Automatic reconstruction of stationary 3-D objects from multiple uncalibrated camera views

A system for the automatic reconstruction of real-world objects from multip le uncalibrated camera views is presented. The camera position and orientat ion for all views, the 3-D shape of the rigid object, as well as the associ ated color information, are recovered from the image sequence. The system p roceeds in four steps. First, the internal camera parameters describing the imaging geometry are calibrated using a reference object, Second, an initi al 3-D description of the object is computed from two views, This model inf ormation is then used in a third step to estimate the camera positions for all available views using a no, el linear 3-D motion and shape estimation a lgorithm. The main feature of this third step is the simultaneous estimatio n of 3-D camera-motion parameters and object shape refinement with respect to the initial 3-D model. The initial 3-D shape model exhibits only a few d egrees of freedom and the object shape refinement is defined as flexible de formation of the initial shape model. Our formulation of the shape deformat ion allows the object texture to slide on the surface, which differs from t raditional flexible body modeling. This novel combined shape and motion est imation using sliding texture considerably improves the calibration data of the individual views in comparison to fixed-shape model-based camera-motio n estimation, Since the shape model used for model-based camera-motion esti mation is only approximate, a volumetric 3-D reconstruction process is init iated in the fourth step that combines the information from all views simul taneously. The recovered object consists of a set of voxels with associated color information that describes even fine structures and details of the o bject, New views of the object can be rendered from the recovered 3-D model , which has potential applications in virtual reality or multimedia systems and the emerging field of video coding using 3-D scene models.