Where your photo is taken: Geolocation prediction for social images

Abstract

Social image-sharing websites have attracted a large number of users. These systems allow users to associate geolocation information with their images, which is essential for many interesting applications. However, only a small fraction of social images have geolocation information. Thus, an automated tool for suggesting geolocation is essential to help users geotag their images. In this article, we use a large data set consisting of 221 million Flickr images uploaded by 2.2 million users. For the first time, we analyze user uploading patterns, user geotagging behaviors, and the relationship between the taken-time gap and the geographical distance between two images from the same user. Based on the findings, we represent a user profile by historical tags for the user and build a multinomial model on the user profile for geotagging. We further propose a unified framework to suggest geolocations for images, which combines the information from both image tags and the user profile. Experimental results show that for images uploaded by users who have never done geotagging, our method outperforms the state-of-the-art method by 10.6 to 34.2%, depending on the granularity of the prediction. For images from users who have done geotagging, a simple method is able to achieve very high accuracy.

Analysis of Social Networking Service Data for Smart Urban Planning

“…Other proposals are focused on discovering popular tourist attractions within the urban areas through geo-tagged images posted in Flickr [47] or Panoramio [48]. These platforms are SNSs where users can share, organize, and find multimedia content composed mainly of pictures [49,50].…”

“We Can Remember It for You”: Location, Memory, and Commodification in Social Networking Sites

“…This is what makes geotagged images and status updates valuable to commercial companies for advertising, user profiling, and other uses such as studying consumption patterns on a particular street. Because "a large portion of photos uploaded to social image-sharing services contain no geolocation information" (Liu et al, 2014), companies have created automated systems that detect the location shown in the photograph to prompt users to geotag both new and older content. A study of images posted to Flickr found that less than 50% were geotagged (Liu et al, 2014).…”

“…Because "a large portion of photos uploaded to social image-sharing services contain no geolocation information" (Liu et al, 2014), companies have created automated systems that detect the location shown in the photograph to prompt users to geotag both new and older content. A study of images posted to Flickr found that less than 50% were geotagged (Liu et al, 2014). Interestingly, this study used the time lapse between capturing the image and geotagging it as one of the methods of analyzing user geotagging behavior (Liu et al, 2014).…”

“…A study of images posted to Flickr found that less than 50% were geotagged (Liu et al, 2014). Interestingly, this study used the time lapse between capturing the image and geotagging it as one of the methods of analyzing user geotagging behavior (Liu et al, 2014). Facebook is increasingly becoming a location-sharing platform, promoting features such as opt-in or opt-out of sharing location (Wilken, 2014) and allowing geotagging of both new and old content (Wilken, 2014(Wilken, , p. 1093.…”

A Survey of Geo-tagged Multimedia Content Analysis within Flickr

“…Kennedy and Naaman [39] used visual features and tags, in order to extract the most representative tags and views for landmarks, working on a corpus of 110K Flickr photos from San Fransisco. Finally, Liu et al [40] were the first to consider user uploading patterns, geotagging behaviors, and the relationship between the temporal and the spatial gap of two photos from the same user.…”