Extended view morphing
OwnerMatthew H. Klapmaninventor
Life of this patent
Estimating the term…
Abstract
A method and apparatus are provided for synthesizing an image from a plurality of video streams. The method includes the steps of determining an orientation of a first and second video source of the plurality of video streams with respect to a video object lying within a field of view of both the first and second video sources and selecting an orientation of a synthesized video source with respect to the video object. The method further includes the step of interpolating among corresponding images of the first and second video sources to form a stream of perspective images of the video object from the selected orientation of the synthesized source.
Claims19
A method of synthesizing at least one image from a plurality of video streams, such method comprising the steps of:
- determining an orientation of at least two video sources of the plurality of video streams with respect to each other, the at least two video sources having a video object lying within a field of view of the at least two video sources;
- selecting a location of a synthesized video source with respect to the video object;
- and
- interpolating among corresponding synchronized frames from the at least two video sources to synthesize the at least one image of the video object as viewed from the selected location of the synthesized source, wherein the at least one image is synthesized without using a three-dimensional model of the video object.
Apparatus for synthesizing at least one image from a plurality of video streams, such apparatus comprising:
- means for determining an orientation of at least two video sources of the plurality of video streams with respect to each other, the at least two video sources having a video object lying within a field of view of the at least two video sources;
- means for selecting a location of a synthesized video source with respect to the video object;
- and
- means for interpolating among corresponding synchronized frames from the at least two video sources to synthesize the at least one image of the video object as viewed from the selected location of the synthesized source, wherein the at least one image is synthesized without using a three dimensional model of the video object.
Apparatus for synthesizing at least one image from a plurality of video streams, such apparatus comprising:
- at least two orientation sensors, located respectively at at least two videos sources, adapted to determine an orientation of the at least two video sources of the plurality of video streams with respect to each other, the at least two video sources having a video object lying within a field of view of the at least two video sources;
- a view point controller adapted to select a location of a synthesized video source with respect to the video object;
- and
- a video processor adapted to interpolate among corresponding synchronized frames received from the at least two video sources to synthesize the at least one image of the video object as viewed form the selected location of the synthesized source, wherein the at least one image is synthesized without using a three-dimensional model of the video object.
Dependencies are inferred from the claim text.