Visual and semantic context modeling for scene-centric image annotation

Abstract

Automatic image annotation enables efficient indexing and retrieval of the images in the large-scale image collections, where manual image labeling is an expensive and labor intensive task. This paper proposes a novel approach to automatically annotate images by coherent semantic concepts learned from image contents. It exploits sub-visual distributions from each visually complex semantic class, disambiguates visual descriptors in a visual context space, and assigns image annotations by modeling image semantic context. The sub-visual distributions are discovered through a clustering algorithm, and probabilistically associated with semantic classes using mixture models. The clustering algorithm can handle the inner-category visual diversity of the semantic concepts with the curse of dimensionality of the image descriptors. Hence, mixture models that formulate the sub-visual distributions assign relevant semantic classes to local descriptors. To capture non-ambiguous and visual-consistent local descriptors, the visual context is learned by a probabilistic Latent Semantic Analysis (pLSA) model that ties up images and their visual contents. In order to maximize the annotation consistency for each image, another context model characterizes the contextual relationships between semantic concepts using a concept graph. Therefore, image labels are finally specialized for each image in a scene-centric view, where images are considered as unified entities. In this way, highly consistent annotations are probabilistically assigned to images, which are closely correlated with the visual contents and true semantics of the images. Experimental validation on several datasets shows that this method outperforms state-of-the-art annotation algorithms, while effectively captures consistent labels for each image.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

References

  1. 1.

    Amiri SH, Jamzad M (2015) Automatic image annotation using semi-supervised generative modeling. Pattern Recognit 48(1):174–188

    Article  Google Scholar 

  2. 2.

    Ankerst M, Breunig MM, Kriegel H et al. (1999) “OPTICS: ordering points to identify the clustering structure,”. Proc SIGMOD’99 ACM SIGMOD Int Conf Manag Data 49–60

  3. 3.

    Assari SM, Zamir AR, Shah M et al. (2014) “Video classification using semantic concept co-occurrences,”. Proc 2014 I.E. Conf Comput Vision Pattern Recognit (CVPR) 2529–2536

  4. 4.

    Bannour H, Hudelot C (2014) Building and using fuzzy multimedia ontologies for semantic image annotation. Multimed Tools Appl 72(3):2107–2141

    Article  Google Scholar 

  5. 5.

    Bao B-K, Li T, Yan S (2012) Hidden-concept driven multilabel image annotation and label ranking. IEEE Trans Multimed 14(1):199–210

    Article  Google Scholar 

  6. 6.

    Barnard K, Duygulu P, Forsyth D, de Freitas N, Blei DM, Jordan MI (2003) Matching words and pictures. J Mach Learn Res 3:1107–1135

    MATH  Google Scholar 

  7. 7.

    Bouveyron C, Girard S, Schmid C (2007) High-dimensional data clustering. Comput Stat Data Anal 52(1):502–519

    MathSciNet  Article  MATH  Google Scholar 

  8. 8.

    Campello R, Moulavi D, Sander J et al. (2013) “Density-based clustering based on hierarchical density estimates,”. Proc 17th Pacific-Asia Conf Adv Knowledge Discovery Data Mining (PAKDD) Springer, LNAI 160–172

  9. 9.

    Carneiro G, Chan AB, Moreno PJ, Vasconcelos N (2007) Supervised learning of semantic classes for image annotation and retrieval. Pattern Anal Mach Intell IEEE Trans 29(3):394–410

    Article  Google Scholar 

  10. 10.

    Chapelle O, Haffner P, Vapnik VN (1999) Support vector machines for histogram-based image classification. Neural Networks, IEEE Trans 10(5):1055–1064

    Article  Google Scholar 

  11. 11.

    Chen Z, Fu H, Chi Z, Feng D (2012) An adaptive recognition model for image annotation. Syst Man, Cybern Part C Appl Rev IEEE Trans 42(6):1120–1127

    Article  Google Scholar 

  12. 12.

    Chua T, Tang J, Hong R et al. (2009) “NUS-WIDE: a real-world web image database from national university of Singapore,”. Proc ACM Int Conf Image Video Retriev, Greece 48–53

  13. 13.

    Csurka G, Dance CR, Fan L et al. (2004) “Visual categorization with bags of keypoints,”. Proc Workshop Statistical Learn Comput vision (ECCV 2004) 1–22

  14. 14.

    Donoho DL, Grimes C (2003) Hessian eigenmaps: locally linear embedding techniques for high-dimensional data. Proc Natl Acad Sci 100(10):5591–5596

    MathSciNet  Article  MATH  Google Scholar 

  15. 15.

    Escalante HJ, Hernández CA, Gonzalez JA, López-López A, Montes M, Morales EF, Enrique Sucar L, Villaseñor L, Grubinger M (2010) The segmented and annotated IAPR TC-12 benchmark. Comput Vis Image Underst 114(4):419–428

    Article  Google Scholar 

  16. 16.

    Ester M, Kriegel H, Sander J et al. (1996) “A density-based algorithm for discovering clusters in large spatial databases with noise,”. Proc Second Int Conf Knowledge Discovery Data Mining (KDD-96)

  17. 17.

    Fernando B, Fromont E, Tuytelaars T et al. (2012) “Effective use of frequent itemset mining for image classification,”. Proc Europe Conf Comput Vision (ECCV) 214–227

  18. 18.

    Gao Y, Ji R, Liu W, Dai Q, Hua G (2014) Weakly supervised visual dictionary learning by harnessing image attributes. Imag Process IEEE Trans 23(12):5400–5411

    MathSciNet  Article  Google Scholar 

  19. 19.

    Gao S, Tsang IW, Chia L (2013) Laplacian sparse coding, hypergraph laplacian sparse coding, and applications. Pattern Anal Mach Intell IEEE Trans 35(1):92–104

    Article  Google Scholar 

  20. 20.

    Gould S, Rodgers J, Cohen D, Elidan G, Koller D (2008) Multi-class segmentation with relative location prior. Int J Comput Vis 80(3):300–316

    Article  Google Scholar 

  21. 21.

    Guha S, Rastogi R, Shim K et al. (1998) “CURE: an efficient clustering algorithm for large databases,”. Proc SIGMOD’98 ACM SIGMOD Int Conf Manag Data 73–84

  22. 22.

    He K, Chan W, Zhu G, Lin L, Zhou X (2014) Image region labeling by exploring contextual information of visual spatial and semantic concepts. Proc 15th Pacific Rim Conf Adv Multimed Inform Process–PCM 1:93–102

    Google Scholar 

  23. 23.

    He X, Zemel RS, Carreira-Perpinan MA (2004) “Multiscale conditional random fields for image labeling,”. Proc 2004 I.E. Comput Soc Conf Comput Vision Pattern Recognit (CVPR)

  24. 24.

    Huiskes M, Thomee B, Lew M et al. (2010) “New trends and ideas in visual concept detection: the MIR flickr retrieval evaluation initiative,”. Proc 11th ACM Int Conf Multimed Inform Retriev 527–536

  25. 25.

    Izadinia H, Shah M (2012) Recognizing complex events using large margin joint low-level event model. Proc Europ Conf Comput Vision (ECCV) 7575:430–444

    Google Scholar 

  26. 26.

    Jiang W, Chang SF, Loui AC (2007) Context-based concept fusion with boosted conditional random fields. Proc 2007 I.E. Int Conf Acoustics, Speech, Sign Process I(Pts 1-3):949–952

    Google Scholar 

  27. 27.

    Ladick’ L, Russell C, Kohli P et al (2009) “Associative hierarchical crfs for object class image segmentation,”. Proc 12th IEEE Int Conf Comput Vision (ICCV) 739–746

  28. 28.

    Lazebnik S, Raginsky M (2009) Supervised learning of quantizer codebooks by information loss minimization. Pattern Anal Mach Intell IEEE Trans 31(7):1294–1309

    Article  Google Scholar 

  29. 29.

    Lazebnik S, Schmid C, Ponce J (2006) Beyond bags of features: spatial pyramid matching for recognizing natural scene categories. Proc 2006 I.E. Comput Soc Conf Comput Vision Pattern Recognit (CVPR) 2:2169–2178

    Article  Google Scholar 

  30. 30.

    Leung CHC, Chan AWS, Milani A, Liu J, Li Y (2012) Intelligent social media indexing and sharing using an adaptive indexing search engine. ACM Trans Intell Syst Technol 3(3):1–27

    Article  Google Scholar 

  31. 31.

    Li L, Jiang S, Huang Q (2012) Learning hierarchical semantic description via mixed-norm regularization for image understanding. IEEE Trans Multimed 14(5):1401–1413

    Article  Google Scholar 

  32. 32.

    Li Z, Liu J, Tang J, Lu H (2014) Projective matrix factorization with unified embedding for social image tagging. Comput Vis Image Underst 124:71–78

    Article  Google Scholar 

  33. 33.

    Li J, Wang JZ (2008) Real-time computerized annotation of pictures. IEEE Trans Pattern Anal Mach Intell 30(6):985–1002

    Article  Google Scholar 

  34. 34.

    Liu W, Tao D (2013) Multiview hessian regularization for image annotation. Imag Process IEEE Trans 22(7):2676–2687

    MathSciNet  Article  Google Scholar 

  35. 35.

    Liu W, Tao D, Cheng J, Tang Y (2014) Multiview hessian discriminative sparse coding for image annotation. Comput Vis Image Underst 118:50–60

    Article  Google Scholar 

  36. 36.

    Llorente A, Manmatha R, Rüger S et al. (2010) “Image retrieval using Markov random fields and global image features,”. Proc ACM Int Conf Imag Video Retriev 243

  37. 37.

    Ma Z, Nie F, Yang Y, Uijlings JRR, Sebe N (2012) Web image annotation via subspace-sparsity collaborated feature selection. IEEE Trans Multimed 14(4):1021–1030

    Article  Google Scholar 

  38. 38.

    Makadia A, Pavlovic V, Kumar S (2010) Baselines for image annotation. Int J Comput Vis 90(1):88–105

    Article  Google Scholar 

  39. 39.

    Nguyen C-T, Kaothanthong N, Tokuyama T, Phan X-H (2013) A feature-word-topic model for image annotation and retrieval. ACM Trans Web 7(3):1–24

    Article  Google Scholar 

  40. 40.

    Parsons L, Haque E, Liu H (2004) Subspace clustering for high dimensional data: a review. ACM SIGKDD Explor Newsl 6(1):90–105

    Article  Google Scholar 

  41. 41.

    Rabinovich A, Vedaldi A, Galleguillos C et al. (2007) “Objects in context,”. Proc 11th IEEE Int Conf Comput Vision (ICCV) 1–8

  42. 42.

    Rasiwasia N, Vasconcelos N (2009) “Holistic context modeling using semantic co-occurrences,”. Proc 2009 I.E. Conf Comput Vision Pattern Recognit (CVPR) 1889–1895

  43. 43.

    Ritendra D, Joshi D, Li J, Wang JZ (2008) Image retrieval: ideas, influences, and trends of the new age. ACM Comput Surv 40(2):1–60

    Google Scholar 

  44. 44.

    Russell BC, Efros AA, Sivic J, Freeman WT, Zisserman A (2006) Using multiple segmentations to discover objects and their extent in image collections. Proc 2006 I.E. Comput Soc Conf Comput Vision Pattern Recognit (CVPR) 2:1605–1614

    Article  Google Scholar 

  45. 45.

    Schultz M, Joachims T (2004) “Learning a distance metric from relative comparisons,”. Adv Neural Inf Process Syst 41–48

  46. 46.

    Shotton J, Winn J, Rother C et al. (2006) “Textonboost: joint appearance, shape and context modeling for multi-class object recognition and segmentation,”. Proc Europ Conf Comput Vision (ECCV), Springer Berlin Heidelberg 1–15

  47. 47.

    Sivic J, Zisserman A (2003) Video Google: a text retrieval approach to object matching in videos. Proc 9th IEEE Int Conf Comput Vision (ICCV) 2:1470–1477

    Article  Google Scholar 

  48. 48.

    Song X, Jiang S, Wang S, Li L, Huang Q (2014) Polysemious visual representation based on feature aggregation for large scale image applications. Multimed Tools Appl 74(2):595–611

    Article  Google Scholar 

  49. 49.

    Su J-H, Huang W-J, Yu PS, Tseng VS (2011) Efficient relevance feedback for content-based image retrieval by mining user navigation patterns. Knowl Data Eng IEEE Trans 23(3):360–372

    Article  Google Scholar 

  50. 50.

    Tao D, Jin L, Liu W, Li X (2013) Hessian regularized support vector machines for mobile image annotation on the cloud. Multimed, IEEE Trans 15(4):833–844

    Article  Google Scholar 

  51. 51.

    Tirilly P, Claveau V, Gros P et al. (2008) “Language modeling for bag-of-visual word image categorization,”. Proc CIVR’08 - Int Conf Content-Based Imag Video Retriev 249–258

  52. 52.

    Vailaya A, Figueiredo MAT, Jain AK, Zhang H-J (2001) Image classification for content-based indexing. Imag Process IEEE Trans 10(1):117–130

    Article  MATH  Google Scholar 

  53. 53.

    Van Gemert JC, Snoek CGM, Veenman CJ, Smeulders AWM, Geusebroek J (2010) Comparing compact codebooks for visual categorization. Comput Vis Image Underst 114(4):450–462

    Article  Google Scholar 

  54. 54.

    Vogel J, Schiele B (2007) Semantic modeling of natural scenes for content-based image retrieval. Int J Comput Vis 72(2):133–157

    Article  Google Scholar 

  55. 55.

    Wang X, Du J, Wu S, Li X, Xin H, Zhang Y, Li F (2015) High-level semantic image annotation based on hot Internet topics. Multimed Tools Appl 74(6):2055–2084

    Article  Google Scholar 

  56. 56.

    Wang H, Lu T, Wang Y et al. (2014) “Weakly-supervised region annotation for understanding scene images,”. Multimed Tools Appl 1–25

  57. 57.

    Wang Y, Mei T, Gong S, Hua X-S (2009) Combining global, regional and contextual features for automatic image annotation. Pattern Recognit 42(2):259–266

    Article  MATH  Google Scholar 

  58. 58.

    Xiang Y, Zhou X, Liu Z et al. (2010) “Semantic context modeling with maximal margin Conditional Random Fields for automatic image annotation,”. Proc 2010 I.E. Conf Comput Vision Pattern Recognit (CVPR) 3368–3375

  59. 59.

    Xu C, Tao D, Xu C (2014) Large-margin multi-view information bottleneck. Pattern Anal Mach Intell IEEE Trans 36(8):1559–1572

    Article  Google Scholar 

  60. 60.

    Zha Z.-J, Hua X.-S, Mei T et al. (2008) “Joint multi-label multi-instance learning for image classification,”. Proc 2008 I.E. Conf Comput Vision Pattern Recognit (CVPR) 1–8

  61. 61.

    Zhang D, Islam MM, Lu G (2012) A review on automatic image annotation techniques. Pattern Recognit 45(1):346–362

    Article  Google Scholar 

  62. 62.

    Zhang D, Monirul Islam M, Lu G (2013) Structural image retrieval using automatic image annotation and region based inverted file. J Vis Commun Image Represent 24(7):1087–1098

    Article  Google Scholar 

  63. 63.

    Zhang S, Tian Q, Hua G, Huang Q, Gao W (2014) ObjectPatchNet: towards scalable and semantic image annotation and retrieval. Comput Vis Image Underst 118:16–29

    Article  Google Scholar 

  64. 64.

    Zhao S, Yao H, Yang Y et al. (2014) “Affective image retrieval via multi-graph learning,”. Proc ACM Int Conf Multimed 1025–1028

  65. 65.

    Zheng Y-T, Neo S-Y, Chua T-S, Tian Q (2009) Toward a higher-level visual representation for object-based image retrieval. Vis Comput 25(1):13–23

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Mohsen Zand.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zand, M., Doraisamy, S., Abdul Halin, A. et al. Visual and semantic context modeling for scene-centric image annotation. Multimed Tools Appl 76, 8547–8571 (2017). https://doi.org/10.1007/s11042-016-3500-5

Download citation

Keywords

  • Automatic image annotation
  • Visual diversity
  • Mixture model
  • Visual context
  • Semantic context