Deep Human Pose Estimation Method Based on Mixture Articulated Limb Model
LIU Binghan1, 2, LI Zhenda1, 2, KE Xiao1, 2
1.College of Mathematics and Computer Science, Fuzhou University, Fuzhou 350116; 2.Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing, Fuzhou University, Fuzhou 350116
Abstract:A flexible mixture model is proposed to solve the problems of human pose estimation. The model is composed of joint appearance and inner-joint relationship models, and it is trained through a deep convolutional neural network (DCNN). Firstly, a graphical model is constructed to represent joints and limbs of human body. Secondly, images are decomposed into several image blocks centered on the joints and used as training input data. Finally, a multiple classification DCNN network is obtained to perform human pose estimation.The proposed method is more flexible for human body representation, and the detection rate of joint points and the correct detection rate are effectively improved.
刘秉瀚, 李振达, 柯逍. 基于混合关节肢体模型的深度人体姿态估计方法[J]. 模式识别与人工智能, 2019, 32(2): 97-107.
LIU Binghan, LI Zhenda, KE Xiao. Deep Human Pose Estimation Method Based on Mixture Articulated Limb Model. , 2019, 32(2): 97-107.
[1] BARADEL F, WOLF C, MILLE J.Human Action Recognition: Pose-Based Attention Draws Focus to Hands // Proc of the IEEE International Conference on Computer Vision Workshop. Washington, USA: IEEE, 2017: 604-613. [2] VASILEIADIS M, MALASSIOTIS S, GIAKOUMIS D, et al. Robust Human Pose Tracking for Realistic Service Robot Applications // Proc of the IEEE International Conference on Computer Vision Workshop. Washington, USA: IEEE, 2017: 1363-1372. [3] GONG K, LIANG X D, ZHANG D Y, et al. Look into Person: Self-supervised Structure-Sensitive Learning and a New Benchmark for Human Parsing // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2017: 932-940. [4] YANG Y, RAMANAN D.Articulated Pose Estimation with Flexible Mixtures-of-Parts // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2011: 1385-1392. [5] CHEN X J, YUILLE A.Articulated Pose Estimation by a Gra-phical Model with Image Dependent Pairwise Relations // GHAHRAMANI Z, WELLING M, GORTES C, et al., eds. Advances in Neural Information Processing Systems 27. Cambridge, USA: The MIT Press, 2014: 1736-1744. [6] FISCHLER M A, ELSCHLAGER R A.The Representation and Matching of Pictorial Structures. IEEE Transactions on Computers, 1973, 22(1): 67-92. [7] JOHNSON S, EVERINGHAM M.Learning Effective Human Pose Estimation from Inaccurate Annotation // Proc of the 24th IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2011: 1465-1472. [8] TSOCHANTARIDIS I, HOFMANN T, JOACHIMS T, et al. Su-pport Vector Machine Learning for Interdependent and Structured Output Spaces // Proc of the 21st International Conference on Machine Learning. New York, USA: ACM, 2004. DOI:10.1145/1015330.1015341. [9] CHEN X J, YUILLE A L.Parsing Occluded People by Flexible Compositions // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2015: 3945-3954. [10] CHU X, OUYANG W L, LI H S, et al. Structured Feature Lear-ning for Pose Estimation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 4715-4723. [11] FAN X C, ZHENG K, LIN Y W, et al. Combining Local Appea-rance and Holistic View: Dual-Source Deep Neural Networks for Human Pose Estimation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2015: 1347-1355. [12] OUYANG W L, CHU X, WANG X G.Multi-source Deep Lear-ning for Human Pose Estimation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2014: 2337-2344. [13] TOSHEV A, SZEGEDY C.DeepPose: Human Pose Estimation via Deep Neural Networks // Proc of the IEEE Conference on Compu-ter Vision and Pattern Recognition. Washington, USA: IEEE, 2014: 1653-1660. [14] JOHNSON S, EVERINGHAM M. Clustered Pose and Nonlinear Appearance Models for Human Pose Estimation // Proc of the British Machine Vision Conference. Dundee, Britain: BMVA, 2010: 12.1-12.11. [15] SAPP B, TASKAR B.MODEC: Multimodal Decomposable Models for Human Pose Estimation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2013: 3674-3681. [16] RAMANAN D.Learning to Parse Images of Articulated Bodies // SCHÖLKOPF B, PLATT J C, HOFFMAN T, eds. Advances in Neural Information Processing Systems 19. Cambridge, USA: The MIT Press, 2006: 1129-1136. [17] PISHCHULIN L, ANDRILUKA M, GEHLER P, et al. Poselet Conditioned Pictorial Structures // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2013: 588-595. [18] RAMAKRISHNA V, MUNOZ D, HEBERT M, ,et al. Pose Machines: Articulated Pose Estimation via Inference Machines // Proc of the 13th European Conference on Computer Vision. Berlin. Pose Machines: Articulated Pose Estimation via Inference Machines // Proc of the 13th European Conference on Computer Vision. Berlin, Germany: Springer, 2014, II: 33-47. [19] YANG W, OUYANG W L, LI H S, et al. End-to-End Learning of Deformable Mixture of Parts and Deep Convolutional Neural Networks for Human Pose Estimation // Proc of the IEEE Conference on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 3073-3082. [20] TOMPSON J J, JAIN A, LECUN Y, et al.Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation // GHAHRAMANI Z, WELLING M, CORTES C, et al., eds. Advances in Neural Information Processing Systems 27. Cambridge, USA: The MIT Press, 2014: 1799-1807.