parlab research> activities

Research projects

Free-form 3D object digitizing

Christian Schütz, Timothée Jost

Virtual objects

There exists a high demand for digitizing systems which allow to generate a virtual 3D model from a real world 3D object. Applications such as teleshopping, virtual museum, rapid prototyping, reverse engineering and 3D object recognition need a virtual representation of the 3D geometry of a real object. Manual object digitizing is a tedious task and can be replaced by 3D scanners as for example range scanners which provide an accurate and fast way to digitize solid objects. Since only one view of an object can be captured at once, several views have to be combined in order to obtain a description of the complete object surface.

Digitizing system

In this work a digitizing system is developed which captures and triangulates views of a real world 3D object and semi-automatically registers and integrates them into a virtual model. First, an object is placed at different poses and its surfaces are sensed by a range scanner. Then, an interactive 3D environment allows the operator to enter an estimate of the relative pose of the different surfaces which are then aligned automatically. The automatic matching uses the common parts of two views to align them precisely. Finally, the overlapping triangle meshes of the registered surfaces are fused in order to obtain one unique mesh for the entire object.

The presented 3D object digitizing system successfully registers and integrates object views acquired by a range scanner. It combines interactive and automatic modules in order to provide maximal flexibility. Human perception is used for the rough registration of single views. The automatic matching registers the approximately aligned surfaces at high precision. The individual surfaces are easily triangulated using the range image grid. The integration of the different triangle meshes tries to keep a maximum of the triangulated surface mesh and joins them together with a gap filling algorithm.

References

H. Hugli, Ch. Schutz, D. Semitekos, "Geometric matching for free-form 3D object recognition," ACCV, Singapore, Vol 3, pp.819-823, 1995.

Christian Schutz, Emerico Natonek, Charles Baur, Heinz Hugli, "3D Vision in a Virtual Reality Robotics Environment", SPIE Photonics East, Telemanipulator and Telepresence Technologies III, Boston, Proc. SPIE, Vol. 3012, pp. 138-146, Nov. 1996.

Christian Schütz, Heinz Hügli, "Augmented reality using range images", SPIE Photonics West, The Engineering Reality of Virtual Reality, San Jose, Proc. SPIE, Vol. 3012- IV, pp. 472-478, Feb. 1997

C. Schütz, T. Jost and H. Hügli, "Semi-Automatic 3D Object Digitizing System Using Range Images", Proc. ACCV'98, Hong-Kong, pp. 490-497, Jan 1998

Christian Schütz, Timothée Jost, Heinz Hügli, "Free-form 3D object reconstruction from range images", Proc. Int. Conf. Virtual Systems and MultiMedia, VSMM'97, Geneva, IEEE Computer Society, pp. 69-70, Aug. 1997

H. Hügli, "Numériser sans contraintes de positionnement: par assemblage de vues 3D arbitraires; Getting rid of positioning constraints: Modeling from arbitrary 3D scans", Congrès industriel sur la digitalisaton & la création industrielle, Paris (F) , 19-20 mai 1999

hu / 23.04.2008
-