Using Geotags to Derive Rich Tag-Clouds for Image Annotation

Joshi, Dhiraj; Luo, Jiebo; Yu, Jie; Lei, Phoury; Gallagher, Andrew

doi:10.1007/978-0-85729-436-4_11

Cited by 14 publications

(7 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Naaman [3] first proposed using a database of named geographic locations to automatically label geo-tagged photos, although their capabilities were limited to cities, states, and parks, and they did not support search. More recent work has looked at combining personal tags with community tags, using GPStagged photos on flickr to find nearby tags [4], [5]. Many authors have used date and time information to organize photo collections and group them into events, e.g., [6].…”

Section: Related Workmentioning

confidence: 99%

Photo Recall: Using the Internet to Label Your Photos

Kumar

Seitz

2014

2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops

View full text Add to dashboard Cite

We describe a system for searching your personal photos using an extremely wide range of text queries, including dates and holidays (Halloween), named and categorical places (Empire State Building or park ), events and occasions (Radiohead concert or wedding), activities (skiing), object categories (whales), attributes (outdoors), and object instances (Mona Lisa), and any combination of these -all with no manual labeling required. We accomplish this by correlating information in your photos -the timestamps, GPS locations, and image pixels -to information on the Internet. This includes matching dates to holidays listed on Wikipedia, GPS coordinates to places listed on Wikimapia, places and dates to find named events using Google, and visual categories and object instances using classifiers either pre-trained on ImageNet or trained on-the-fly using results from Google Image Search. We tie all of these disparate sources of information together in a unified way, allowing for fast and accurate searches using whatever information you remember about a photo. We quantitatively evaluate several aspects of our system and show excellent performance in all respects. Please watch a video demonstrating our system in action on a large range of queries at http://youtu.be/Se3bemzhAiY

show abstract

Section: Related Workmentioning

confidence: 99%

Photo Recall: Using the Internet to Label Your Photos

Kumar

Seitz

2014

2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops

View full text Add to dashboard Cite

show abstract

“…A location-driven tag suggestion system is proposed in [7]. This system use sources such as a public Geographic Names Information System (GNIS) database, community tags from Flickr pictures, and personal tags shared through users' photo collections.…”

Section: Related Workmentioning

confidence: 99%

TagPix: Automatic Real-Time Landscape Photo Tagging for Smartphones

Debnath¹,

Borcea²

2013

2013 International Conference on MOBILe Wireless MiddleWARE, Operating Systems, and Applications

View full text Add to dashboard Cite

Abstract-This paper presents TagPix, a lightweight smartphone photo tagging app that provides good tagging accuracy, works in real-time, and protects user privacy. The main novelty of TagPix consists in leveraging the phone sensors and a placetag database to tag landscape photos which include landmarks. GPS location is used to identify landmark tags in a given region. Then, TagPix computes the angular distance for the object in the camera focus using the orientation sensors. This allows the app to select a small subset of landmark tags for the photo. For further tag accuracy, we devised new usable methods to estimate the Euclidean distance between the user and the landmark in the camera focus. These methods employ simple user actions and lightweight trigonometric calculations. TagPix is implemented and tested using several Android phones and Google Places API. The app was tested in 8 cities across USA. Using only angular distance, TagPix achieves 86% tagging accuracy. Adding Euclidean distance estimation leads to 93% accuracy.

show abstract

“…The work of Jones [70] and Popescu and Moellic [119] describe annotation of landmark images, where identification of landmark and subsequent annotation relies on a reference image corpus collected from Flickr [119] and from Flickr and Panoramio [70]. In the work of Joshi [71] location-driven tags are suggested for geo-tagged images by collecting tag clouds from three different sources; a point-of-interest database, Flickr photos and personal photos. Tags are ranked according to distance and visual similarity to the query image.…”

Section: Image Annotation In Online Social Networkmentioning

confidence: 99%

“…Much of the referred works that provide image tagging are based on content analysis of images [70,71,103,119,133,137], in some cases in combination with geographic position or user provided keyword(s) [55,70,71,88,103,119,133,137,148]. Our work contrasts with the referred works as we do not include analysis of visual features in images, but rather provide automatic image annotation based solely on image metadata.…”

Section: Related Workmentioning

confidence: 99%

“…A number of systems annotate query images by selecting tags from related images gathered from online image collections based on a combination of geographic position and visual similarity (Jones et al, 2010;Joshi et al, 2011;Sergieh et al, 2012;Popescu and Moëllic, 2009;Silva and Martins, 2011). The general technique is first to collect a set of images within a certain radius of the query image, narrow down the set by using visual similarity techniques, and finally collect tags from the remaining images.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A Context Centric Approach for Semantic Image Annotation and Retrieval

Elahi

Karlsen

Akselsen

2009

2009 Computation World: Future Computing, Service Computation, Cognitive, Adaptive, Content, Patterns

View full text Add to dashboard Cite

I would also like to thank master student Einar, Martin and Steffen Hageland for their participation in project development and other research fellows Mehmmod and Kostas Antypas for their valuable discussions and giving me this believe that I can pursue and fulfill my research goals. Last but not the least, I would like to thank my family: my wife Sana, my mother Riffat Ara, my father and my siblings for supporting me spiritually throughout writing this thesis and my life in general.

show abstract

Using Geotags to Derive Rich Tag-Clouds for Image Annotation

Cited by 14 publications

References 6 publications

Photo Recall: Using the Internet to Label Your Photos

Photo Recall: Using the Internet to Label Your Photos

TagPix: Automatic Real-Time Landscape Photo Tagging for Smartphones

A Context Centric Approach for Semantic Image Annotation and Retrieval

Contact Info

Product

Resources

About