Visual-Textual Joint Relevance Learning for Tag-Based Social Image Search

@article{Gao2013VisualTextualJR,
  title={Visual-Textual Joint Relevance Learning for Tag-Based Social Image Search},
  author={Yue Gao and Meng Wang and Zhengjun Zha and Jialie Shen and Xuelong Li and Xindong Wu},
  journal={IEEE Transactions on Image Processing},
  year={2013},
  volume={22},
  pages={363-376}
}
Due to the popularity of social media websites, extensive research efforts have been dedicated to tag-based social image search. Both visual information and tags have been investigated in the research field. However, most existing methods use tags and visual characteristics either separately or sequentially in order to estimate the relevance of images. In this paper, we propose an approach that simultaneously utilizes both visual and textual information to estimate the relevance of user tagged… 
Joint Hypergraph Learning for Tag-Based Image Retrieval
TLDR
This paper proposes a global and local visual features fusion approach to learn the relevance of images by hypergraph approach and adopts the hypergraph learning algorithm to calculate the relevance score of each image to the query.
Tag-based social image search with hyperedges correlation
TLDR
A hypergraph with correlated hyperedge (CHH) is proposed, which introduces high-order relationship of hyperedges into hypergraph learning and a pairwise visual-textual correlation hypergraph (VTCH) model is used for tag-based social image search.
Multimodal information joint learning for geotagged image search
  • Yi Xie, Huimin Yu, Roland Hu
  • Computer Science
    2014 IEEE International Conference on Multimedia and Expo Workshops (ICMEW)
  • 2014
TLDR
A hypergraph-based framework which integrates image content, user-generated tags and geo-location information into image ranking problem and simultaneously optimizes the ranking scores and hyperedge weights is proposed.
Social image search exploiting joint visual-textual information within a fuzzy hypergraph framework
TLDR
A novel and efficient method is proposed, exploiting visual and textual information simultaneously in image search, powered by a fuzzy hypergraph powered by the term-frequency and inverse-document-frequency (tf-idf) weighting scheme.
Exploiting visual and textual neighborhood information to improve image-tag relevance
TLDR
This work improves image-tag relevance by combining visual neighborhood of images and textual neighborhood of tags and boosts the ranking of informative tags of an image by exploiting tags of visual neighbors of the images.
An Effective and Efficient Re-ranking Framework for Social Image Search
TLDR
This paper proposes an effective and efficient re-ranking framework for social image search, which can quickly and accurately return ranking results, and develops an index to accelerate the optimization process, which ensures the efficiency of the framework.
Optimized tag ranking based on visual vocabulary for social images in compressed domain
TLDR
Experimental results show that the proposed optimization method can significantly reduce the time of image tag ranking under ensuring the ranking accuracy of social image tags.
Scalable Hypergraph-Based Image Retrieval and Tagging System
TLDR
Based on the hypergraph, HIRT is proposed, a scalable image retrieval and tagging system, which uses Personalized PageRank to measure vertex similarity, and employs top-k search to support image retrieved and tagging.
Socializing the Semantic Gap
TLDR
A two-dimensional taxonomy is introduced to structure the growing literature, understand the ingredients of the main works, clarify their connections and difference, and recognize their merits and limitations.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 72 REFERENCES
Tag-based social image search with visual-text joint hypergraph learning
TLDR
This paper simultaneously investigates the bag-of-words and bag- of-visual-words representations of images and accomplish the relevance estimation with a hypergraph learning approach and results demonstrate the effectiveness of this approach.
Towards a Relevant and Diverse Search of Social Images
TLDR
A diverse relevance ranking scheme that is able to take relevance and diversity into account by exploring the content of images and their associated tags, and it is shown that the diversity of search results can be enhanced while maintaining a comparable level of relevance.
Social Image Search with Diverse Relevance Ranking
TLDR
A diverse relevance ranking scheme which is able to simultaneously take relevance and diversity into account and takes advantage of both the content of images and their associated tags.
Unsupervised multi-feature tag relevance learning for social image retrieval
TLDR
Two simple and effective unsupervised methods based on the classical Borda Count, comparable to a state-of-the-art supervised alternative, but without the need of any training data are introduced.
Learning tag relevance by neighbor voting for social image retrieval
TLDR
A novel algorithm that scalably and reliably learns tag relevance by accumulating votes from visually similar neighbors is proposed, and treated as tag frequency, learned tag relevance is seamlessly embedded into current tag-based social image retrieval paradigms.
Boost search relevance for tag-based social image retrieval
TLDR
This paper proposes a relevance-based ranking scheme for social image search, aiming to automatically rank images according to their relevance to the query tag, and shows that an iterative method to solve the optimization problem can thus be accomplished.
Learning Social Tag Relevance by Neighbor Voting
TLDR
This paper proposes a neighbor voting algorithm which accurately and efficiently learns tag relevance by accumulating votes from visual neighbors and proves that the algorithm is a good tag relevance measurement for both image ranking and tag ranking.
Image Retagging Using Collaborative Tag Propagation
TLDR
The proposed image retagging scheme is formulated as a multiple graph-based multi-label learning problem, which simultaneously explores the visual content of the images, semantic correlation of the tags as well as the prior information provided by users.
Image tag clarity: in search of visual-representative tags for social images
TLDR
Evaluated on a large real-world dataset containing more than 269K images and their associated tags, it is shown that NITC score can effectively identify the visual-representative tags from all tags contributed by users.
Visual diversification of image search results
TLDR
Based on a performance evaluation, the outcome of the methods for visual diversification of image search results closely resembles human perception of diversity, which was established in an extensive clustering experiment carried out by human assessors.
...
1
2
3
4
5
...