Topology Estimation Using Bundle Adjustment

Part of the Springer Tracts in Advanced Robotics book series (STAR, volume 82)

Abstract

As discussed in the previous chapter, obtaining the topology is a requirement for getting globally coherent mosaics from image information alone.

In this chapter, a generic framework for FIM is proposed. This framework is capable of obtaining a topology with a reduced number of matching attempts, and the best possible trajectory estimate. Innovative aspects include the use of a fast image similarity criterion combined with a Minimum Spanning Tree (MST) solution to obtain a tentative initial topology. This topology is then improved by attempting image matching over the pairs for which there is more overlap evidence. Unlike the standard approaches for large-area mosaicing, the proposed framework is able to deal naturally with cases where time-consecutive images cannot be matched successfully, such as completely unordered image sets.

Keywords

Minimum Span Tree Image Pair Image Match Virtual Link Bundle Adjustment 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag GmbH Berlin Heidelberg 2013

Authors and Affiliations

  1. 1.Department of Mathematical EngineeringYildiz Technical UniversityIstanbulTurkey
  2. 2.Computer Vision and Robotics GroupUniversity of GironaGironaSpain

Personalised recommendations