Localization with Omnidirectional Images using the Radial Trifocal Tensor

Carlos Sagüés, A. C. Murillo, José Jesús Guerrero, Toon Goedemé, Tinne Tuytelaars, Luc Van Gool

Abstract

In this paper we present a technique to linearly recover 2D structure and motion in man made environments from three uncalibrated omnidirectional views. We use vertical lines from the scene which are projected as radial lines in the images and are automatically matched. The algorithm is based on a 1D radial trifocal tensor which encodes the relations of the three views and the projected lines. We include experiments with real images, which demonstrate the good performance of the method and its application to robotic tasks, such as robot localization based in a database of reference images or to obtain the initial values of robot and landmarks localization in SLAM algorithms
BibTeX
@inproceedings{sagues2006localization,
  author = {Carlos Sagüés and A. C. Murillo and José Jesús Guerrero and Toon Goedemé and Tinne Tuytelaars and Luc Van Gool},
  title = {{Localization with Omnidirectional Images using the Radial Trifocal Tensor}},
  booktitle = {IEEE International Conference on Robotics and Automation},
  year = {2006},
  doi = {10.1109/robot.2006.1641768},
}

Loading related content...

Figure