METHOD FOR LOCATING A CAMERA AND FOR 3D RECONSTRUCTION IN A PARTIALLY KNOWN ENVIRONMENT

摘要

The invention relates to a method for locating a camera and for the 3D reconstruction of the static environment in which said camera is located, said environment comprising an object of interest, the 3D model of which is known, said method including the following steps: a) calculating an initial exposure of the camera in the environment and an initial reconstruction; b) calculating the exposure of the camera for each new image by pairing 3D primitives of the environment with 2D primitives of said image, and reconstructing 3D primitives of the environment by triangulation; and c) simultaneously optimizing the exposures of the camera and the 3D primitives by minimizing a reprojection error over a plurality of images. The 3D model is a geometric description of the object of interest, the reprojection error has only two types of terms, i.e. a first type of terms associated with primitives constrained by the 3D model, and a second type of terms associated with primitives of the environment other than the object, and the optimization includes a sub-step of associating the primitives with the environment or with the 3D model.