Abstract :
Establishing correspondence between moving objects is a key problem in multi-camera surveillance, and Field of View (FOV) line is an efficient tool to resolve the consistency in labeling objects. In this paper, we get the camera field of view lines by using Scale-Invariant Features Transform (SIFT) algorithm and homograph, without knowing the camera calibration information. First, by using SIFT algorithm, matching points are automatically generated between two images sharing a joint region; we chose more than four pairs of matching points which are coplanar in space. These points are then utilized to compute the homography matrix of the two images. Finally, the camera FOV lines are obtained by using the homography matrix and boundary points of the images. Experimental results show the accuracy and robustness of our method.
