A research project is described on the development of an active computer vision system for automatically producing three-dimensional models of physical objects. The objective is to meet the increasing demand for such models for reverse engineering and computer graphics purposes. The system under development uses a camera mounted on a gantry robot arm to view an object placed on a turntable from any desired angle or position. An occluding contour technique is described which is employed by the system to generate three-dimensional data. Examples are presented of the type of model that the system can produce and current work that will further enhance the performance of the system is reviewed.
ZhaoC. S.MohrR.Global three-dimensional surface reconstruction from occluding contours. J. Computer Vision and Image Understanding, 1996, 64(1), 62–96.
2.
BoyerE.BergerM. O.3D surface reconstruction using occluding contours. Int. J. Computer Vision, 1997, 22(3), 219–233.
3.
SzeliskiR.WeissR.Robust shape recovery from occluding contours using a linear smoother. Int. J. Computer Vision, 1998, 28(1), 27–44.
4.
TsaiR. Y.A versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV cameras and lenses. IEEE J. Robotics Automn, 1987, 3(4), 323–344.
5.
HeikkilJ.SilvnO.Four-step camera calibration procedure with implicit image correction. In Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, San Juan, Puerto Rico, 1997, pp. 1106–1112.
