TU Darmstadt / ULB / TUbiblio

World-wide Scale Geotagged Image Dataset for Automatic Image Annotation and Reverse Geotagging

Mousselly-Sergieh, Hatem and Watzinger, Daniel and Huber, Bastian and Döller, Mario and Egyed-Zsigmond, Elöd and Kosch, Harald (2014):
World-wide Scale Geotagged Image Dataset for Automatic Image Annotation and Reverse Geotagging.
In: Proceedings of the 5th ACM Multimedia Systems Conference, ACM, Singapore, Singapore, In: MMSys '14, ISBN 978-1-4503-2705-3,
DOI: 10.1145/2557642.2563673, [Online-Edition: https://dl.acm.org/citation.cfm?id=2563673&dl=ACM&coll=DL],
[Conference or Workshop Item]

Abstract

In this paper, a dataset of geotagged photos on a world-wide scale is presented. The dataset contains a sample of more than 14 million geotagged photos crawled from Flickr with the corresponding metadata. To guarantee the spatial representativeness of the dataset, a crawling approach based on the small-world phenomena and the Flickr friendship's graph is applied. Furthermore, the noisiness of user-provided tags is reduced through an automatic tag cleaning approach. To enable efficient retrieval, photos in the dataset are indexed based on their location information using quad-tree data structure. The dataset can assists different applications, especially, search-based automatic image annotation and reverse geotagging.

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: World-wide Scale Geotagged Image Dataset for Automatic Image Annotation and Reverse Geotagging
Language: English
Abstract:

In this paper, a dataset of geotagged photos on a world-wide scale is presented. The dataset contains a sample of more than 14 million geotagged photos crawled from Flickr with the corresponding metadata. To guarantee the spatial representativeness of the dataset, a crawling approach based on the small-world phenomena and the Flickr friendship's graph is applied. Furthermore, the noisiness of user-provided tags is reduced through an automatic tag cleaning approach. To enable efficient retrieval, photos in the dataset are indexed based on their location information using quad-tree data structure. The dataset can assists different applications, especially, search-based automatic image annotation and reverse geotagging.

Title of Book: Proceedings of the 5th ACM Multimedia Systems Conference
Series Name: MMSys '14
Publisher: ACM
ISBN: 978-1-4503-2705-3
Uncontrolled Keywords: Flickr, geotagging, image dataset, quad-tree
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Ubiquitous Knowledge Processing
Event Location: Singapore, Singapore
Date Deposited: 31 Dec 2016 14:29
DOI: 10.1145/2557642.2563673
Official URL: https://dl.acm.org/citation.cfm?id=2563673&dl=ACM&coll=DL
Identification Number: TUD-CS-2014-1077
Export:

Optionen (nur für Redakteure)

View Item View Item