Advertisement

Knowledge and Information Systems

, Volume 50, Issue 2, pp 585–605 | Cite as

Visual domain adaptation via transfer feature learning

  • Jafar Tahmoresnezhad
  • Sattar Hashemi
Regular Paper

Abstract

One of the serious challenges in computer vision and image classification is learning an accurate classifier for a new unlabeled image dataset, considering that there is no available labeled training data. Transfer learning and domain adaptation are two outstanding solutions that tackle this challenge by employing available datasets, even with significant difference in distribution and properties, and transfer the knowledge from a related domain to the target domain. The main difference between these two solutions is their primary assumption about change in marginal and conditional distributions where transfer learning emphasizes on problems with same marginal distribution and different conditional distribution, and domain adaptation deals with opposite conditions. Most prior works have exploited these two learning strategies separately for domain shift problem where training and test sets are drawn from different distributions. In this paper, we exploit joint transfer learning and domain adaptation to cope with domain shift problem in which the distribution difference is significantly large, particularly vision datasets. We therefore put forward a novel transfer learning and domain adaptation approach, referred to as visual domain adaptation (VDA). Specifically, VDA reduces the joint marginal and conditional distributions across domains in an unsupervised manner where no label is available in test set. Moreover, VDA constructs condensed domain invariant clusters in the embedding representation to separate various classes alongside the domain transfer. In this work, we employ pseudo target labels refinement to iteratively converge to final solution. Employing an iterative procedure along with a novel optimization problem creates a robust and effective representation for adaptation across domains. Extensive experiments on 16 real vision datasets with different difficulties verify that VDA can significantly outperform state-of-the-art methods in image classification problem.

Keywords

Transfer learning Unsupervised domain adaptation Domain invariant clustering Domain shift Invariant feature representation 

References

  1. 1.
    Gopalan R, Li R, Chellappa R (2014) Unsupervised adaptation across domain shifts by generating intermediate data representations. IEEE Trans Pattern Anal Mach Intell 36(11):2288–2302CrossRefGoogle Scholar
  2. 2.
    Gong B, Grauman K, Sha F (2013) Connecting the dots with landmarks: Discriminatively learning domain-invariant features for unsupervised domain adaptation. In: Proceedings of the 30th international conference on machine learning, pp 222–230Google Scholar
  3. 3.
    Bergamo A, Torresani L (2010) Exploiting weakly-labeled web images to improve object classification: a domain adaptation approach. In: Advances in neural information processing systems, pp 181–189Google Scholar
  4. 4.
    Hoffman J, Kulis B, Darrell T, Saenko K (2012) Discovering latent domains for multisource domain adaptation. In: Computer Vision–ECCV 2012, pp 702–715. SpringerGoogle Scholar
  5. 5.
    Chen M, Weinberger KQ, Blitzer J (2011) Co-training for domain adaptation. In: Advances in neural information processing systems, pp 2456–2464Google Scholar
  6. 6.
    Baktashmotlagh M, Harandi MT, Lovell BC, Salzmann M (2013) Unsupervised domain adaptation by domain invariant projection. In: 2013 IEEE international conference on computer vision (ICCV), pp 769–776Google Scholar
  7. 7.
    Gheisari M, Baghshah MS (2015) Unsupervised domain adaptation via representation learning and adaptive classifier learning. Neurocomputing 165:300–311CrossRefGoogle Scholar
  8. 8.
    Tahmoresnezhad J, Hashemi S (2015) Common feature extraction in multi-source domains for transfer learning. In IEEE 2015 7th Conference on information and knowledge technology (IKT), pp 1–5Google Scholar
  9. 9.
    Fernando B, Habrard A, Sebban M, Tuytelaars T (2013) Unsupervised visual domain adaptation using subspace alignment. In: 2013 IEEE international conference on computer vision (ICCV), pp 2960–2967Google Scholar
  10. 10.
    Xiong C, McCloskey S, Hsieh S-H, Corso JJ (2014) Latent domains modeling for visual domain adaptation. In: Proceedings of AAAI conference on artificial intelligence (AAAI)Google Scholar
  11. 11.
    Gong B, Grauman K, Sha F (2014) Learning kernels for unsupervised domain adaptation with applications to visual object recognition. Int J Comput Vision 109(1–2):3–27MathSciNetCrossRefzbMATHGoogle Scholar
  12. 12.
    Cuong V, Duin RPW, Piqueras-Salazar I, Loog M (2013) A generalized fisher based feature extraction method for domain shift. Pattern Recognit 46(9):2510–2518CrossRefGoogle Scholar
  13. 13.
    Gopalan R, Li R, Chellappa R (2011) Domain adaptation for object recognition: an unsupervised approach. In: 2011 IEEE international conference on computer vision (ICCV), pp 999–1006Google Scholar
  14. 14.
    Ben-David S, Blitzer J, Crammer K, Pereira F et al (2007) Analysis of representations for domain adaptation. Adv Neural Inf Process Syst 19:137Google Scholar
  15. 15.
    Uguroglu S, Carbonell J (2011) Feature selection for transfer learning. In: Machine learning and knowledge discovery in databases, pp 430–442. SpringerGoogle Scholar
  16. 16.
    Pan SJ, Kwok JT, Yang Q (2008) Transfer learning via dimensionality reduction. AAAI 8:677–682Google Scholar
  17. 17.
    Quanz B, Huan J, Mishra M (2012) Knowledge transfer with low-quality data: a feature extraction issue. IEEE Trans Knowl Data Eng 24(10):1789–1802CrossRefGoogle Scholar
  18. 18.
    Zhong E, Fan W, Peng J, Zhang K, Ren J, Turaga D, Verscheure O (2009) Cross domain distribution adaptation via kernel mapping. In: Proceedings of the 15th ACMSIGKDD international conference on Knowledge discovery and data mining, pp 1027–1036Google Scholar
  19. 19.
    Sun Q, Chattopadhyay R, Panchanathan S, Ye J (2011) A two-stage weighting framework for multi-source domain adaptation. In: Advances in neural information processing systems, pp 505–513Google Scholar
  20. 20.
    Long M, Wang J, Ding G, Sun J, YuPhilip S (2013) Transfer feature learning with joint distribution adaptation. In: 2013 IEEE international conference on computer vision (ICCV), pp 2200–2207Google Scholar
  21. 21.
    Tahmoresnezhad J, Hashemi S (2015) A generalized kernel-based random k-sample sets method for transfer learning. Iran J Sci Technol Trans Electrical Eng 39:193–207Google Scholar
  22. 22.
    Gretton A, Borgwardt KM, Rasch M, Schölkopf B, Smola AJ (2006) A kernel method for the two-sample-problem. In: Advances in neural information processing systems, pp 513–520Google Scholar
  23. 23.
    Jolliffe I (2002) Principal component analysis. Wiley, New YorkzbMATHGoogle Scholar
  24. 24.
    Jie L, Behbood V, Hao P, Zuo H, Xue S, Zhang G (2015) Transfer learning using computational intelligence: a survey. Knowl Based Syst 80:14–23CrossRefGoogle Scholar
  25. 25.
    Pan SJ, Yang Q (2010) A survey on transfer learning. IEEE Trans Knowl Data Eng 22(10):1345–1359CrossRefGoogle Scholar
  26. 26.
    Blitzer J, McDonald R, Pereira F (2006) Domain adaptation with structural correspondence learning. In: Proceedings of the 2006 conference on empirical methods in natural language processing, Association for Computational Linguistics, pp 120–128Google Scholar
  27. 27.
    Pan SJ, Ni X, Sun J-T, Yang Q, Chen Z (2010) Cross-domain sentiment classification via spectral feature alignment. In: Proceedings of the 19th international conference on World wide web, ACM, pp 751–760Google Scholar
  28. 28.
    Huang J, Gretton A, Borgwardt KM, Schölkopf B, Smola AJ (2006) Correcting sample selection bias by unlabeled data. In: Advances in neural information processing systems, pp 601–608Google Scholar
  29. 29.
    Pan SJ, Tsang IW, Kwok JT, Yang Q (2011) Domain adaptation via transfer component analysis. Neural Netw 22(2):199–210CrossRefGoogle Scholar
  30. 30.
    Daumé H III (2009) Frustratingly easy domain adaptation.arXiv preprint arXiv:0907.1815
  31. 31.
    Duan L, Tsang IW, Xu D, Maybank SJ (2009)Domain transfer SVM for video concept detection. In: IEEE Conference on computer vision and pattern recognition, CVPR 2009, pp 1375–1381Google Scholar
  32. 32.
    Gong B, Shi Y, Sha F, Grauman K (2012) Geodesic flow kernel for unsupervised domain adaptation. In: 2012 IEEE Conference on computer vision and pattern recognition (CVPR), pp 2066–2073Google Scholar
  33. 33.
    Saenko K, Kulis B, Fritz M, Darrell T (2010)Adapting visual category models to new domains. In: Computer vision–ECCV 2010, pp 213–226. SpringerGoogle Scholar
  34. 34.
    Chen Q, Xue B, Zhang M (2015) Generalisation and domain adaptation in GP with gradient descent for symbolic regression. In: 2015 IEEE congress on evolutionary computation (CEC), pp 1137–1144Google Scholar
  35. 35.
    Iqbal M, Browne WN, Zhang M (2014) Reusing building blocks of extracted knowledge to solve complex, large-scale Boolean problems. IEEE Trans Evol Comput 18(4):465–480CrossRefGoogle Scholar
  36. 36.
    Jiang J, Zhai CX (2007) Instance weighting for domain adaptation in NLP. ACL 7:264–271CrossRefGoogle Scholar
  37. 37.
    Jiang J (2008) A literature survey on domain adaptation of statistical classifiers. URL: http://sifaka.cs.uiuc.edu/jiang4/domainadaptation/survey
  38. 38.
    Duan L, Tsang IW, Xu D, Chua T-S (2009) Domain adaptation from multiple sources via auxiliary classifiers. In: Proceedings of the 26th annual international conference on machine learning, ACM, pp 289–296Google Scholar
  39. 39.
    Long M, Wang J, Ding G, Pan SJ, Yu PS (2014) Adaptation regularization: a general framework for transfer learning. IEEE Trans Knowl Data Eng 26(5):1076–1089CrossRefGoogle Scholar
  40. 40.
    Bruzzone L, Marconcini M (2010) Domain adaptation problems: a DASVM classification technique and a circular validation strategy. IEEE Trans Pattern Anal Mach Intell 32(5):770–787CrossRefGoogle Scholar
  41. 41.
    Satpal S, Sarawagi S (2007) Domain adaptation of conditional probability models via feature subsetting. In: Knowledge discovery in databases: PKDD 2007, pp 224–235. SpringerGoogle Scholar
  42. 42.
    Si S, Tao D, Geng B (2010) Bregman divergence-based regularization for transfer subspace learning. IEEE Trans Knowl Data Eng 22(7):929–942CrossRefGoogle Scholar
  43. 43.
    Jhuo I-H, Liu D, Lee DT, Chang S-F et al (2012) Robust visual domain adaptation with low-rank reconstruction. In: 2012 IEEE conference on computer vision and pattern recognition (CVPR), pp 2168–2175Google Scholar
  44. 44.
    Qiu Q, Patel VM, Turaga P, Chellappa R (2012) Domain adaptive dictionary learning. In: Computer vision–ECCV2012, pp 631–645. SpringerGoogle Scholar
  45. 45.
    Roy SD, Mei T, Zeng W, Li S (2012) Social transfer: cross-domain transfer learning from social streams for media applications. In: Proceedings of the 20th ACM international conference on multimedia, pp 649–658Google Scholar
  46. 46.
    Long M, Wang J, Ding G, Sun J, Yu PS (2014) Transfer joint matching for unsupervised domain adaptation. In: 2014 IEEE conference on computer vision and pattern recognition (CVPR), pp 1410–1417Google Scholar
  47. 47.
    Long M, Wang J, Sun J, Yu PS (2015) Domain invariant transfer kernel learning. IEEE Trans Knowl Data Eng 27(6):1519–1532CrossRefGoogle Scholar
  48. 48.
    Griffin G, Holub A, Perona P (2007) Caltech-256 object category datasetGoogle Scholar
  49. 49.
    Bay H, Tuytelaars T, Van Gool L (2006) Surf: speeded up robust features. In: Computer vision–ECCV 2006, pp 404–417. SpringerGoogle Scholar
  50. 50.
    Nene SA, Nayar SK, Murase H et al (1996) Columbia object image library (coil-20). Technical report, TechnicalReport CUCS-005-96Google Scholar

Copyright information

© Springer-Verlag London 2016

Authors and Affiliations

  1. 1.CSE and IT DepartmentShiraz UniversityShirazIran

Personalised recommendations