In the conventional structured light approach to 3D object surface reconstruction, it is well known that finding the correspondences between the camera rays and light rays is a difficult problem. In this research we present a novel approach that does not directly use the ray correspondences to derive the surface points as the intersection of the corresponding camera rays and light rays. Instead, we use a virtual linear grid line originating at a grid junction on the object surface being considered. The length and the orientation of the virtual linear grid line in the 3D space are first calculated. Then the 3D positions of the grid junction can be determined. Detailed derivations and experimental results using the real object data are reported
Published in:
Pattern Recognition, 1996., Proceedings of the 13th International Conference on
(Volume:1
)
Date of Conference: 25-29 Aug 1996