| 摘要 |
A method for inserting a target image into images of an image stream of a real site taken by a camera (108), the method comprising: receiving a stream of original images taken by a camera (108) of a real, three-dimensional site; placing at least one target image into a three-dimensional model of the site on at least one target area (502, 504); for each original image (400) in the image stream, rendering the three-dimensional model using data on a position and pointing direction of the camera (108) in the real site to generate a mask (900) and a rendered image containing said at least one target image (604); determining a background area (400a) in the original image (400) separate from the at least one target area (502, 504) using the mask (900); and combining the background area (400a) of the original image (400), and the at least one target image (604) into an output image (400e). |
