Combining Line and Point Correspondences for Homography Estimation

@inproceedings{Dubrofsky2008CombiningLA,
  title={Combining Line and Point Correspondences for Homography Estimation},
  author={Elan Dubrofsky and Robert J. Woodham},
  booktitle={ISVC},
  year={2008}
}
This paper presents a method to extend the normalized direct linear transform (DLT) algorithm for homography estimation. Previously, only point correspondences were used. Now, line correspondences can be included as well. This extension is point-centric in that the lines are normalized to fit with the normalization used for points. Therefore, this method will be most useful if there are more point correspondences than line correspondences. The main contribution of this paper is the derivation… 

A unified normalization method for homography estimation using combined point and line correspondences

Data normalization is an essential and imperative step when using the direct linear transform method for homography estimation. However, the existing data normalization methods only rely on either

On Line-based Homographies in Urban Environments

TLDR
A new method to extract only those line segments that belong to contours of buildings that is able to separate single facades in a building and the known crucial instability of line homography is overcome.

Calibration and object correspondence in camera networks with widely separated overlapping views

TLDR
A semi-automatic scene calibration method is proposed to deal with mapping a network of cameras with overlapped fields of view onto a single ground plane view, even when the overlap is not substantial.

Using Line and Ellipse Features for Rectification of Broadcast Hockey Video

TLDR
A new method to use line and ellipse features along with key-point based matches to estimate the homography is described, which is applicable to real world data and has the ability to track long sequences on the order of 1,000 frames.

A Two-Point Method for PTZ Camera Calibration in Sports

TLDR
A two-point method, which requires only two point correspondences given the prior knowledge of base location and orientation of a pan-tilt-zoom (PTZ) camera, is proposed, which achieves superior performance over the state-of-the-art method.

A methodology for precise camera calibration for data collection applications in urban traffic scenes

TLDR
This study presents the development details of a robust camera calibration approach based on integrating a collection of geometric information found in urban traffic scenes in a consistent optimization framework that was tested on six datasets obtained from urban intersections in British Columbia, California, and Kentucky.

Locally warping-based image stitching by imposing line constraints

TLDR
A locally warping-based image stitching by imposing line constraints to jointly estimate local homographies with point correspondences that yields more encouraging image stitching results in contrast with state-of-the-art methods.

Image Stitching by Line-guided Local Warping with Global Similarity Constraint

Camera Calibration for Urban Traffic Scenes: Practical Issues and a Robust Approach

TLDR
This study presents the development details of a robust camera calibration approach based on integrating a collection of geometric information found in urban traffic scenes in a consistent optimization framework and tested on six datasets obtained from urban intersections in British Columbia, California, and Kentucky.

Automatic initialization for broadcast sports videos rectification by Shervin

TLDR
This thesis comprises four key modules, namely preprocessing, keyframe selection, key frame matching, and frame-to-frame homography estimation, that work together in order to automatically initialize any homography estimator method that can be used for broadcast sports videos.

References

SHOWING 1-6 OF 6 REFERENCES

Real-Time Detection of Out-of-Plane Objects in Stereo Vision

TLDR
An automatic approach to detecting objects appearing in front of planar background using a planar homography estimated with high accuracy in an off-line initialization phase and normalized cross-correlation measure to measure the similarity between neighborhoods of overlapping pixels.

In Defense of the Eight-Point Algorithm

  • R. Hartley
  • Computer Science
    IEEE Trans. Pattern Anal. Mach. Intell.
  • 1997
TLDR
This paper shows that by preceding the eight-point algorithm with a very simple normalization (translation and scaling) of the coordinates of the matched points, results are obtained comparable with the best iterative algorithms.

AUTOMATIC RECTIFICATION OF LONG IMAGE SEQUENCES

TLDR
This paper addresses the problem of automatically computing homographies between successive frames in image sequences and compensating for the panning, tilting and zooming of the cameras by combining elements of two previous approaches.

Good features to track

  • Jianbo ShiCarlo Tomasi
  • Computer Science
    1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition
  • 1994
TLDR
A feature selection criterion that is optimal by construction because it is based on how the tracker works, and a feature monitoring method that can detect occlusions, disocclusions, and features that do not correspond to points in the world are proposed.

Multiple view geometry in computer vision

  • 2000

Multiple View Geomerty in Computer Vision, 2nd edn

  • Multiple View Geomerty in Computer Vision, 2nd edn
  • 2003