Obtain the depth of field based on two-view

this paper attempts to propose a method that uses images taken in different locations by a single camera or from a video stream interception to restore depth information of the scene. At first, the SIFT algorithm is used to process two images, to find the feature points and match them into pairs, then these pairs are used to calculate the corresponding fundamental matrix. From the fundamental matrix the epipolar can be got. The feature points are matched with the epipolar constraint to improve accuracy. With these matched points, the binocular stereo vision method is used to find the depth information. For a visual representation, the coordinates of the points are input to Matlab to show the result. Experiments show that the method can get information of depth of the field without a known position relationship between images.