Contextual visual localization: cascaded submap classification, optimized saliency detection, and fast view matching
Please use this identifier to cite or link to this item:
http://hdl.handle.net/10045/23402
Title: | Contextual visual localization: cascaded submap classification, optimized saliency detection, and fast view matching |
---|---|
Authors: | Escolano, Francisco | Bonev, Boyan | Suau Pérez, Pablo | Aguilar, Wendy | Frauel, Yann | Sáez Martínez, Juan Manuel | Cazorla, Miguel |
Research Group/s: | Robótica y Visión Tridimensional (RoViT) | Laboratorio de Investigación en Visión Móvil (MVRLab) |
Center, Department or Service: | Universidad de Alicante. Departamento de Ciencia de la Computación e Inteligencia Artificial |
Keywords: | Contextual visual localization | Fast view-matching algorithm | Global localization problem | Minimal-complexity classifier | Saliency detector | Supervised classifier | Visual database |
Knowledge Area: | Ciencia de la Computación e Inteligencia Artificial |
Issue Date: | 2007 |
Publisher: | IEEE |
Citation: | ESCOLANO, Francisco, et al. "Contextual visual localization: cascaded submap classification, optimized saliency detection, and fast view matching". En: Proceedings of the 2007 IEEE/RSJ International Conference on Intelligent Robots and Systems : San Diego, CA, USA, Oct 29-Nov 2, 2007. Piscataway, N.J. : IEEE, 2007. ISBN 978-1-4244-0912-9, pp. 1715-1722 |
Abstract: | In this paper, we present a novel coarse-to-fine visual localization approach: contextual visual localization. This approach relies on three elements: (i) a minimal-complexity classifier for performing fast coarse localization (submap classification); (ii) an optimized saliency detector which exploits the visual statistics of the submap; and (iii) a fast view-matching algorithm which filters initial matchings with a structural criterion. The latter algorithm yields fine localization. Our experiments show that these elements have been successfully integrated for solving the global localization problem. Context, that is, the awareness of being in a particular submap, is defined by a supervised classifier tuned for a minimal set of features. Visual context is exploited both for tuning (optimizing) the saliency detection process, and to select potential matching views in the visual database, close enough to the query view. |
Sponsor: | This work was supported by Project DPI2005-01280 funded by the Spanish Government, and Project GV06/134 from Generalitat Valenciana. |
URI: | http://hdl.handle.net/10045/23402 |
ISBN: | 978-1-4244-0912-9 |
DOI: | 10.1109/IROS.2007.4399186 |
Language: | eng |
Type: | info:eu-repo/semantics/conferenceObject |
Rights: | © Copyright 2007 IEEE |
Peer Review: | si |
Publisher version: | http://dx.doi.org/10.1109/IROS.2007.4399186 |
Appears in Collections: | INV - RoViT - Comunicaciones a Congresos, Conferencias, etc. |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
2007_Cazorla_etal_IEEE_SanDiego.pdf | Versión revisada (acceso abierto) | 1,93 MB | Adobe PDF | Open Preview |
2007_Cazorla_etal_IEEE_SanDiego_final.pdf | Versión final (acceso restringido) | 1,05 MB | Adobe PDF | Open Request a copy |
Items in RUA are protected by copyright, with all rights reserved, unless otherwise indicated.