3D Surface Reconstruction from Single 2D Image

Abstract:

The image-based 3D reconstruction technique is an important method of 3D data passive acquired in computer vision. Aimed at the limitation of traditional SFS based 3D surface reconstruction method, a combined image-based 3D surface reconstruction method is given in this paper. Firstly, the pretreatment method of gradation is applied to convert the input color image to 256 gray image; secondly, shape from shading method is applied to recover the 3D profile of object; finally, the B-spline technique is adopted to reconstruct the model of surface and improve the resolution of surface. The key techniques such as the principle and algorithm of B-spline interpolation reconstruction are analyzed, and the experimental results are also given. The single-image based 3D reconstruction method given in this paper will supply a helpful condition for the extensive used of SFS technique.

• Periodical: Key Engineering Materials (Volume 522)
• Pages: 886-890
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.522.886
• Online since: August 2012
• Authors: Feng He Wu, Jin Fen Wang, Jun Wang, Yu Bo Ren
• Keywords: 3D Reconstruction, B-Spline, Image, Resolution, SFS

References

[1] S. Dimitris and M. Dimitris: IEEE. PAMI, Vol. 25(2003), pp.247-264.

[2] K. Ikeuchi and B.K.P. Horn: Artificial Intelligence, Vol. 17(1981), pp.141-184.

[3] C.H. Lee and A. Rosenfeld: Artificial Intelligence, Vol. 26 (1985), pp.125-143.

[4] A.P. Pentland: Second Int. Conf. on Computer Vision (Los Alamitos, USA, December 5-8, 1988). pp.404-413.

[5] P. S Tsai and M Shah: Image and Vision Computing, Vol. 12(1994) No. 8, pp.487-498.

[6] F.H. Wu, F.Z. Shi and Z.Q. Hu: Chinese Mechanical Engineering Vol. 13(2004), p.1163~1166 (In Chinese).