An apparatus and method for displaying a three-dimensional virtual endoscopic image are provided. In the method, information on a virtual endoscopic image is input in the form of volume data expressed as a function of three-dimensional position. A two-dimensional reference image, a three-dimensional reference image, and a virtual endoscopic image are detected from the volume data. The detected images are displayed on one screen. Virtual cameras are respectively displayed on areas, in which the two and three-dimensional reference images are respectively displayed. Here, a camera display sphere and a camera display circle are defined on the basis of a current position of each virtual camera. When information regarding to one image, among the two and three-dimensional reference images and the virtual endoscopic image on one screen, is changed by a user's operation, information regarding to the other images is changed based on the information changed by the user's operation.