US Patent 7236622 Method for forming a depth image

A computer vision/image processing method generates a depth map. In one embodiment, the present invention provides a method of forming a relative depth map from at least three x-ray images of a patient taken from different perspectives. The method includes the steps of: (a) selecting one of the x-ray images as a reference image; (b) identifying corresponding objects within the reference image and the other images; (c) computing a disparity between the identified corresponding objects; and (d) generating the relative depth map from the disparities.