A method for integrating range images with different resolutions for 3-D model construction

This paper proposes a new method for integrating range images with different resolutions to generate a whole surface model of a 3D object. Generally, a modeling method using range images needs an overlapping area between the images for the registration process. Accordingly, there are redundant data in the overlapping area. In order to produce a seamless and nonredundant model, we integrate the data, and represent the overlapping area with triangular patches. However, few conventional methods have paid attention to integrating images with different resolutions. We propose a new integration method in which geometric and topological information are used. Experimental results show that our proposed method is efficient for integrating multiple range images