Automatic image annotation using visual content and folksonomies

Lindstaedt, Stefanie; Mörzinger, Roland; Sorschag, Robert; Pammer, Viktoria; Thallinger, Georg

doi:10.1007/s11042-008-0247-7

Cited by 56 publications

(25 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Two systems that can potentially handle unlimited vocabulary are presented in [14] and [15]. Li et al present a content based image retrieval system [14] that propagates textual features from visually similar images.…”

Section: Related Workmentioning

confidence: 99%

“…Li et al present a content based image retrieval system [14] that propagates textual features from visually similar images. Lindstaedt et al [15] use a combination of off-line supervised classification of images into selected concepts followed by tag propagation from visual similar images. However, these systems need to analyze very large datasets and do not address the localized relevance of tags in a folksonomic setting.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Quest for relevant tags using local interaction networks and visual content

Sawant

Datta

et al. 2010

Proceedings of the International Conference on Multimedia Information Retrieval

View full text Add to dashboard Cite

Typical tag recommendation systems for photos shared on social networks such as Flickr, use visual content analysis, collaborative filtering or personalization strategies to produce annotations. However, the dependence on manual intervention and the knowledge of sufficient personal preferences coupled with the folksonomic issues limit the scope of these strategies. In this paper, we present a fully automatic and folksonomically scalable tag recommendation model that can recommend tags for a user's photos without an explicit knowledge of the user's personal tagging preferences. The model is learned using the collective tagging behavior of other users in the user's local interaction network, which we believe approximates the user's preferences, at least partially. The tag recommendation model generates content-based annotations and then uses a Naïve Bayes formulation to translate these annotations to a set of folksonomic tags selected from the tags used by the users in the local interaction network. Quantitative and qualitative comparisons with 890 Flickr networks show that this approach is highly useful for tag recommendation in the presence of insufficient information of a user's own preferences.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Quest for relevant tags using local interaction networks and visual content

Sawant

Datta

et al. 2010

Proceedings of the International Conference on Multimedia Information Retrieval

View full text Add to dashboard Cite

show abstract

“…For instance, Linstaedt et al use sophisticated computer programs to analyze still images found within Flickr and then use this analysis to process new images and to propagate relevant user tags to those images. 21 In a slightly more complicated example, Liu and Qin employ machine--learning techniques to initially process and assign metadata, including subject terms, to a repository of documents related to the computer science profession. 22 However, this proof of concept project also permits users to edit the fields of the metadata once established.…”

Section: Social Taggingmentioning

confidence: 99%

Evaluation of Semi-Automatic Metadata Generation Tools: A Survey of the Current State of the Art

Park

Brenza

2015

ITAL

View full text Add to dashboard Cite

show abstract

“…Moreover, tags allow reusing already existing text-based search techniques. However, as for instance pointed out in [5] and [6], the presence of non-relevant tags hampers the effective organization and retrieval of usercontributed images, motivating the design of techniques that allow differentiating relevant tags from non-relevant tags. In this paper, we consider a tag to be non-relevant when users with common knowledge are not able to easily and consistently relate the tag to the image content, a definition also used by the authors of [7], [8], and [9].…”

Section: Introductionmentioning

confidence: 99%

Visually weighted neighbor voting for image tag relevance learning

Lee

Neve

2013

Multimed Tools Appl

View full text Add to dashboard Cite

The presence of non-relevant tags in image folksonomies hampers the effective organization and retrieval of user-contributed images. In this paper, we propose to learn the relevance of user-supplied tags by means of visually weighted neighbor voting, a variant of the popular baseline neighbor voting algorithm proposed by Xirong Li et al. in 2009. To gain insight into the effectiveness of baseline and visually weighted neighbor voting, we qualitatively analyze the difference in tag relevance when using a different number of neighbors, for both tags relevant and tags not relevant to the content of a seed image. Our qualitative analysis shows that tag relevance values computed by means of visually weighted neighbor voting are more stable and representative than tag relevance values computed by means of baseline neighbor voting. This is quantitatively confirmed through extensive experimentation with MIRFLICKR-25000, studying the variation of tag relevance values as a function of the number of neighbors used (for both tags relevant and tags not relevant with respect to the content of a seed image), as well as the influence of tag relevance learning on the effectiveness of image tag refinement, tag-based image retrieval, and image tag recommendation.

show abstract

Automatic image annotation using visual content and folksonomies

Cited by 56 publications

References 18 publications

Quest for relevant tags using local interaction networks and visual content

Quest for relevant tags using local interaction networks and visual content

Evaluation of Semi-Automatic Metadata Generation Tools: A Survey of the Current State of the Art

Visually weighted neighbor voting for image tag relevance learning

Contact Info

Product

Resources

About