Homography From Two Orientation- and Scale-Covariant Features

Baráth, Dániel; Kúkelová, Zuzana

doi:10.1109/iccv.2019.00118

Cited by 42 publications

(31 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…From five such point correspondences extended by the rotational angles of the features the fundamental matrix can be computed [2]. Similarly, the homography can be estimated by using two correspondences when including the corresponding rotational angles and scales of the features [4]. Of high interest are methods which use affine correspondences obtained by an affine-covariant feature detector, such as ASIFT [27] and MODS [26].…”

Section: Related Workmentioning

confidence: 99%

“…We choose a point in each random plane randomly, so there are also 50 points in the random planes. The corresponding affine transformation related to each point correspondence is calculated from the homography, which is estimated by using four projected image points from the same plane [4]. The baseline between two views is set to be 2 meters.…”

Section: Experiments On Synthetic Datamentioning

confidence: 99%

See 1 more Smart Citation

Minimal Solutions for Relative Pose With a Single Affine Correspondence

Guan

Zhao

et al. 2020

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

In this paper we present four cases of minimal solutions for two-view relative pose estimation by exploiting the affine transformation between feature points and we demonstrate efficient solvers for these cases. It is shown, that under the planar motion assumption or with knowledge of a vertical direction, a single affine correspondence is sufficient to recover the relative camera pose. The four cases considered are two-view planar relative motion for calibrated cameras as a closed-form and a least-squares solution, a closedform solution for unknown focal length and the case of a known vertical direction. These algorithms can be used efficiently for outlier detection within a RANSAC loop and for initial motion estimation. All the methods are evaluated on both synthetic data and real-world datasets from the KITTI benchmark. The experimental results demonstrate that our methods outperform comparable state-of-the-art methods in accuracy with the benefit of a reduced number of needed RANSAC iterations.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Experiments On Synthetic Datamentioning

confidence: 99%

Minimal Solutions for Relative Pose With a Single Affine Correspondence

Guan

Zhao

et al. 2020

2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

View full text Add to dashboard Cite

show abstract

“…Conventional methods for estimating the homography between two images first extract hand-crafted features, such as SIFT [30], SURF [5], HOG [11] or ORB [42] from each image. Recently, Barath et al [3] proposed two general constraints on the orientation-and scale-covariant features (e.g. SIFT).…”

Section: Yang MI Kang Zheng and Song Wangmentioning

confidence: 99%

Homography estimation along short videos by recurrent convolutional regression network

Mi¹,

Zheng²,

Wang³

2020

Mathematical Foundations of Computing

View full text Add to dashboard Cite

Many moving-camera video processing and analysis tasks require accurate estimation of homography across frames. Estimating homography between non-adjacent frames can be very challenging when their camera view angles show large difference. In this paper, we propose a new deep-learning based method for homography estimation along videos by exploiting temporal dynamics across frames. More specifically, we develop a recurrent convolutional regression network consisting of convolutional neural network (CNN) and recurrent neural network (RNN) with long short-term memory (LSTM) cells,followed by a regression layer for estimating the parameters of homography. In the experiments, we evaluate the proposed method on both the synthesized and real-world short videos. The experimental results verify that the proposed method can estimate the homographies along short videos better than several existing methods.

show abstract

“…Nevertheless, parts of the affine features can be obtained from widely-used feature detectors. For example, SIFT [22] and SURF [23] provide orientation-and scale-covariant features, which allows homography estimation from two correspondences [24]. ORB [25] provides oriented features, and has been successfully used for fundamental matrix estimation [26].…”

Section: Introductionmentioning

confidence: 99%

“…al. [24]. Our new solvers exploit the orientation and scale constraints on Euclidean homographies introduced in [24] and apply them for homography-based egomotion estimation with a known vertical direction.…”

Section: Introductionmentioning

confidence: 99%

Homography-Based Egomotion Estimation Using Gravity and SIFT Features

Ding

Baráth

Kúkelová

2021

Computer Vision – ACCV 2020

Self Cite

View full text Add to dashboard Cite

Camera systems used, e.g., in cars, UAVs, smartphones, and tablets, are typically equipped with IMUs (inertial measurement units) that can measure the gravity vector. Using the information from an IMU, the y-axes of cameras can be aligned with the gravity, reducing their relative orientation to a single DOF (degree of freedom). In this paper, we use the gravity information to derive extremely efficient minimal solvers for homography-based egomotion estimation from orientation-and scalecovariant features. We use the fact that orientation-and scale-covariant features, such as SIFT or ORB, provide additional constraints on the homography. Based on the prior knowledge about the target plane (horizontal/vertical/general plane, w.r.t. the gravity direction) and using the SIFT/ORB constraints, we derive new minimal solvers that require fewer correspondences than traditional approaches and, thus, speed up the robust estimation procedure significantly. The proposed solvers are compared with the state-of-the-art point-based solvers on both synthetic data and real images, showing comparable accuracy and significant improvement in terms of speed.

show abstract

Homography From Two Orientation- and Scale-Covariant Features

Cited by 42 publications

References 30 publications

Minimal Solutions for Relative Pose With a Single Affine Correspondence

Minimal Solutions for Relative Pose With a Single Affine Correspondence

Homography estimation along short videos by recurrent convolutional regression network

Homography-Based Egomotion Estimation Using Gravity and SIFT Features

Contact Info

Product

Resources

About