自動追蹤影片中人體姿勢變化不但具挑戰性，且有許多實際應用需求。在真實場景中，由於動作複雜、遮蔽、光影變化等等，要追蹤人體姿勢更極具挑戰性。因此，我們提出線上學習方法，使用潛在結構自動向量機（Latent structured SVM）來解決此問題。首先，我們初始化潛在的身體部位，接著經過四個步驟訓練潛在結構自動向量機模型（Latent structured SVM）。追蹤影片中的連續影像過程中，此模型會持續更新。此外，為解決遮蔽問題，我們使用Prize-Collecting Steiner Tree (PCST) 演算法來偵測身體部位。實驗結果證明此方法優於其他最新的人體姿勢追蹤方法。

Human pose tracking in a video is a challenging problem and a desirable requirement in many applications. The problem is challenging in realistic scenes due to complicated movement, occlusion, a lighting change, and etc. We propose an online learning approach for tracking human pose using latent structured SVM. Firstly, we initialize body and latent parts, then we train the model by using a four-stage training process of latent structured SVM. We update the model for each image sequence of video during tracking process. To solve the problem of occlusion, we use body part detection by Prize-Collecting Steiner Tree algorithm (PCST). The experimental results veri ed that our proposed method outperforms the other state-of-the-art human pose approaches.

[1] T. Lim, S. Hong, B. Han, and J. Hee Han, "Joint segmentation and pose
tracking of human in natural videos," in The IEEE International Conference
on Computer Vision (ICCV), December 2013.
[2] C. Huang, E. Boyer, N. Navab, and S. Ilic, "Human shape and pose tracking using keyframes," in 2014 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014, Columbus, OH, USA, June 23-28, 2014, pp. 3446-3453, IEEE, 2014.
[3] D. Park and D. Ramanan, "N-best maximal decoders for part models.," in
ICCV, pp. 2627{2634, IEEE, 2011.
[4] V. Ramakrishna, T. Kanade, and Y. Sheikh, "Tracking human pose by tracking symmetric parts," in CVPR'13, pp. 3728{3735, 2013.
[5] J. Tian, L. Li, and W. Liu, \Multi-scale human pose tracking in 2D monocular images," Journal of Computer and Communications, vol. 02, no. 02, p. 78 84, 2014.
[6] Y. Yang and D. Ramanan, "Articulated pose estimation with flexible mixtures-of-parts," in CVPR, 2011.
[7] C.-H. Huang, E. Boyer, and S. Ilic, "Robust human body shape and pose
tracking," in 3DV'13, pp. 287{294, 2013.
[8] Y. Tian, C. L. Zitnick, and S. G. Narasimhan, "Exploring the spatial hierarchy of mixture models for human pose estimation.," in ECCV (5), vol. 7576 of Lecture Notes in Computer Science, pp. 256{269, Springer, 2012.
[9] C. Ionescu, F. Li, and C. Sminchisescu, "Latent structured models for human pose estimation," in IEEE International Conference on Computer Vision,
ICCV 2011, Barcelona, Spain, November 6-13, 2011, pp. 2220{2227, IEEE,
2011.
[10] R. Yao, Q. Shi, C. Shen, Y. Zhang, and A. van den Hengel, "Part-based visual tracking with online latent structural learning," in IEEE Conference on Computer Vision and Pattern Recognition (CVPR'13), (Oregon, USA), 2013.
[11] P. F. Felzenszwalb, R. B. Girshick, D. McAllester, and D. Ramanan, "Object detection with discriminatively trained part-based models," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, no. 9, pp. 1627-1645, 2010.
[12] C.-N. J. Yu and T. Joachims, "Learning structural svms with latent variables," in Proceedings of the 26th Annual International Conference on Machine Learning, ICML '09, (New York, NY, USA), pp. 1169-1176, ACM, 2009.
[13] L. Zhu, Y. Chen, A. L. Yuille, and W. T. Freeman, "Latent hierarchical structural learning for object detection.," in CVPR, pp. 1062{1069, IEEE, 2010.
[14] S. Hare, A. Sa ari, and P. H. S. Torr, "Struck: Structured output tracking with kernels.," in ICCV, pp. 263{270, IEEE, 2011.
[15] M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, "The PASCAL Visual Object Classes (VOC) challenge," International Journal of Computer Vision, vol. 88, no. 2, pp. 303-338, 2010.
[16] P. Viola and M. Jones, \Rapid object detection using a boosted cascade of simple features," in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 1, pp. 511{518, Dec. 2001.
[17] N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," in International Conference on Computer Vision & Pattern Recognition, vol. 2, pp. 886-893, June 2005.
[18] S. Vijayanarasimhan and K. Grauman, "Effi cient region search for object detection.," in CVPR, pp. 1401-1408, IEEE, 2011.
[19] Z. Li, X.-M.Wu, and S.-F. Chang, "Segmentation using superpixels: A bipartite
graph partitioning approach.," in CVPR, pp. 789-796, IEEE, 2012.
[20] I. Ljubic, R. Weiskircher, U. Pferschy, G. W. Klau, P. Mutzel, and M. Fischetti, "An algorithmic framework for the exact solution of the prize-collecting steiner tree problem," Math. Program., vol. 105, no. 2-3, pp. 427-449, 2006.
[21] X. Wang, C. Ning, A. Shi, and G. Lv, "An improved similarity measure in particle lters for robust object tracking," in Image and Signal Processing (CISP), 2013 6th International Congress on, vol. 01, pp. 46-50, 2013.