3D Reconstruction from few Silhouettes using Statistical Models and Landmark Points

3D model reconstruction is an important issue in computer graphics, computer vision and virtual reality. This paper presents a new method in rapid reconstruction of detailed 3D surface models from a small number, e. g. 3-4, of 2D silhouettes of objects taken from different views. A statistical shape model is used to fit a set of landmarks points, which are automatically created from the 2D silhouettes, to reconstruct a full detailed surface. The landmark points are optimally distributed to guarantee that the salient features of objects are included in the reconstruction process. Some proof-of-concept computational experiments have been conducted to reconstruct a test object. The results show that the proposed method is capable of reconstructing an acceptable detailed 3D surface.