Method, apparatus and stream of formatting an immersive video for legacy and immersive rendering devices
The present disclosure relates to methods, apparatus or systems for formatting of backward compatible immersive video streams. At least one legacy rectangular video is captured from an immersive video obtained from a source ( 82 ). A set of camera control data are used to determine which parts of the immersive video will constitute legacy videos ( 84 ). These part are removed from the immersive video ( 83 ) and every prepared videos are packaged in a stream ( 85 ). The structure from the stream is a container. Information data about the location and size of removed parts may be added in the stream.
1. A method of generating a stream from an immersive video prepared for a convex mapping surface, the method comprising:
generating at least one first video, wherein a frame of a first video is a part of a given frame of the immersive video and is rectified to map a flat rectangular mapping surface;
generating a second video, wherein the second video is the immersive video in which parts corresponding to the at least one first video are replaced by a default content, wherein a shape of the parts depends on the convex mapping surface; and
generating the stream comprising a first syntax element relative to the at least one first video and a second syntax element relative to the second video.
2. The method of claim 1 , the stream further comprising a third syntax element comprising metadata representative of a description of the parts corresponding to the at least one first video.
3. The method of claim 2 , wherein the metadata change over time.
4. The method of claim 2 , wherein generating the at least one first video further comprises reorienting the mapping surface according to the metadata.
5. A device configured to generate a stream from an immersive video prepared for a convex mapping surface, the device comprising:
a video generator configured to generate at least one first video, wherein a frame of a first video is a part of a frame of the immersive video and is rectified to map a flat rectangular mapping surface;
a video generator configured to generate a second video from the immersive video in which parts corresponding to the at least one first video are replaced by a default content, wherein a shape of the parts depends on the convex mapping surface; and
a stream generator configured to generate the stream comprising a first syntax element relative to the at least one first video and at a second syntax element relative to the second video.
6. The device of claim 5 , wherein the stream generator is further configured to add a third syntax element into the stream, the third syntax element relative to metadata representative of a description of the parts corresponding to the at least one first video.
7. The device of claim 6 , wherein the metadata change over time.
8. A non-transitory computer storage medium having stored instructions, when executed by a processor, to perform a method of generating a stream from an immersive video prepared for a convex mapping surface, the method comprising:
generating at least one first video, wherein a frame of a first video is a part of a given frame of the immersive video and is rectified to map a flat rectangular mapping surface;
generating a second video, wherein the second video is the immersive video in which parts corresponding to the at least one first video are replaced by a default content, wherein a shape of the parts depends on the convex mapping surface; and
generating the stream comprising metadata comprising:
a first syntax element relative to at least one first video, wherein a frame of a first video is a part of a frame of an immersive video prepared for a convex mapping surface, wherein the frame is rectified to map a flat rectangular surface; and
a second syntax element relative to a second video, wherein the second video is the immersive video in which parts corresponding to the at least one first video are replaced by a default content, wherein the shape of the parts depends on the convex mapping surface.
9. The non-transitory computer storage medium of claim 8 , wherein the metadata further comprises a third syntax element relative to metadata representative of a description of the parts corresponding to the at least one first video.
10. A method of composing a video for a rendering device from a stream carrying metadata representative of an immersive video prepared for a convex mapping surface, the method comprising:
obtaining, from a first syntax element of the stream, at least one first video, wherein a frame of a first video is a part of a frame of the immersive video and is rectified to map a flat rectangular mapping surface;
obtaining a second video from a second syntax element of the stream, wherein the second video is the immersive video from which the parts corresponding to the at least one first video are replaced by a default content, wherein a shape of the parts depends on the convex mapping surface; and
composing the video by overlaying the at least one first video on the second video.
11. The method of claim 10 , wherein the stream further comprises a third syntax element comprising metadata representative of a description of the parts corresponding to the at least one first video.
12. The method of claim 11 , wherein the composing of the video comprises distorting, scaling, and translating each first video according to the description of the parts corresponding to the at least one first video for overlaying on the second video.
13. A device configured to compose a video from a stream carrying metadata representative of an immersive video prepared for a convex mapping surface for a rendering device, the device comprising a memory associated with a processor configured to:
obtain, from a first syntax element of the stream, at least one first video, wherein a frame of a first video is a part of a frame of the immersive video and is rectified to map a flat rectangular surface; and
obtain a second video from a second syntax element of the stream, wherein the second video is the immersive video from which parts corresponding to the at least one first video are replaced by a default content, wherein a shape of the parts depends on the convex mapping surface; and
compose the video by overlaying each of the at least one first video on the second video.
14. The device of claim 13 , wherein the stream further comprises a third syntax element comprising metadata representative of a description of the parts corresponding to the at least one first video.
15. The device of claim 14 , wherein the composing of the video comprises distorting, scaling, and translating each first video according to the description of the parts corresponding to the at least one first video for overlaying on the second video.