3D object model construction from multiple range views

We present a complete system for building complex 3D models using multiple range views measured from a fiber-optic projected-fringe interferometry system, and from a structured light range scanner. For the interferometry images, the BLS phase unwrapping algorithm was used to obtain true depth values. A robust registration algorithm is proposed that uses randomly selected control points, and the interval least medium squares error estimator. It is not necessary to triangulate the raw data before the algorithm is applied. Also, a 'holes-and-gaps' filling algorithm is described to improve the display of the integrated 3D models. The experimental results show that the system robustly integrates different views into a single 3D model, and that the holes-and-gaps algorithm improves the visual results.

[1]  Ray Jarvis,et al.  3D free-form surface registration and object recognition , 2004, International Journal of Computer Vision.

[2]  Zhengyou Zhang,et al.  Parameter estimation techniques: a tutorial with application to conic fitting , 1997, Image Vis. Comput..

[3]  Anil K. Jain,et al.  Two-dimensional phase unwrapping using a block least-squares method , 1999, IEEE Trans. Image Process..

[4]  Katsuhiko Sakaue,et al.  Registration and integration of multiple range images for 3-D model construction , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[5]  Gang Wang,et al.  From images to models: automatic 3D object model construction from multiple views , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[6]  Glenn Beheim,et al.  Fiber-optic projected-fringe digital interferometry , 1990 .