Techniques are disclosed for displaying a graphical element in a manner that simulates three-dimensional (3D) visibility (including parallax and shadowing). More particularly, a number of images, each captured with a known spatial relationship to a target 3D object, may be used to construct a lighting model of the target object. In one embodiment, for example, polynomial texture maps (PTM) using spherical or hemispherical harmonics may be used to do this. Using PTM techniques a relatively small number of basis images may be identified. When the target object is to be displayed, orientation information may be used to generate a combination of the basis images so as to simulate the 3D presentation of the target object.