Domain Adaptive Fisher Vector for Visual Recognition

详细信息    查看全文

• 关键词：Domain adaptation ; Fisher vector
• 刊名：Lecture Notes in Computer Science
• 出版年：2016
• 出版时间：2016
• 年：2016
• 卷：9910
• 期：1
• 页码：550-566
• 全文大小：550 KB
• 参考文献：1.Jaakkola, T.S., Haussler, D., et al.: Exploiting generative models in discriminative classifiers. In: NIPS (1999)
  2.Perronnin, F., Sánchez, J., Mensink, T.: Improving the fisher kernel for large-scale image classification. In: Daniilidis, K., Maragos, P., Paragios, N. (eds.) ECCV 2010, Part IV. LNCS, vol. 6314, pp. 143–156. Springer, Heidelberg (2010)CrossRef
  3.Simonyan, K., Vedaldi, A., Zisserman, A.: Deep Fisher networks for large-scale image classification. In: NIPS (2013)
  4.Wei, Y., Xia, W., Huang, J., Ni, B., Dong, J., Zhao, Y., Yan, S.: CNN: Single-label to multi-label (2014). arXiv preprint arXiv:​1406.​5726
  5.Uricchio, T., Bertini, M., Seidenari, L., Bimbo, A.: Fisher encoded convolutional bag-of-windows for efficient image retrieval and social image tagging. In: ICCV (2015)
  6.Wang, H., Schmid, C.: Action recognition with improved trajectories. In: ICCV (2013)
  7.Peng, X., Zou, C., Qiao, Y., Peng, Q.: Action recognition with stacked fisher vectors. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part V. LNCS, vol. 8693, pp. 581–595. Springer, Heidelberg (2014)
  8.Dixit, M., Rasiwasia, N., Vasconcelos, N.: Adapted Gaussian models for image classification. In: CVPR (2011)
  9.Torralba, A., Efros, A.A.: Unbiased look at dataset bias. In: CVPR (2011)
  10.Kim, H.J., Adluru, N., Banerjee, M., Vemuri, B.C., Singh, V.: Interpolation on the manifold of K component GMMs. In: ICCV (2015)
  11.Huang, J., Smola, A., Gretton, A., Borgwardt, K., Scholkopf, B.: Correcting sample selection bias by unlabeled data. In: NIPS (2007)
  12.Bruzzone, L., Marconcini, M.: Domain adaptation problems: A DASVM classification technique and a circular validation strategy. T-PAMI 32(5), 770–787 (2010)CrossRef
  13.Gong, B., Shi, Y., Sha, F., Grauman, K.: Geodesic flow kernel for unsupervised domain adaptation. In: CVPR (2012)
  14.Gopalan, R., Li, R., Chellappa, R.: Domain adaptation for object recognition: an unsupervised approach. In: ICCV (2011)
  15.Fernando, B., Habrard, A., Sebban, M., Tuytelaars, T.: Unsupervised visual domain adaptation using subspace alignment. In: ICCV (2013)
  16.Baktashmotlagh, M., Harandi, M.T., Lovell, B.C., Salzmann, M.: Unsupervised domain adaptation by domain invariant projection. In: ICCV (2013)
  17.Pan, S.J., Tsang, I.W., Kwok, J.T., Yang, Q.: Domain adaptation via transfer component analysis. T-NN 22(2), 199–210 (2011)
  18.Aljundi, R., Emonet, R., Muselet, D., Sebban, M.: Landmarks-based kernelized subspace alignment for unsupervised domain adaptation. In: CVPR (2015)
  19.Shekhar, S., Patel, V., Nguyen, H., Chellappa, R.: Generalized domain-adaptive dictionaries. In: CVPR (2013)
  20.Shao, M., Kit, D., Fu, Y.: Generalized transfer subspace learning through low-rank constraint. IJCV 109(1–2), 74–93 (2014)MathSciNet CrossRef MATH
  21.Yuan, M., Lin, Y.: Model selection and estimation in regression with grouped variables. J. Royal Stat. Soc. Ser. B (Statistical Methodology) 68(1), 49–67 (2006)MathSciNet CrossRef MATH
  22.Perronnin, F., Dance, C.: Fisher kernels on visual vocabularies for image categorization. In: CVPR (2007)
  23.Liu, L., Shen, C., Wang, L., van den Hengel, A., Wang, C.: Encoding high dimensional local features by sparse coding based Fisher vectors. In: NIPS (2014)
  24.Gong, Y., Wang, L., Guo, R., Lazebnik, S.: Multi-scale orderless pooling of deep convolutional activation features. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part VII. LNCS, vol. 8695, pp. 392–407. Springer, Heidelberg (2014)
  25.Oneata, D., Verbeek, J., Schmid, C.: Action and event recognition with Fisher vectors on a compact feature set. In: ICCV (2013)
  26.Caseiro, R., Henriques, J.F., Martins, P., Batista, J.: Beyond the shortest path: Unsupervised domain adaptation by sampling subspaces along the spline flow. In: CVPR (2015)
  27.Kulis, B., Saenko, K., Darrell, T.: What you saw is not what you get: Domain adaptation using asymmetric kernel transforms. In: CVPR (2011)
  28.Sun, B., Feng, J., Saenko, K.: Return of frustratingly easy domain adaptation. In: AAAI (2016)
  29.Duan, L., Tsang, I.W., Xu, D.: Domain transfer multiple kernel learning. T-PAMI 34, 465–479 (2012)CrossRef
  30.Duan, L., Xu, D., Tsang, I.W., Luo, J.: Visual event recognition in videos by learning from web data. T-PAMI 34, 1667–1680 (2012)CrossRef
  31.Li, W., Niu, L., Xu, D.: Exploiting privileged information from web data for image categorization. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014, Part V. LNCS, vol. 8693, pp. 437–452. Springer, Heidelberg (2014)
  32.Niu, L., Li, W., Xu, D.: Exploiting privileged information from web data for action and event recognition. In: IJCAI (2016)
  33.Jhuo, I.H., Liu, D., Lee, D., Chang, S.F., et al.: Robust visual domain adaptation with low-rank reconstruction. In: CVPR (2012)
  34.Zhou, X., Cui, N., Li, Z., Liang, F., Huang, T.S.: Hierarchical Gaussianization for image classification. In: ICCV (2009)
  35.Fe-Fei, L., Fergus, R., Perona, P.: A bayesian approach to unsupervised one-shot learning of object categories. In: ICCV (2003)
  36.Ramirez, I., Sprechmann, P., Sapiro, G.: Classification and clustering via dictionary learning with structured incoherence and shared features. In: CVPR (2010)
  37.Farquhar, J., Szedmak, S., Meng, H., Shawe-Taylor, J.: Improving bag-of-keypoints image categorisation: Generative models and pdf-kernels. Technical report, University of Southampton (2005)
  38.Müller, K.R., Mika, S., Rätsch, G., Tsuda, K., Schölkopf, B.: An introduction to kernel-based learning algorithms. T-NN 12(2), 181–201 (2001)
  39.Bergamo, A., Torresani, L.: Exploiting weakly-labeled web images to improve object classification: a domain adaptation approach. In: NIPS (2010)
  40.Hoffman, J., Rodner, E., Donahue, J., Darrell, T., Saenko, K.: Efficient learning of domain-invariant image representations. In: ICLR (2013)
  41.Uijlings, J.R., van de Sande, K.E., Gevers, T., Smeulders, A.W.: Selective search for object recognition. IJCV 104(2), 154–171 (2013)CrossRef
  42.Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: NIPS, pp. 1097–1105 (2012)
  43.Jia, Y., Shelhamer, E., Donahue, J., Karayev, S., Long, J., Girshick, R., Guadarrama, S., Darrell, T.: Caffe: Convolutional architecture for fast feature embedding (2014). arXiv preprint arXiv:​1408.​5093
  44.Cheng, Z., Qin, L., Ye, Y., Huang, Q., Tian, Q.: Human daily action analysis with multi-view and color-depth data. In: Fusiello, A., Murino, V., Cucchiara, R. (eds.) ECCV 2012. LNCS, vol. 7584, pp. 52–61. Springer, Heidelberg (2012). doi:10.​1007/​978-3-642-33868-7_​6
  
• 作者单位：Li Niu (17)
  Jianfei Cai (18)
  Dong Xu (19)
  
  17. Interdisciplinary Graduate School, Nanyang Technological University, Singapore, Singapore
  18. School of Computer Engineering, Nanyang Technological University, Singapore, Singapore
  19. School of Electrical and Information Engineering, University of Sydney, Sydney, Australia
  
• 丛书名：Computer Vision ¨C ECCV 2016
• ISBN：978-3-319-46466-4
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349
• 卷排序：9910

文摘

In this paper, we consider Fisher vector in the context of domain adaptation, which has rarely been discussed by the existing domain adaptation methods. Particularly, in many real scenarios, the distributions of Fisher vectors of the training samples (i.e., source domain) and test samples (i.e., target domain) are considerably different, which may degrade the classification performance on the target domain by using the classifiers/regressors learnt based on the training samples from the source domain. To address the domain shift issue, we propose a Domain Adaptive Fisher Vector (DAFV) method, which learns a transformation matrix to select the domain invariant components of Fisher vectors and simultaneously solves a regression problem for visual recognition tasks based on the transformed features. Specifically, we employ a group lasso based regularizer on the transformation matrix to select the components of Fisher vectors, and use a regularizer based on the Maximum Mean Discrepancy (MMD) criterion to reduce the data distribution mismatch of transformed features between the source domain and the target domain. Comprehensive experiments demonstrate the effectiveness of our DAFV method on two benchmark datasets.