Attentional landmark selection for visual SLAM
2006 (English)In: 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems, Vols 1-12, NEW YORK: IEEE , 2006, 2582-2587 p.Conference paper (Refereed)
In this paper, we introduce a new method to automatically detect useful landmarks for visual SLAM. A biologically motivated attention system detects regions of interest which "pop-out" automatically due to strong contrasts and the uniqueness of features. This property makes the regions easily redetectable and thus they are useful candidates for visual landmarks. Matching based on scene prediction and feature similarity allows not only short-term tracking of the regions, but also redetection in loop closing situations. The paper demonstrates how regions are determined and how they are matched reliably. Various experimental results on real-world data show that the landmarks are useful with respect to be tracked in consecutive frames and to enable closing loops.
Place, publisher, year, edition, pages
NEW YORK: IEEE , 2006. 2582-2587 p.
Feature extraction, Reliability, Tracking (position)
IdentifiersURN: urn:nbn:se:kth:diva-42059DOI: 10.1109/IROS.2006.281711ISI: 000245452402115ScopusID: 2-s2.0-34250676831ISBN: 978-1-4244-0258-8OAI: oai:DiVA.org:kth-42059DiVA: diva2:446115
IEEE/RSJ International Conference on Intelligent Robots and Systems. Beijing, PEOPLES R CHINA Date: OCT 09-13, 2006
QC 201110062011-10-062011-10-052012-01-10Bibliographically approved