A Computational Approach to Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Computer graphics: principles and practice (2nd ed.)
Computer graphics: principles and practice (2nd ed.)
CVGIP: Image Understanding
Techniques for Assessing Polygonal Approximations of Curves
IEEE Transactions on Pattern Analysis and Machine Intelligence
Robust detection of lines using the progressive probabilistic Hough transform
Computer Vision and Image Understanding - Special issue on robusst statistical techniques in image understanding
Multiple view geometry in computer visiond
Multiple view geometry in computer visiond
Automatic line matching across views
CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Making Good Features Track Better
CVPR '98 Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
| Hi-index | 0.00 |
We propose an image representation method appropriate for real-time visual geometry applications. If the expressive power of raw segments is augmented with robust color information from their two sides, the most relevant geometric and photometric structure in the image can be concisely captured. In this paper we describe an efficient algorithm to compute this kind of representation, which can be successfully exploited in several projective geometry problems, such as 3D reconstruction, motion estimation or calibration, and in interpretation related tasks. We also show how these enhanced primitives are powerful enough to recover a very acceptable approximation of the original image, especially for partially structured scenes, like interior of buildings, man-made objects, and so on. The algorithm works at frame rate for medium size images (PAL/2), using low-cost hardware, such as an off-the-self image acquisition card and a standard PC. This makes it very useful as an on-line feature extraction method for robot visual navigation, where more elaborated (and slower) methods can not be used. We describe some applications of the algorithm in this kind of tasks.