Multiple View Geometry in Computer Vision
Multiple View Geometry in Computer Vision
A Stereo Matching Algorithm Based on Multiresolution and Epipolar Constraint
ICIG '04 Proceedings of the Third International Conference on Image and Graphics
Hi-index | 0.00 |
In this paper an approach to estimate the topography of indoor environments is presented. The system is based on a single camera system only. The topography of areas is often a matter of particular interest and has a wide range of navigational application. The ground surface is assumed to be completely planar; this is usually satisfied in indoor environments like apartments or even industrial facilities. While the camera system is moving images are acquired cyclically. Pixel correspondences and homographic relations are used to estimate the height above ground level for each visible point. Finally a high resolution elevation map is computed through inverse perspective mapping.