A method including obtaining x-ray data for a multi-component object; processing the x-ray data to obtain at least first and second 3D representations of the object with respective first and second resolutions, the first resolution being higher than the second resolution; identifying a plurality of regions of the second 3D representation, each region corresponding to one of a number of components of the object, by at least identifying a number of initial regions of the second 3D representation, each initial region having pixel values in one of a plurality of ranges of pixel values, and adjusting each of the number of initial regions based on a comparison between the initial region and features derived from 2D sections of the object from the first 3D representation; and obtaining a 3D model.