System and method for generating three dimensional presentations

摘要

A virtual three-dimensional presentation in accordance with the present invention includes three stereoscopic layers, a stereoscopic background layer, a stereoscopic video insert layer, and a stereoscopic three-dimensional foreground layer. A surface, the Z-sphere, is defined in the virtual three-dimensional space of the three-dimensional scene with reference to the aim point, the distance between the cameras, and the distance from the camera of the video scene, of a virtual stereoscopic camera looking at the three-dimensional scene being created. The separation of the virtual stereoscopic camera lens and distance to the aim point are set based on the stereoscopic cameras that are creating the video insert layer. The stereoscopic three-dimensional foreground layer scene may include a key which defines the transparency of objects in the foreground layer. The stereoscopic video insert layer may be generated by positioning a presenter, or other object, in a chroma-key set which is captured with a stereoscopic camera. The stereoscopic video insert layer may be distorted before being composited with the background and foreground such that elements of the stereoscopic video appear to move across a three-dimensional floor of the stereoscopic three-dimensional scene.

主权项

1. A computer-implemented method for generating a virtual video presentation, comprising the steps of:
(a) defining a surface within a three-dimensional space of a virtual three dimensional scene to be rendered, the surface dividing the virtual three-dimensional scene into a virtual three-dimensional background scene to be rendered and a virtual three-dimensional foreground scene to be rendered; (b) rendering the virtual three-dimensional background scene to be located behind the surface within the virtual video presentation; (c) rendering the virtual three-dimensional foreground scene to be located in front of the surface within the virtual video presentation; (d) generating two stereoscopic insert videos and key signals using at least two virtual video cameras defining an aim point that is positioned within the virtual three-dimensional background scene, wherein the aim point is not on the surface dividing the virtual three-dimensional scene into the virtual three-dimensional background scene to be rendered and the virtual three-dimensional foreground scene to be rendered; and (e) compositing the stereoscopic insert videos and key signals, the rendered virtual three-dimensional background scene, and the rendered virtual three-dimensional foreground scene together to form the virtual video presentation, wherein each of the stereoscopic insert videos appears in front of the three-dimensional background scene and the three dimensional foreground scene appears in front of each of the stereoscopic insert videos and the three-dimensional background scene.