Semi-supervised adaptive feature analysis and its application for multimedia understanding

Abstract

Multimedia understanding for high dimensional data is still a challenging work, due to redundant features, noises and insufficient label information it contains. Graph-based semi-supervised feature learning is an effective approach to address this problem. Nevertheless, Existing graph-based semi-supervised methods usually depend on the pre-constructed Laplacian matrix but rarely modify it in the subsequent classification tasks. In this paper, an adaptive local manifold learning based semi-supervised feature selection is proposed. Compared to the state-of-the-art, the proposed algorithm has two advantages: 1) Adaptive local manifold learning and feature selection are integrated jointly into a single framework, where both the labeled and unlabeled data are utilized. Besides, the correlations between different components are also considered. 2) A group sparsity constraint, i.e. l 2 , 1-norm, is imposed to select the most relevant features. We also apply the proposed algorithm to serval kinds of multimedia understanding applications. Experimental results demonstrate the effectiveness of the proposed algorithm.

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

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

References

  1. 1.

    Boutell MR, Luo J, Shen X, Brown CM (2004) Learning multi-label scene classification. Pattern Recogn 37:1757–1771

    Article  Google Scholar 

  2. 2.

    Cai D, Zhang C, He X (2010) Unsupervised Feature Selection for Multi-Cluster Data. conference on Knowledge discovery and data 333–342

  3. 3.

    Chang X, Yang Y (2016) Semisupervised feature analysis by mining correlations among multiple tasks. IEEE Transactions on Neural Networks and Learning Systems PP:1–12. doi:10.1109/TNNLS.2016.2582746

    MathSciNet  Article  Google Scholar 

  4. 4.

    Chang X, Nie F, Yang Y, Huang H (2014) A Convex Formulation for Semi-supervised Multi-label Feature Selection. In: Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence. pp 1171–1177

  5. 5.

    Chang X, Shen H, Wang S, et al (2014) Semi-supervised feature analysis for multimedia annotation by mining label correlation. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 8444 LNAI:74–85

  6. 6.

    Chang X, Yang Y, Hauptmann AG, et al (2015) Semantic Concept Discovery for Large-Scale Zero-Shot Event Detection. In: Proceedings of the 24th International Conference on Artificial Intelligence. AAAI Press, pp 2234–2240

  7. 7.

    Chang X, Nie F, Yang Y et al (2016) Convex Sparse PCA for Unsupervised Feature Learning. ACM Trans Knowl Discov Data 11:1–16. doi:10.1145/2910585

    Article  Google Scholar 

  8. 8.

    Chang X, Ma Z, Yang Y et al (2017) Bi-Level Semantic Representation Analysis for Multimedia Event Detection. IEEE Trans Cybern 47:1180–1197. doi:10.1109/TCYB.2016.2539546

    Article  Google Scholar 

  9. 9.

    Chen C, Jafari R, Kehtarnavaz N (2015) UTD-MHAD: A multimodal dataset for human action recognition utilizing a depth camera and a wearable inertial sensor. In: Proceedings - International Conference on Image Processing, ICIP. pp 168–172

  10. 10.

    Chen B, Yang J, Jeon B, Zhang X (2017) Kernel quaternion principal component analysis and its application in RGB-D object recognition. Neurocomputing. doi:10.1016/j.neucom.2017.05.047

    Article  Google Scholar 

  11. 11.

    Du L, Shen Y-D (2015) Unsupervised Feature Selection with Adaptive Structure Learning. Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 209–218

  12. 12.

    Duda RO, Hart PE, Stork DG (2000) Pattern Classification, 2Nd edn. Wiley-Interscience

  13. 13.

    He X, Cai D, Niyogi P (2005) Laplacian Score for Feature Selection. Adv Neural Inf Proces Syst 18:507–514

    Google Scholar 

  14. 14.

    Hou C, Nie F, Li X et al (2014) Joint embedding learning and sparse regression: A framework for unsupervised feature selection. IEEE Trans Cybern 44:793–804

    Article  Google Scholar 

  15. 15.

    Hou C, Nie F, Yi D, Tao D (2015) Discriminative Embedded Clustering: A Framework for Grouping High-Dimensional Data. IEEE Trans Neural Netw Learn Syst 26:1287–1299

    MathSciNet  Article  Google Scholar 

  16. 16.

    Li H, Wang M, Hua XS (2009) MSRA-MM 2.0: A large-scale web multimedia dataset. In: ICDM Workshops 2009 - IEEE International Conference on Data Mining. pp 164–169

  17. 17.

    Li Z, Yang Y, Liu J, et al (2012) Unsupervised Feature Selection Using Nonnegative Spectral Analysis. Twenty-Sixth AAAI Conference on Artificial Intelligence Unsupervised 1026–1032

  18. 18.

    Liu Y, Nie F, Wu J, Chen L (2013) Efficient semi-supervised feature selection with noise insensitive trace ratio criterion. Neurocomputing 105:12–18

    Article  Google Scholar 

  19. 19.

    Luo M, Chang X, Nie L, et al (2017) An Adaptive Semisupervised Feature Analysis for Video Semantic Recognition. IEEE Transactions on Cybernetics 1–13. doi:10.1109/TCYB.2017.2647904

    Article  Google Scholar 

  20. 20.

    Ma Z, Nie F, Yang Y et al (2012) Discriminating Joint Feature Analysis for Multimedia Data Understanding. IEEE Trans Multimed 14:1662–1672

    Article  Google Scholar 

  21. 21.

    Ma Z, Nie F, Yang Y et al (2012) Web Image Annotation via Subspace-Sparsity Collaborated Feature Selection. IEEE Tran Multimed 14:1021–1030

    Article  Google Scholar 

  22. 22.

    Ma Z, Yang Y, Sebe N, Hauptmann AG (2014) Multiple Features But Few Labels? In: Proceedings of the ACM International Conference on Multimedia - MM ‘14. ACM Press, New York, New York, USA, pp 77–86

  23. 23.

    Nene SA, Nayar SK, Murase H (1996) Columbia University Image Library (COIL-20)

  24. 24.

    Nie F, Xu D, Tsang IWH, Zhang C (2010) Flexible manifold embedding: A framework for semi-supervised and unsupervised dimension reduction. IEEE Trans Image Process 19:1921–1932

    MathSciNet  Article  Google Scholar 

  25. 25.

    Nie F, Huang H, Cai X, Ding C (2010) Efficient and Robust Feature Selection via Joint l2,1-Norms Minimization. Adv Neural Inf Proces Syst 23:1813–1821

    Google Scholar 

  26. 26.

    Nie F, Wang H, Huang H, Ding C (2011) Unsupervised and semi-supervised learning via l1-norm graph. 2011 International Conference on Computer Vision 2268–2273

  27. 27.

    Nie F, Wang X, Huang H (2014) Clustering and projected clustering with adaptive neighbors. Proceedings of the 20th ACM SIGKDD international conference on Knowledge discovery and data mining - KDD ‘14 977–986

  28. 28.

    Nie F, Zhu W, Li X (2016) Unsupervised Feature Selection with Structured Graph Optimization. Proceedings of the 30th Conference on Artificial Intelligence (AAAI 2016) 13:1302–1308

  29. 29.

    Siddiqi MH, Ali R, Idris M et al (2016) Human facial expression recognition using curvelet feature extraction and normalized mutual information feature selection. Multimed Tools Appl 75:935–959

    Article  Google Scholar 

  30. 30.

    Sigal L, Balan AO, Black MJ (2010) HumanEva: Synchronized video and motion capture dataset and baseline algorithm for evaluation of articulated human motion. Int J Comput Vis 87:4–27

    Article  Google Scholar 

  31. 31.

    Song J, Yang Y, Huang Z et al (2013) Effective multiple feature hashing for large-scale near-duplicate video retrieval. IEEE Trans Multimed 15:1997–2008

    Article  Google Scholar 

  32. 32.

    Wang X, Zhang X, Zeng Z et al (2016) Unsupervised spectral feature selection with l 1 -norm graph. Neurocomputing 200:47–54

    Article  Google Scholar 

  33. 33.

    Wang X, Chen R-C, Hong C et al (2017) Semi-supervised multi-label feature selection via label correlation analysis with l 1 -norm graph embedding. Image Vis Comput 63:10–23. doi:10.1016/j.imavis.2017.05.004

    Article  Google Scholar 

  34. 34.

    Yang Y, Shen HT, Ma Z, et al (2011) l2,1-Norm regularized discriminative feature selection for unsupervised learning. IJCAI International Joint Conference on Artificial Intelligence 1589–1594

  35. 35.

    Yang Y, Nie F, Xu D et al (2012) A Multimedia Retrieval Framework Based on Semi-Supervised Ranking and Relevance Feedback. IEEE Trans Pattern Anal Mach Intell 34:723–742

    Article  Google Scholar 

  36. 36.

    Yang Y, Ma Z, Hauptmann AG et al (2013) Feature Selection for Multimedia Analysis by Shareing Information Among Multiple Tasks. IEEE Trans Multimed 15:661–669

    Article  Google Scholar 

  37. 37.

    Yang Y, Ma Z, Nie F et al (2014) Multi-Class Active Learning by Uncertainty Sampling with Diversity Maximization. Int J Comput Vis 113:113–127

    MathSciNet  Article  Google Scholar 

  38. 38.

    Yang XK, He L, Qu D, Zhang W (2016) Semi-supervised minimum redundancy maximum relevance feature selection for audio classification. Multimedia Tools and Applications 1–27

  39. 39.

    Yuan C, Sun X, Lv R (2016) Fingerprint liveness detection based on multi-scale LPQ and PCA. China Commun 13:60–65. doi:10.1109/CC.2016.7559076

    Article  Google Scholar 

  40. 40.

    Zeng Z, Wang X, Zhang J, Wu Q (2016) Semi-supervised feature selection based on local discriminative information. Neurocomputing 173:102–109

    Article  Google Scholar 

  41. 41.

    Zhang Z, Bai L (2015) Unsupervised Feature Selection by Graph Optimization. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol 9279, pp 130–140

    Google Scholar 

  42. 42.

    Zhao Z, Liu H (2007) Semi-supervised feature selection via spectral analysis. Proceedings of the 2007 siam International Conference on Data Mining 641–646

  43. 43.

    Zhou Z-H, Zhang M-L (2007) Multi-Instance Multi-Label Learning with Application to Scene Classification. In: Schölkopf B, Platt J, Hoffman T (eds) Advances in Neural Information Processing Systems. MIT Press, Cambridge, pp 1609–1616

    Google Scholar 

Download references

Acknowledgements

This paper is supported by National Natural Science Foundation of China (Grant No. 61502405), National Natural Science Foundation of Fujian Province, China (Grant Nos. 2016 J01324, 2016J01327, 2017 J01511), the International Science and Technology Cooperation Program of Xiamen university of technology (No.E201400400), Xiamen Science and Technology Planning Project (Nos.3502Z20143030, 3502Z20103037, 3502Z20133043), Scientific Research Fund of Fujian Provincial Education Department (Nos. JA15385, JAT160357), and Ministry of Science and Technology, Taiwan, (Grant Nos. MOST-104-2221-E-324-019-MY2, MOST-103-2632-E-324-001-MY3).

Author information

Affiliations

Authors

Corresponding author

Correspondence to Rung-Ching Chen.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wang, X., Chen, R., Yan, F. et al. Semi-supervised adaptive feature analysis and its application for multimedia understanding. Multimed Tools Appl 77, 3083–3104 (2018). https://doi.org/10.1007/s11042-017-4990-5

Download citation

Keywords

  • feature selection
  • semi-supervised learning
  • adaptive learning
  • image annotation
  • 3D human action recognition