This paper describes how the generation of 3-D models may be achieved from a small number of images. It uses an automatic segmentation and a two stage iterative camera calibration and volumetric intersection algorithm to produce a set of 3D points in the form of a cloud of voxels. From this a polygonal or B-spline based surface representation of an object can be generated. The B-spline surface representation can be used to implicitly smooth the modelled surface and also assists texture extraction and portability. The technique, which has been developed on a standard work station, is both fast and robust and requires no specialised equipment.