Human Action Recognition Based On Ridgelet Transform And Svm

• Main Authors: A. Ouanane, A. Serir
• Format: Text
• Language: English
• Published: Zenodo 2013
• Subjects: Human action; Ridgelet Transform; PCA; K-means; SVM.; laptev
• Online Access: https://dx.doi.org/10.5281/zenodo.1055730; https://zenodo.org/record/1055730

In this paper, a novel algorithm based on Ridgelet Transform and support vector machine is proposed for human action recognition. The Ridgelet transform is a directional multi-resolution transform and it is more suitable for describing the human action by performing its directional information to form spatial features vectors. The dynamic transition between the spatial features is carried out using both the Principal Component Analysis and clustering algorithm K-means. First, the Principal Component Analysis is used to reduce the dimensionality of the obtained vectors. Then, the kmeans algorithm is then used to perform the obtained vectors to form the spatio-temporal pattern, called set-of-labels, according to given periodicity of human action. Finally, a Support Machine classifier is used to discriminate between the different human actions. Different tests are conducted on popular Datasets, such as Weizmann and KTH. The obtained results show that the proposed method provides more significant accuracy rate and it drives more robustness in very challenging situations such as lighting changes, scaling and dynamic environment : {"references": ["J. C. Niebles and F.Li. \"A hierarchical model of shape and appearance\nfor human action classification,\" in Proc. IEEE Conf. Comput. Vis.\nPattern Recog., Jun.17-22,pp. 1-8. 2007.", "C.Rougier, J.Meunier, A. St-Arnaud, and J.Rousseau. \"Fall detection\nfrom human shape and motion history using video surveillance,\" in\nProc.21st Int. Conf. Adv. Inf. Netw. Appl. Workshops, pp. 875-880.\n2007.", "S. Ali and M. Shah. \"Human Action Recognition in Videos Using\nKinematic Features and Multiple Instance Learning\", IEEE Transactions\non Pattern Analysis and Machine Intelligence (PAMI), Volume 32, Issue\n2, pp: 288-303, 2010.", "E.B. Ermis, V. Saligrama, P. Jodoin and J. Konrad. \"Motion\nsegmentation and abnormal behavior detection via behavior clustering,\"\nin Proc. IEEE Int. Conf. Image Process., Oct. 12-15, pp. 769-772.2008.", "J. Li, Q. Pan, H. Zhang, P. Cui. \"Image recognition using Radon\ntransform,\" Intelligent Transportation Systems,. Proceedings. IEEE,\nvol.1, no., pp. 741- 744. 2003.", "M. Singh, M. Mandal, A. Basu. \"Pose recognition using the Radon\ntransform\". In: 48th Midwest Symposium on Circuits and Systems, pp.\n1091-1094. 2005.", "Y. Wang, K. Huang, T. Tan. \"Human activity recognition based on R\ntransform\". In: IEEE Conference on Computer Vision and Pattern\nRecognition, CVPR, pp. 1-8. 2007.", "A. Ouanane and A. Serir. \"Fingerprint Compression by Ridgelet\nTransform\", IEEE ISSPIT, 16-19 D\u00e9cembre 2008 Sarajevo", "Z. Zhang, H. Yu, J. Zhang, X. Zhang. \"Digital image watermark\nembedding and blind extracting in the ridgelet domain\", Journal of\nCommunication and Computer, vol.3, No.5.pp.1-7, may 2006.\n[10] Candes, E. \" Ridgelets: theory and applications,\" Ph.D. thesis,\nDepartment of Statistics, Stanford University; 1998.\n[11] J. L. Starck, E. J. Cand\u00e8s and D. L. Donoho. The curvelet transform for\nimage denoising. IEEE Transactions on Image Processing, vol.11, pp.\n670-684. 2000\n[12] L. Sirovich and M. Kirby. \"Low dimensional procedure for the\ncharacterization of human faces\". Journal of the Optical Society of\nAmerica. A, Optics, Image Science, and Vision, vol 4(3), pp.519-524.\n1987.\n[13] I.T.Jolliffe. \" Principal component analysis\". Springer, New York. 2002.\n[14] Y. Wang, H. Jiang, M. Drew, Z. Li, G.Mori. \"Unsupervised discovery of\naction classes\". IEEE Computer Vision and Pattern Recognition. vol.2,\nno., pp.1654-1661, 2006.\n[15] V.N. Vapnik. \"The Nature of Statistical Learning Theory\". New York:\nSpringer-Verlag,1995.\n[16] M.Blank, L.Gorelick, E.Shechtman, M.Irani and R. Basri. \"Actions as\nspace-time shapes,\" in Proc. IEEE Int. Conf. Comput. Vision, pp. 1395-\n1402. 2005.\n[17] C. Schuldt, I. Laptev, and B.I Caputo. \"Recognizing human actions: a\nlocal SVM approach\". ICPR (17), pp. 32-36, 2004.\n[18] C.-W. Hsu and C.-J. Lin. \"A comparison of methods for multi-class\nsupport vector machines,\" IEEE Trans. Neural Netw. , vol. 13, no. 2, pp.\n415-425, Mar. 2002.\n[19] C.C. Chang and C.J Lin. \"LIBSVM: A Library for Support Vector\nMachines\", 2001.http://www.csie.ntu.edu.tw/~cjlin/libsvm.\n[20] J. Liu and M. Shah. \"Learning Human Actions via Information\nMaximization,\" Proc. IEEE Conf. Computer Vision and Pattern\nRecognition, 2008.\n[21] I. Laptev.I, M. Marsza\u00b6\u00c7\u00e2\u00ed, C.Schmid and B. Rozenfeld. \"Learning realistic\nhuman actions from movies\". CVPR, pp 1-8, 2008.\n[22] K. Schindler and L.V. Gool. \"Action snippets: how many frames does\nhuman action recognition require\". In: CVPR, pp. 1-8, 2008."]}