Browse wiki

Jump to: navigation, search
Viral: visual image retrieval and localization
Abstract New applications are emerging every day exNew applications are emerging every day exploiting the huge data volume in community photo collections. Most focus on popular subsets, e.g., images containing landmarks or associated to Wikipedia articles. In this work we are concerned with the problem of accurately finding the location where a photo is taken without needing any metadata, that is, solely by its visual content. We also recognize landmarks where applicable, automatically linking them to Wikipedia. We show that the time is right for automating the geo-tagging process, and we show how this can work at large scale. In doing so, we do exploit redundancy of content in popular locations—but unlike most existing solutions, we do not restrict to landmarks. In other words, we can compactly represent the visual content of all thousands of images depicting e.g., the Parthenon and still retrieve any single, isolated, non-landmark image like a house or a graffiti on a wall. Starting from an existing, geo-tagged dataset, we cluster images into sets of different views of the same scene. This is a very efficient, scalable, and fully automated mining process. We then align all views in a set to one reference image and construct a 2D scene map. Our indexing scheme operates directly on scene maps. We evaluate our solution on a challenging one million urban image dataset and provide public access to our service through our online application, VIRaL.ice through our online application, VIRaL.
Added by wikilit team Added on initial load  +
Collected data time dimension Cross-sectional  +
Comments We also recognize landmarks and points of interest by cross-validating location, photo title, frequent tags and geo-referenced Wikipedia article titles in an efficient online process.
Conclusion Sub-linear indexing is not typically exploSub-linear indexing is not typically exploited in landmark recognition applications, while geo-tags are not typically exploited in large scale 3D reconstruction applications. We have combined both, along with a novel scene representation that is directly encoded in our retrieval engine. The result is a considerable increase in retrieval performance, even compared to query expansion methods, at the cost of a slight increase in query time. Memory requirements for the index are also considerably reduced compared to a baseline system.bly reduced compared to a baseline system.
Data source Documents  + , Experiment responses  + , Wikipedia pages  +
Doi 10.1007/s11042-010-0651-7 +
Google scholar url http://scholar.google.com/scholar?ie=UTF-8&q=%22Viral%3A%2Bvisual%2Bimage%2Bretrieval%2Band%2Blocalization%22  +
Has author Yannis Kalantidis + , Giorgos Tolias + , Yannis Avrithis + , Marios Phinikettos + , Evaggelos Spyrou + , Phivos Mylonas + , Stefanos Kollias +
Has domain Geography + , Information science +
Has topic Multimedia information retrieval +
Month November  +
Peer reviewed Yes  +
Publication type Journal article  +
Published in Multimedia Tools and Applications +
Research design Experiment  +
Research questions In this work we are concerned with the proIn this work we are concerned with the problem of accurately finding the location where a photo is taken without needing any metadata, that is, solely by its visual content. We also recognize landmarks where applicable, automatically linking them to Wikipediae, automatically linking them to Wikipedia
Revid 11,030  +
Theories Undetermined
Theory type Design and action  +
Title Viral: visual image retrieval and localization
Unit of analysis Article  +
Url http://dx.doi.org/10.1007/s11042-010-0651-7  +
Wikipedia coverage Other  +
Wikipedia data extraction Secondary dataset  +
Wikipedia language Not specified  +
Wikipedia page type Article  +
Year 2010  +
Creation dateThis property is a special property in this wiki. 15 March 2012 20:32:38  +
Categories Multimedia information retrieval  + , Geography  + , Information science  + , Publications  +
Modification dateThis property is a special property in this wiki. 30 January 2014 20:32:15  +
hide properties that link here 
  No properties link to this page.
 

 

Enter the name of the page to start browsing from.