This paper explores some existing methods commonly used in three dimensional (3D) model generation, in particular those which typically employ multiple cameras or multiple frames from one camera for image acquisition. A novel method of model generation is described and evaluated. This method captures multiple images from different viewpoints with a stereo camera to create separate three dimensional meshes, which are then joined to form one single continuous mesh which represents the captured object in 3D. The results from the method provide a good base for 3D models, and although the quality of the results may not be as high as other more expensive methods, they can be achieved at low cost and with minimal set up time and calibration.