TU Darmstadt / ULB / TUbiblio

Folkioneer: Efficient Browsing of Community Geotagged Images on a Worldwide Scale

Mousselly-Sergieh, Hatem and Watzinger, Daniel and Huber, Bastian and Döller, Mario and Egyed-Zsigmond, Elöd and Kosch, Harald (2014):
Folkioneer: Efficient Browsing of Community Geotagged Images on a Worldwide Scale.
In: MultiMedia Modeling - 20th Anniversary International Conference, MMM 2014, Dublin, Ireland, January 6-10, 2014, Proceedings, Part II, Springer International Publishing, Dublin, Ireland, ISBN 978-3-319-04117-9,
DOI: 10.1007/978-3-319-04117-9_36, [Online-Edition: https://link.springer.com/chapter/10.1007%2F978-3-319-04117-...],
[Conference or Workshop Item]

Abstract

In this paper, we introduce Folkioneer, a novel approach for browsing and exploring community-contributed geotagged images. Initially, images are clustered based on the embedded geographical information by applying an enhanced version of the CURE algorithm, and characteristic geodesic shapes are derived using Delaunay triangulation. Next, images of each geographical cluster are analyzed and grouped according to visual similarity using SURF and restricted homography estimation. At the same time, LDA is used to extract representative topics from the provided tags. Finally, the extracted information is visualized in an intuitive and user-friendly manner with the help of an interactive map.

Item Type: Conference or Workshop Item
Erschienen: 2014
Creators: Mousselly-Sergieh, Hatem and Watzinger, Daniel and Huber, Bastian and Döller, Mario and Egyed-Zsigmond, Elöd and Kosch, Harald
Title: Folkioneer: Efficient Browsing of Community Geotagged Images on a Worldwide Scale
Language: English
Abstract:

In this paper, we introduce Folkioneer, a novel approach for browsing and exploring community-contributed geotagged images. Initially, images are clustered based on the embedded geographical information by applying an enhanced version of the CURE algorithm, and characteristic geodesic shapes are derived using Delaunay triangulation. Next, images of each geographical cluster are analyzed and grouped according to visual similarity using SURF and restricted homography estimation. At the same time, LDA is used to extract representative topics from the provided tags. Finally, the extracted information is visualized in an intuitive and user-friendly manner with the help of an interactive map.

Title of Book: MultiMedia Modeling - 20th Anniversary International Conference, MMM 2014, Dublin, Ireland, January 6-10, 2014, Proceedings, Part II
Publisher: Springer International Publishing
ISBN: 978-3-319-04117-9
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Ubiquitous Knowledge Processing
Event Location: Dublin, Ireland
Date Deposited: 31 Dec 2016 14:29
DOI: 10.1007/978-3-319-04117-9_36
Official URL: https://link.springer.com/chapter/10.1007%2F978-3-319-04117-...
Identification Number: TUD-CS-2014-1074
Export:

Optionen (nur für Redakteure)

View Item View Item