leading to Adaptive Real-Time Random Sample Consensus

Raguram

Frahm

Pollefeys

2008

Lecture Notes in Computer Science

382

243

View full text Add to dashboard Cite

Abstract. The Random Sample Consensus (RANSAC) algorithm is a popular tool for robust estimation problems in computer vision, primarily due to its ability to tolerate a tremendous fraction of outliers. There have been a number of recent efforts that aim to increase the efficiency of the standard RANSAC algorithm. Relatively fewer efforts, however, have been directed towards formulating RANSAC in a manner that is suitable for real-time implementation. The contributions of this work are two-fold: First, we provide a comparative analysis of the state-of-the-art RANSAC algorithms and categorize the various approaches. Second, we develop a powerful new framework for real-time robust estimation. The technique we develop is capable of efficiently adapting to the constraints presented by a fixed time budget, while at the same time providing accurate estimation over a wide range of inlier ratios. The method shows significant improvements in accuracy and speed over existing techniques.

show abstract

Locally Optimized RANSAC

Cited by 654 publications

References 12 publications

Singular Vector Methods for Fundamental Matrix Computation

Singular Vector Methods for Fundamental Matrix Computation

Automatic Registration of Images to Untextured Geometry Using Average Shading Gradients

A Comparative Analysis of RANSAC Techniques Leading to Adaptive Real-Time Random Sample Consensus

Contact Info

Product

Resources

About