This paper presents a metric based on information theory principles that compares 3D object models to images. The metric is based on the formulation of the mutual information between the model and the images. The technique does not require a priori information about the surface properties of the object and is robust with respect to variations of illumination. As result the method is quite general and may be used in a wide variety of applications.

Experiments are presented that demonstrate the approach by evaluating reconstructed 3D building models in aerial images.