3-D Reconstruction of Image Sequence Based on Independent Three-View

Abstract:

In order to realize efficient and stable 3-D reconstruction of image sequence, a novel 3-D reconstruction method is proposed based on independent three-view in this paper. Firstly, three-view is used as reconstruction unit. Consequently, robust fundamental matrix is calculated under the constraint of three-view. Then 3-D point cloud of 3-D scene corresponding to this three-view is rebuilt by triangulation. Because three-view point cloud is rigid body and local stable, 3-D point clouds of whole image sequence are unified to the same space coordinate system via calculating and using transformation parameters between adjacent three-view. Finally 3-D point cloud reconstruction of whole image sequence is achieved. As shown in experiment, method presented in this paper is simple, efficient and reliable.

• Periodical: Advanced Materials Research (Volumes 989-994)
• Pages: 3844-3850
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.989-994.3844
• Online since: July 2014
• Authors: Cong Li*, Hong Rui Zhao, Gang Fu
• Keywords: 3-D Reconstruction, Computer Vision, Fundamental Matrix, Photogrammetry, Three-View, Transformation Parameters

References

[1] R. Hartley, A. Zisserman: Multiple View Geometry in Computer Vision (Cambridge University Press , UK 2004).

[2] H.T. Shan: The study on object 3D reconstruction theory and kty technology based on image sequences(PLA Information Engineering University, Zhengzhou, 2011).

[3] L. Fang: Research on feature based 3D scene reconstruction techniques from image sequence ( Huazhong University of science and technology, Wuhan 2006).

[4] D.G. Lowe: International Journal of Computer Vision (2004).

[5] Z,Y. Zhang, in: The Proceedings of the Seventh IEEE International Conference on  Computer Vision, edited by IEEE, Kerkyra (1999).

[6] A.I. Carro, R. Morros, in: 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video , edited by IEEE, Zurich (2012).

DOI: 10.1109/3dtv.2012.6365452

[7] Y. Li, S. Velipasalar, and M.C. Gursoy, in: 10th IEEE International Conference on Advanced Video and Signal Based Surveillance, edited by IEEE, Krakow (2013).

DOI: 10.1109/avss.2013.6636644

[8] X.B. Shi, F. Liu, and Y. Wang, in: International Conference on Virtual Reality and Visualization, edited by IEEE, Beijing (2011).

[9] D.B. Calderon, T. Maria, in: 6th Colombian Computing Congress, edted by IEEE, Manizales(2011).

[10] S.S. Brant: Journal of Mathematical Imaging and Vision (2006).

[11] F.X. Chen, R.S. Wang: Journal of computer-aided design & computer graphics (2005).

[12] S. Lu, Y. J. Lei, and W. W. Kong: Control and Decision(2012).

[13] P. J. Wang, Y.M. Xu: Photogrammetry(Wuhan University Press, Wuhan 2010).

[14] N. David : IEEE Transactions on Pattern Analysis and Machine Intelligence (2004).