Abstract: The present invention provides a target object elimination method based on a panoramic camera, the panoramic camera comprising at least two independent lenses, wherein the method comprises following steps: capturing one basic panoramic image by each of the independent lenses; recognizing a target object from each of the basic panoramic images; dividing each of the basic panoramic images into a plurality of image parts having a same or different size, wherein at least one of the image parts comprises the target object; deleting the one image part comprising the target object in each of the basic panoramic images; and combining all remaining image parts in each of the basic panoramic images to obtain a corrected panoramic image from which the target object is eliminated. By the present invention, the target object can be eliminated quickly and efficiently while ensuring the quality of the image.
Abstract: A method and a system for visual localization based on dual dome cameras is based on two synchronized panoramic video streams output by dual dome cameras to solve the problem of fewer feature points and tracking failure, thereby achieving stable visual SLAM tracking. The depth information of the scene is restored via the two panoramic video streams based on the principle of triangulation measurement. The positions and postures of the dual dome cameras are calculated based on a principle of binocular vision based SLAM, so that accurate map information are obtained finally by evaluating the positions and postures of the dual dome cameras corresponding to the key frames and the depth information in the key frames. The disclosure makes up for inaccurate and incomplete depth information of the scenes in passive scene recoveries, which is suitable for vehicle and robot positioning, obstacle detection and free space estimation.