A technique for reducing the time required before a tracking process is performed again in calculating the position and orientation of an image capturing apparatus or an object. Images captured by the apparatus are acquired, the acquired images are stored in order per frame, straight lines detected in the captured images and straight lines in a three-dimensional space model are matched, the position and orientation of the apparatus are initialized by obtaining correspondences having the highest consistency, a captured image to be processed is selected from the captured images, the change in position and orientation of the apparatus is measured using the change in position and orientation of a target object in the selected image based on the initialized position and orientation of the apparatus, and if the measured change in position and orientation is not within a predetermined threshold, initialization is performed using the selected image.