Pose And Motion Recovery From Feature Correspondences And a Digital Terrain Map
.
Pose and Motion Recovery from Feature Correspondences and a Digital Terrain Map.
IEEE Trans. Pattern Anal. Mach. Intell., 2006
Online Version
A pdf version is available for download.
Abstract
A novel algorithm for pose and motion estimation using corresponding features and a Digital Terrain Map is proposed. Using a Digital Terrain (or Digital Elevation) Map (DTM or DEM) as a global reference enables the elimination of the ambiguity present in vision-based algorithms for motion recovery. As a consequence, the absolute position and orientation of a camera can be recovered with respect to the external reference frame. In order to do this, the DTM is used to formulate a constraint between corresponding features in two consecutive frames. Explicit reconstruction of the 3D world is not required. When considering a number of feature points, the resulting constraints can be solved using non-linear optimization in terms of the position, orientation and motion. Such procedure requires an initial guess of these parameters that can be obtained from dead-reckoning or any other source. The feasibility of the algorithm is established through extensive experimentation. Performance is compared with state of the art alternative algorithm which intermediately reconstruct the 3D structure and then register it to the DTM. Clear advantage for the novel algorithm is demonstrated in variety of scenarios.
Keywords
Co-authors
Bibtex Entry
@article{LernerRR06a,
title = {Pose and Motion Recovery from Feature Correspondences and a Digital Terrain Map},
author = {Ronen Lerner and Ehud Rivlin and Hector Rotstein},
year = {2006},
journal = {IEEE Trans. Pattern Anal. Mach. Intell.},
keywords = {Pose estimation; Vision-based navigation; DTM; Structure from motion},
abstract = {A novel algorithm for pose and motion estimation using corresponding features and a Digital Terrain Map is proposed. Using a Digital Terrain (or Digital Elevation) Map (DTM or DEM) as a global reference enables the elimination of the ambiguity present in vision-based algorithms for motion recovery. As a consequence, the absolute position and orientation of a camera can be recovered with respect to the external reference frame. In order to do this, the DTM is used to formulate a constraint between corresponding features in two consecutive frames. Explicit reconstruction of the 3D world is not required. When considering a number of feature points, the resulting constraints can be solved using non-linear optimization in terms of the position, orientation and motion. Such procedure requires an initial guess of these parameters that can be obtained from dead-reckoning or any other source. The feasibility of the algorithm is established through extensive experimentation. Performance is compared with state of the art alternative algorithm which intermediately reconstruct the 3D structure and then register it to the DTM. Clear advantage for the novel algorithm is demonstrated in variety of scenarios.}
}