ICCV '99 Proceedings of the International Workshop on Vision Algorithms: Theory and Practice
Using geometric corners to build a 2D mosaic from a set of image
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Direct Method for Visual Scene Reconstruction
VSR '95 Proceedings of the IEEE Workshop on Representation of Visual Scenes
Matching Widely Separated Views Based on Affine Invariant Regions
International Journal of Computer Vision
Scale & Affine Invariant Interest Point Detectors
International Journal of Computer Vision
A new method for important points extraction
CompSysTech '04 Proceedings of the 5th international conference on Computer systems and technologies
A Comparison of Affine Region Detectors
International Journal of Computer Vision
Automatic Panoramic Image Stitching using Invariant Features
International Journal of Computer Vision
| Hi-index | 0.00 |
The paper is devoted to the problem of image alignment which is one of the most important in image stitching. The basic requirement of this stage is the high alignment precision. As a rule algorithms for image registration are time consuming. In the paper we propose a fast algorithm for image registration. We suggest only sets of specially extracted control points to be involved in computations. The basic feature of these points, named equipotential points, is that they outline closed contours around brightest and darkest objects in the scene. An algorithm for combining, grouping and reduction of selected control points is also presented. Results, obtained using proposed points and algorithm, are compared with the results, obtained using other control points like corner and edge points.