ABSTRACT
Most existing tag-based social image search engines present search results as a ranked list of images, which cannot be consumed by users in a natural and intuitive manner. In this paper, we present a novel concept-preserving image search results summarization algorithm named Prism. Prism exploits both visual features and tags of the search results to generate high quality summary, which not only breaks the results into visually and semantically coherent clusters but it also maximizes the coverage of the summary w.r.t the original search results. It first constructs a visual similarity graph where the nodes are images in the search results and the edges represent visual similarities between pairs of images. This graph is optimally decomposed and compressed into a set of concept-preserving subgraphs based on a set of summarization objectives. Images in a concept-preserving subgraph are visually and semantically cohesive and are described by a minimal set of tags or concepts. Lastly, one or more exemplar images from each subgraph is selected to form the exemplar summary of the result set. Through empirical study, we demonstrate the effectiveness of Prism against state-of-the-art image summarization and clustering algorithms.
- M. B. Blaschko and C. H. Lampert. Correlational spectral clustering. In IEEE CVPR, 2008.Google ScholarCross Ref
- D.\ Cai, et al. Hierarchical clustering of WWW image search results using visual, textual and link information. In ACM MM, 2004. Google ScholarDigital Library
- H.\ Cheng, et al. Clustering large attributed information networks: an efficient incremental computing approach. Data Mining and Knowledge Discovery, 25(3), March 2012.Google Scholar
- T-S.\ Chua, et al. NUS-WIDE: a real-world web image database from National University of Singapore. In ACM CIVR, 2009. Google ScholarDigital Library
- V. Chvatal. A Greedy Heuristic for the Set-Covering Problem. Mathematics of Operations Research, 4(3):233--235, 1979.Google ScholarDigital Library
- B. J Frey and D. Dueck. Clustering by passing messages between data points. Science, 315(5814):972--976, 2007.Google ScholarCross Ref
- Y.\ Jia, et al. Finding image exemplars using fast sparse affinity propagation. In ACM MM, 2008. Google ScholarDigital Library
- Y.\ Jing and S.\ Baluja. VisualRank: applying PageRank to large-scale image search. IEEE PAMI, 30(11):1877--90, November 2008. Google ScholarDigital Library
- Z.\ Li, et al. Grouping WWW Image Search Results by Novel Inhomogeneous Clustering Method. In Proc. IEEE International Multimedia Modelling Conference, 2005. Google ScholarDigital Library
- P.-A.\ Mo\"ellic, et al. Image clustering based on a shared nearest neighbors approach for tagged collections. In ACM CIVR, 2008. Google ScholarDigital Library
- T.\ Opsahl and P.\ Panzarasa. Clustering in weighted networks. Social Networks, 31(2):155--163, 2009.Google ScholarCross Ref
- R.\ Raguram and S.\ Lazebnik. Computing iconic summaries of general visual concepts. In CVPR Workshops, 2008.Google Scholar
- M.\ Rege, M.\ Dong, and J.\ Hua. Graph theoretical framework for simultaneously integrating visual and textual features for efficient web image clustering. In ACM WWW, 2008. Google ScholarDigital Library
- B. S.\ Seah, et al. Summarizing Social Image Search Results. In ACM WWW, 2014. Google ScholarDigital Library
- I.\ Simon, et al. Scene Summarization for Online Image Collections. In IEEE ICCV, 2007.Google Scholar
- .\ Sun, S. S.\ Bhowmick, et al. Tag-based social image retrieval: An empirical evaluation,textitJASIST, 62(12), 2011. Google ScholarDigital Library
- B. Q.\ Truong, et al. CASIS: a system for concept-aware social image search. In ACM WWW, 2012. Google ScholarDigital Library
- R. H. van Leuken, et al. Visual diversification of image search results. In ACM WWW, 2009. Google ScholarDigital Library
- S.\ Wang, et al. IGroup: presenting web image search results in semantic clusters. In ACM CHI, 2007. Google ScholarDigital Library
- H.\ Xu, et al. Hybrid image summarization. In ACM MM, 2011. Google ScholarDigital Library
Index Terms
- PRISM: concept-preserving social image search results summarization
Recommendations
Summarizing social image search results
WWW '14 Companion: Proceedings of the 23rd International Conference on World Wide WebMost existing social image search engines present search results as a ranked list of images, which cannot be consumed by users in a natural and intuitive manner. Here, we present a novel algorithm that exploits both visual features and tags of the ...
Boost search relevance for tag-based social image retrieval
ICME'09: Proceedings of the 2009 IEEE international conference on Multimedia and ExpoSocial media sharing web sites like Flickr allow users to annotate images with free tags, which greatly facilitate social image search and browsing. However, currently tag-based image search on Flickr does not provide the option of relevance-based ...
Why not, WINE?
WWW '14 Companion: Proceedings of the 23rd International Conference on World Wide WebDespite considerable progress in recent years on Tag-based Social Image Retrieval (TagIR), state-of-the-art TagIR systems fail to provide a systematic framework for end users to ask why certain images are not in the result set of a given query and ...
Comments