In a method for 3D digitization of an object (1), a plurality of camera images of the object are recorded and assembled to determine the 3D coordinates of the object. To improve such method, pictures are taken from the object (1), from which 2D feature points (11, 12, 13; 21, 22, 23) of the object (1) are determined. The 3D coordinates of the 2D feature points are determined. The 2D point correspondences (32, 32, 22) between the 2D feature points of a picture and the 2D feature points of another picture are determined. Several of these 2D point correspondences are selected, and an associated 3D transformation is determined. The quality of this 3D transformation is determined with reference to the transformed 3D coordinates of the 2D feature points. Valid 3D feature points are determined therefrom. For assembling the camera images of the object (1), the 3D coordinates of the valid 3D feature points are used.