| 摘要 |
<p>To allow for a larger selection ofviewpoints for the viewer during the reproduction of 3D image contents, it is proposed to compute the two stereoscopic images from the image signals ofrecording cameras (101, 102, 103), which record images of an object (1) to be reproduced from differentviewpoints of a monoscopic position. The stereoscopic images are computed with an image viewpoint, which correspond to the viewing angle of a hypothetical viewer (400) in the central axis of the monitor screen (304). A change of the actual viewing position with regard to the central axis is signalized to a processing unit (301), which computes the two stereoscopic images with image viewpoints changed accordingly.</p> |
