3D Human Pose Estimation from a Monocular Image Using Model Fitting in Eigenspaces
Geli Bo, Katsunori Onishi, Tetsuya Takiguchi, Yasuo Ariki
.
DOI: 10.4236/jsea.2010.311125   PDF    HTML     4,690 Downloads   8,888 Views  

Abstract

Generally, there are two approaches for solving the problem of human pose estimation from monocular images. One is the learning-based approach, and the other is the model-based approach. The former method can estimate the poses rapidly but has the disadvantage of low estimation accuracy. While the latter method is able to accurately estimate the poses, its computational cost is high. In this paper, we propose a method to integrate the learning-based and model-based approaches to improve the estimation precision. In the learning-based approach, we use regression analysis to model the mapping from visual observations to human poses. In the model-based approach, a particle filter is employed on the results of regression analysis. To solve the curse of the dimensionality problem, the eigenspace of each motion is learned using Principal Component Analysis (PCA). Finally, the proposed method was estimated using the CMU Graphics Lab Motion Capture Database. The RMS error of human joint angles was 6.2 degrees using our method, an improvement of up to 0.9 degrees compared to the method without eigenspaces.

Share and Cite:

G. Bo, K. Onishi, T. Takiguchi and Y. Ariki, "3D Human Pose Estimation from a Monocular Image Using Model Fitting in Eigenspaces," Journal of Software Engineering and Applications, Vol. 3 No. 11, 2010, pp. 1060-1066. doi: 10.4236/jsea.2010.311125.

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1] A. Agarwal and B. Triggs, “3D Human Pose from Silhouettes by Relevance Vector Regression,” IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Vol. 2, 2004, pp. 882-888.
[2] X. Zhao, H. Ning, Y. Liu and T. Huang, “Discriminative Estimation of 3D Human Pose Using Gaussian Processes,” Proceedings of 19th International Conference on Pattern Recognition (ICPR’08), December 2008, pp. 1-4.
[3] C. Sminchisescu, A. Kanaujia and D. N. Metaxas, “ : Discriminative Density Propagation for Visual Tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 29, No. 11, November 2007, pp. 2030-2044.
[4] H. Ning, Y. Hu and T. Huang, “Efficient Initialization of Mixtures of Experts for Human Pose Estimation,” 15th IEEE International Conference on Image Processing (ICIP2008), October 2008, pp. 2164-2167.
[5] M. Lee, I. Cohen, “A Model-Based Approach for Estimating Human 3D Poses in Static Images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 28, No. 6, June 2006, pp. 905-916.
[6] T. Jaeggli, E. Koller-Meier and L. V. Gool, “Learning Generative Models for Monocular Body Pose Estimation,” Proceedings of the 8th Asian Conference on Computer Vision, Vol. 1, 2007.
[7] S. Hou, A. Galata, F. Caillette, N. Thacker and P. Bromiley, “Real-time Body Tracking Using a Gaussian Process Latent Variable Model,” IEEE 11th International Conference on Computer Vision, October 2007, pp. 1-8.
[8] G. Peng, W. Alexander, A. O. Balan and M. J. Black, “Estimating Human Shape and Pose from a Single Image,” IEEE 12th International Conference on Computer Vision (ICCV2009), September 2009, pp. 1381-1388.
[9] N. Dalal and B. Triggs, “Histograms of Oriented Gradients for Human Detection,” IEEE Computer Society Conference on Computer Version and Pattern Recognition (CVPR2005), Vol. 1, June 2005, pp. 886-893.
[10] G. Mori and J. Malik, “Recovering 3D Human Body Configurations using Shape Contexts,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 28, No. 7, 2006, pp. 1052-1062.
[11] K. Onishi, T. Takiguchi and Y. Ariki, “3D Human Posture Estimation Using the HOG Features from Monocular Image,” 19th International Conference on Pattern Recognition (ICPR2008), December 2008, pp. 1-4.
[12] CMU Human Motion Capture Database. Available online: http://mocap.cs.cmu.edu/
[13] A. Fossati, M. Salzmann and P. Fua, “Observable subspaces for 3D human motion recovery,” IEEE Conference on Computer Version and Pattern Recognition (CVPR2009), June 2009, pp. 1137-1144.
[14] M. Isard and A. Blake, “Condensation-Conditional Density Propagation for Visual Tracking,” International Journal of Computer Vision, Vol. 29, No. 1, 1998, pp. 5-28.
[15] J. Deutscher, A. Blake and I. Reid, “Articulated Body Motion Capture by Annealed Particle Filtering,” IEEE Computer Society Conference on Computer Version and Pattern Recognition, Vol. 2, June 2000, pp. 126-133.
[16] L. Ye, Q. Zhang and L. Guan, “Use Hierarchical Genetic Particle Filter to Figure Articulated Human Tracking,” International Conference on Multimedia and Expo (ICME2008), 2008, pp. 1561-1564.
[17] X. Zhao and Y. Liu, “Tracking 3D Human Motion in Compact Base Space,” IEEE Workshop on Applications of Computer Vision (WACV’07), February 2007, p. 39.
[18] H. Sidenbladh, M. Black and D. Fleet, “Stochastic Tracking of 3D Human Figures Using 2D Image Motion,” Computer Vision — ECCV 2000, Vol. 1843, 2000, pp. 702-718.
[19] R. Urtasun, D. Fleet and P. Fua, “Monocular 3D Tracking of the Golf Swing,” IEEE Computer Society Conference on Computer Version and Pattern Recognition (CVPR2005), Vol. 2, June 2005, pp. 932-938.

Copyright © 2024 by authors and Scientific Research Publishing Inc.

Creative Commons License

This work and the related PDF file are licensed under a Creative Commons Attribution 4.0 International License.