Publications

Two useful invariants in stereo vision and their applications

WeiBing Tong, Moulay Akhloufi, Vladimir Polotski and Paul Cohen
More on this project...
Abstract - Invariants can be used to recognize objects in computer vision. In this paper, an invariant of points in stereo vision based on a reference point and two homographies (two planar objects) is identified and compared with the method proposed by Shashua. The other invariant of lines in stereo vision is derived from one homography and one reference line. Basically they are cross ratios in projective geometry, and very useful in vision computations.

Bibtex:

@inproceedings{Tong813,
author    = { WeiBing Tong and Moulay Akhloufi and Vladimir Polotski and Paul Cohen },
title     = { Two useful invariants in stereo vision and their applications },
booktitle = { 6th International Conference on Fuzzy Theory and Technology },
volume    = { 4 },
series    = { International Joint Conference of Information Sciences (JCIS’98) },
pages     = { 398-400 },
year      = { 1998 },
month     = { October },
language  = { English },
web       = { http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.48.4681 }
}

Last modification: 2010/02/21 by akhloufi

		Home \| About \| People \| Research \| Publications \| Events \| Profile
		©2002-. Computer Vision and Systems Laboratory. All rights reserved