| 摘要 |
<p>Video signals representing a real scene are processed to combine a computer generated object with the video signals with the effect that the object appears within the scene. The scene includes a first real object upon which a virtual model of the real object is mapped (s100) so that the virtual model substantially corresponds with the real object. The virtual model has alterable transparency such that it can be rendered as substantially transparent within the scene. Regions upon the virtual model which are determined to be occluded from a virtual light source by the computer generated object are detected (s105). The occluded regions of the virtual model thereafter have their transparency modified (s110), such that when the scene is rendered (s115), the modified transparency regions appear as shadows upon the object.</p> |
