3D B-spline curve construction from orthogonal views with self-overlapping projection segments

Constructing B-spline curves from orthogonal views is a common operation in the process of car modelling. Existing reconstructing methods for 3D free-form curves cannot effectively handle the situation where the projection curves in the input views are self-overlapping, which appears frequently in practice. In this paper we focus on solving the problem above. We track the input views simultaneously and generate an ordered set of sampling space points. Then we reorganize the planar points in each view according to the sampling points. Weight values are assigned to all points, and the points around the overlapping segments are split into multiple ones. Then the resulting curve can be evaluated through minimizing the constructed energy function. Examples show the efficiency of our method. Graphical abstractDisplay Omitted HighlightsWe propose a stepwise tracking method to generate ordered 3D points from two input orthogonal views.We propose a method to construct 3D B-spline curves from planar views through minimizing the energy function with weight values.The method can efficiently detect the situation of self-overlapping for curve segments from input views.

[1]  Les A. Piegl,et al.  The NURBS Book , 1995, Monographs in Visual Communication.

[2]  Allen Y. Yang,et al.  Symmetry-based 3-D reconstruction from perspective images , 2005, Comput. Vis. Image Underst..

[3]  Yi Jun Xiao,et al.  Optimized stereo reconstruction of free-form space curves based on a nonuniform rational B-spline model. , 2005, Journal of the Optical Society of America. A, Optics, image science, and vision.

[4]  Anders Heyden,et al.  Using conic correspondences in two images to estimate the epipolar geometry , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[5]  Yuandi Zhao,et al.  Two-view curve reconstruction based on the snake model , 2011, J. Comput. Appl. Math..

[6]  Fredrik Kahl,et al.  Multiview reconstruction of space curves , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[7]  Sukhendu Das,et al.  Reconstruction of Quadratic Curves in 3-D from Two or More Perspective Views , 2002 .

[8]  Helmut Pottmann,et al.  Fitting B-spline curves to point clouds by curvature-based squared distance minimization , 2006, TOGS.

[9]  Adrien Bartoli,et al.  Reconstruction of 3D Curves for Quality Control , 2007, SCIA.

[10]  Balasubramanian Raman,et al.  RECONSTRUCTION OF CUBIC CURVES FROM TWO OR MORE IMAGES USING GEOMETRIC INTERSECTION , 2009 .

[11]  Sanjeev Kumar,et al.  Reconstruction of free-form space curves using NURBS-snakes and a quadratic programming approach , 2015, Comput. Aided Geom. Des..

[12]  Amnon Shashua,et al.  Multiple View Geometry of General Algebraic Curves , 2004, International Journal of Computer Vision.

[13]  Balasubramanian Raman,et al.  Error estimation in reconstruction of quadratic curves in 3D space , 2007, Int. J. Comput. Math..