A Trajectory Basis Selection Method for Non-Rigid Structure from Motion

Ya Ming Wang; Jun Bao Zheng; Ming Feng Jiang; Ying Li Xiong; Wen Qing Huang

doi:10.4028/www.scientific.net/AMM.644-650.1396

Paper Titles

Three-Dimensional Inversion and Visual Expression of Two-Dimensional Data in High Density Resistivity Method
p.1377

Creep Prediction From Stress Relaxation Coupled With Equivalent Relaxation Rate
p.1382

Have I Ever been to this Place? A SIFT Based Method for UAV
p.1386

Research on Crowd Evacuation Simulation in Complex Mountainous Terrain Area Based on Cellular Automata
p.1391

A Trajectory Basis Selection Method for Non-Rigid Structure from Motion
p.1396

Usability Evaluation of Eye Tracker-Based Smart Phone APP GUI Design
p.1400

Classification of Apple Varieties Using FT-NIR Spectroscopy and Possibilistic Learning Vector Quantization
p.1405

Study on the Time-Grating Sensor Method of the Static Light Field Type Based on Timing-Driven
p.1409

Research on Data Parallel Synchronous Transmission Technology Based Virtual Radio
p.1414

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 644-650A Trajectory Basis Selection Method for Non-Rigid...

A Trajectory Basis Selection Method for Non-Rigid Structure from Motion

Abstract:

The key point of trajectory basis based Non-Rigid Structure from Motion (NRSFM) is how to tune the basis size and how to choose the best basis combination. If the basis size is too small, the trajectory is poorly represented by the basis. But too large basis size makes the system more ill-conditioned and the reconstruction error becomes unbounded. In this paper, an automatic method is proposed to select the trajectory basis, which can select appropriate basis size. The experiment results show that, compared with the empirical value from a lot of repeated experiments, the proposed method can improve both reconstruction accuracy and efficiency of NRSFM.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 644-650)

Pages:

1396-1399

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.644-650.1396

Citation:

Cite this paper

Online since:

September 2014

Authors:

Ya Ming Wang*, Jun Bao Zheng, Ming Feng Jiang, Ying Li Xiong, Wen Qing Huang

Keywords:

Basis Size, Selection Method, Trajectory Basis

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] I. Akhter, Y. Sheikh, S. Khan, and T. Kanade. Nonrigid Structure from Motion in Trajectory Space, Proc. Neural Information Processing Systems, (2008).

Google Scholar

[2] I. Akhter, Y. Sheikh, S. Khan, and T. Kanade. Trajectory Space: A Dual Representation for Nonrigid Structure from Motion. IEEE Computer Society, (2010).

DOI: 10.1109/tpami.2010.201

Google Scholar

[3] Y. Zhu, M. Cox, and S. Lucey. 3D motion reconstruction for real-world camera motion. In CVPR. IEEE, June (2011).

DOI: 10.1109/cvpr.2011.5995650

Google Scholar

[4] J. Valmadre, S Lucey. General Trajectory Prior for Non-Rigid Reconstruction. Computer Vision and Pattern, (2012).

DOI: 10.1109/cvpr.2012.6247826

Google Scholar

[5] M. Salzmann and R. Urtasun. Physically-based motion models for 3D tracking: A convex formulation. In ICCV, pages 2064–2071. IEEE, Nov. (2011).

DOI: 10.1109/iccv.2011.6126480

Google Scholar

[6] P. F. U. Gotardo and A. M. Martinez. Non-rigid structure from motion with complementary rank-3 spaces. In CVPR, pages 3065–3072. IEEE, June (2011).

DOI: 10.1109/cvpr.2011.5995560

Google Scholar

[7] H. Zhou, X. Li, A. Sadka, Nonrigid structure from motion from 2d images using markov chain monte carlo, IEEE Trans. Pattern Anal. Mach. Intell. 14, (2012).

DOI: 10.1109/tmm.2011.2170406

Google Scholar

[8] Y. Dai, H. Li, M. He. A Simple Prior-free Method for Non-Rigid Structure-from-Motion Factorization. Computer Vision and Pattern Recognition, (2012).

DOI: 10.1109/cvpr.2012.6247905

Google Scholar