METHOD AND DEVICE FOR IMAGE PROCESSING

A method for generating a set of respective transformation maps for a set of respective 2D images from a same object and using a parameter-based transformation model, comprises the steps of —receiving said set of respective 2D images and said parameter-based transformation model —detecting matching regions across several pairs of the 2D images, based on the set of 2D images and 3D information of said object, —identifying respective interdependencies of the matching regions over the 2D images, —optimizing the parameters of the parameter-based transformation model over the matching regions of all images as well as over the non-matching regions in all images.