Flickr image quality evaluation by deeply fusing heterogeneous visual cues

  • Yongjun Zheng
  • Weiyu Di
  • Shen Jiang


Flickr is a photo and video hosting site with over 2 million groups. There are more than 35 million new photos uploading every day. But at present, there are no tools to organize these huge numbers of users’ aesthetic tendency. Although Flickr allows users to add different groups manually, they are difficult to maintain updates when new users are added or deleted. In this paper, we put forward a series of Flickr users system that each loop contains similar users aesthetic tendency. We observed: (1) an aesthetic model of thought should be flexible, because different visual features represent different data sets. And (2) significant differences are existing in the number of photos from different Flickr users. So in our work, a new probabilistic topic model is proposed to describe the aesthetic interest of each Flickr user’s potential spatial distribution. After that, an affinity graph is described by aesthetic interests of Flickr users. Obviously, intensive users of Flickr are similar in taste. Thus, these users are divided into different Flickr bounds efficient dense graph discover. It is proposed that the Flickr bound discovery is fully automatic. Experiments show that our proposed method is accurate for 60000 Flickr user community.


Machine learning Multi-cue fusion Aesthetic tendency Flickr Graph mining 



This work is supported by National Science and Technology Major Project of the Ministry of Science and Technology of China (Grant No. 2015ZX02101),National Natural Science Foundation of China (Grant No. 51775530) and by the National Social Science Fund Projects of Art of China (Grant No: 2014CC03652).


  1. 1.
    Bukal M, Markovic I, Petrovic I (2014) Composite distance based approach to Von Mises mixture reduction. Inf Fusion 20:136–145CrossRefGoogle Scholar
  2. 2.
    Celeux G, Govaert G (1993) A classification EM algorithm for clustering and two stochastic versions. Comput Stat Data Anal 14(2):315–332MathSciNetMATHGoogle Scholar
  3. 3.
    Chen YW., Lin CJ. (2006) Combining SVMs with Various Feature Selection Strategies. In: Guyon I., Nikravesh M., Gunn S., Zadeh L.A. (eds) Feature Extraction. Studies in Fuzziness and Soft Computing, vol 207. Springer, BerlinGoogle Scholar
  4. 4.
    Chen Y, Cremers AB, Cao Z (2014) Interactive color image segmentation via iterative evidential labeling. Inf Fusion 20:292–304CrossRefGoogle Scholar
  5. 5.
    Cheng B, Ni B, Yan S, Tian Q (2010) Learning to photograph. ACM Multimed. pp 291–300Google Scholar
  6. 6.
    Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. In: Proc. IEEE Conf. Comput. Vis. Pattern Recognit., June 2005, pp 886–893Google Scholar
  7. 7.
    Erlandsson T, Niklasson L (2014) Automatic evaluation of air mission routes with respect to combat survival. Inf Fusion 20:88–98CrossRefGoogle Scholar
  8. 8.
    Frank M, Streich AP, Basin D, Buhmann JM (2012) Multi-assignment clustering for Boolean data. J March Learn Res 13(1):459–489MathSciNetMATHGoogle Scholar
  9. 9.
    Frey BJ, Dueck D (2007) Clustering by passing messages between data points. Sci 315:972–976MathSciNetCrossRefMATHGoogle Scholar
  10. 10.
    Fu C, Huhns MN, Yang S (2014) A consensus framework for multiple attribute group decision analysis in an evidential reasoning context. Inf Fusion 17:22–35CrossRefGoogle Scholar
  11. 11.
    Govaert G, Nadif M (2007) An EM algorithm for the block mixture model. IEEE T-PAMI 27(4):643–647CrossRefMATHGoogle Scholar
  12. 12.
    Govaert G, Nadif M (2013) Co-clustering: models, algorithms and applications. Wiley-ISTE, Nov. 2013Google Scholar
  13. 13.
    Gregory S (2008) A fast algorithm to find overlapping communities in networks. Mach Learn Knowl Discov Database Springer Berlin Heidelberg 5211:408–423CrossRefGoogle Scholar
  14. 14.
    H.-Viedma E, Cabrerizo FJ, Kacprzyk J, Pedrycz W (2014) A review of soft consensus models in a fuzzy environment. Inf Fusion 17:4–13CrossRefGoogle Scholar
  15. 15.
    Hofmann T (1999) Probabilistic latent semantic analysis. In: Proc. of Uncert. In artifi. Intelli., pp 289–296Google Scholar
  16. 16.
    Huang Z-H, Li W-J, Wang J, Zhang T (2015) Face recognition based on pixel-level and feature-level fusion of the top-level’s wavelet sub-bands. Inf Fusion 181(18):22, 95–22,104Google Scholar
  17. 17.
    Jiang L, Liu H, Cai J (2015) The power average operator for unbalanced linguistic term sets. Inf Fusion 22:85–94CrossRefGoogle Scholar
  18. 18.
    Kimura A, Marcos A, Kataoka K (2013) Image context discovery from socially curated contents. ACM MultimediaGoogle Scholar
  19. 19.
    Kuhn HW (1956) Variants of the Hungarian method for assignment problems. Nav Res Logist Q 24(3):253–258MathSciNetCrossRefMATHGoogle Scholar
  20. 20.
    Lancichinetti A, Fortunato S, Radicchi F (2008) Benchmark graphs for testing community detection algorithms. Phys Rev E 78:046110CrossRefGoogle Scholar
  21. 21.
    Lazebnik S, Schmid C, Ponce J (2006) Beyond bags of features: spatial pyramid matching for recognizing natural scene categories. In: Proc. of CVPRGoogle Scholar
  22. 22.
    Liu H, Yan S (2010) Robust graph mode seeking by graph shift. In: Proc. Int’l Conf. Mach. Learn., pp 671–678Google Scholar
  23. 23.
    Leskovec J, Lang KJ, Mahoney M (2010) Empirical comparison of algorithms for network community detection. In: Proc. Int’l Conf. World Wide Web, pp 631–640Google Scholar
  24. 24.
    Liu X, Song M, Tao D, Liu Z, Zhang L, Bu J, Chen C (2013) Semi-supervised node splitting for random forest construction. IEEE Computer Vision and Pattern Recognition (CVPR), pp 492C499Google Scholar
  25. 25.
    Learning to track multiple targets. IEEE transactions on neural networks and learning systems, 26(5):1060–1073Google Scholar
  26. 26.
    Mignotte M (2014) A label field fusion model with a variation of information estimator for image segmentation. Inf Fusion 20:7–20CrossRefGoogle Scholar
  27. 27.
    Nishiyama M, Okabe T, Sato Y, Sato I (2009) Sensation-based photo cropping. ACM Multimedia, pp 669–672Google Scholar
  28. 28.
    Ochoa SF, Santos RM (2015) Human-centric wireless sensor networks to improve information availability during urban search and rescue activities. Inf Fusion 22:71–84CrossRefGoogle Scholar
  29. 29.
    Poslad S, Kesorn K (2014) A multi-modal incompleteness ontology model (mmio) to enhance information fusion for image retrieval. Inf Fusion 20:136–145CrossRefGoogle Scholar
  30. 30.
    Stricker MA, Orengo M (1995) Similarity of color images. Storage and Retrieval for Image and Video Databases, pp 381–392Google Scholar
  31. 31.
    Yan X-Q, Liu Y, Li B, Liu X-M (2015) A memoryless binary query tree based successive scheme for passive RFID tag collision resolution. Inf Fusion 22:26–38CrossRefGoogle Scholar
  32. 32.
    Yang J, Leskovec J (2012) Community-affiliation graph model for overlapping network community detection. IEEE Int’l Conf. Data Mining, pp 1170–1175Google Scholar
  33. 33.
    Yin X-C, Huang K, Yang C, Hao H-W (2014) Convex ensemble learning with sparsity and diversity. Inf Fusion 20:4959CrossRefGoogle Scholar
  34. 34.
    Yin Y, Shen Z, Zhang L, Zimmermann R (2014) Spatial-temporal tag mining for automatic geospatial video annotations. ACM Trans Multimed Comput Commun Appl (TOMCCAP) 11(2):29Google Scholar
  35. 35.
    Yoshida T (2010) Toward finding hidden communities based on user profile. Workshops of IEEE Int’l Conf. Data Mining, Dec 2010, pp 380–387Google Scholar
  36. 36.
    Zhang L, Song M, Liu Z, Liu X, Jiajun B, Chen C (2013) Probabilistic graphlet cut: exploring spatial structure cue for weakly supervised image segmentation. IEEE Comput Vis Pattern Recognit (CVPR) 2013:1908–1915Google Scholar
  37. 37.
    Zhang L, Tao D, Liu X, Song M, Chen C (2013) Grassmann multimodal implicit feature selection multimedia systems. Multimedia System 20(6):659–674Google Scholar
  38. 38.
    Zhang G, Dong Y, Xu Y (2014) Consistency and consensus measures for linguistic preference relations based on distribution assessments. Inf Fusion 17:46–55CrossRefGoogle Scholar
  39. 39.
    Zhang L, Gao Y, Xia Y, Xu C, Lv K, Shen J, Ji R (2014) Representative discovery of structure cues for weakly-supervised image segmentation. IEEE Trans Multimed 16(2):470–479CrossRefGoogle Scholar
  40. 40.
    Zhang L, Yang Y, Gao Y, Wang C, Yu Y, Li X (2014) A probabilistic associative model for segmenting weakly-supervised images. IEEE Trans Image Process 23(9):4150–4159MathSciNetCrossRefMATHGoogle Scholar
  41. 41.
    Zhang L, Gao Y, Hong C, Feng Y, Zhu J, Deng C (2014) Feature correlation hypergraph: exploiting high-order potentials for multimodal recognition. IEEE Trans Cybern (T-CYB) 44(8):1408–1419CrossRefGoogle Scholar
  42. 42.
    Zhang L, Gao Y, Ji R, Ke L, Shen J (2014) Representative discovery of structure cues for weakly-supervised image segmentation. IEEE Trans Multimed (T-MM) 16(2):470C 479Google Scholar
  43. 43.
    Zhang L, Song M, Yang Y, Zhao Q, Chen Z, Sebe N (2014) Weakly supervised photo cropping. IEEE Trans Multimed 16(1):94–107CrossRefGoogle Scholar
  44. 44.
    Zhang L, Ji R, Xia Y, Li X (2014) Learning a probabilistic topology discovering model for scene categorization. IEEE Trans Neural Netw Learn Syst (T-NNLS) 26(8):1622–1634MathSciNetCrossRefGoogle Scholar
  45. 45.
    Zhang Y, Zhang L, Zimmermann R (2014) Aesthetics-guided summarization from multiple user generated videos. ACM Trans Multimed Comput Commun Appl (TOMCCAP) 11(2):24Google Scholar
  46. 46.
    Zhang L, Gao Y, Zhang C, Tian Q, Zimmermann R (2014) Perception-guided multimodal aesthetics discovery for photo quality assessment. ACM Multimedia.
  47. 47.
    Zhang L, Gao Y, Hong R, Hu Y, Ji R, Dai Q (2015) Probabilistic skimlet fusion for summarizing multiple consumer landmark videos. IEEE Trans Multimed (T-MM) 71(1):40–49CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2017

Authors and Affiliations

  1. 1.College of Metrology and Measurement EngineeringChina Jiliang UniversityZhejiangChina
  2. 2.School of Journalism & CommunicationsBeijing Normal UniversityBeijingChina

Personalised recommendations