Method and apparatus for decoding a 3D video

1. A method for decoding a volumetric video, the method comprising:

receiving a data stream representative of information for selecting, according to a rendering viewpoint, at least one atlas comprising color and depth data patches associated with a viewpoint in said volumetric video, said color and depth data patches being generated with respect to depth and color reference data acquired from a reference viewpoint in said volumetric video, wherein said information comprises a list of adaptive parallax parameters comprising:

a list of reference viewpoints comprising at least one reference viewpoint, and for each of said reference viewpoints:

associated coordinates of said at least one reference viewpoint in a volumetric cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said reference viewpoint;

for each of said reference viewpoints in the list, a number of virtual cameras, each of said virtual cameras corresponding to a viewpoint in said volumetric video; and

for each of said virtual cameras:

associated coordinates of said viewpoint corresponding to said virtual camera in said 3D cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said virtual camera.

2. The method of claim 1 , further comprising:

requesting from a server, using said information:

at least one data stream representative of a reference view comprising said color and depth reference data, and

at least one data stream representative of at least one selected atlas; and

decoding said at least one data stream representative of said reference view and said at least one data stream representative of said at least one selected atlas.

3. The method of claim 1 , wherein selecting the at least one atlas according to a rendering viewpoint of said volumetric video, comprises:

determining a network bandwidth available for requesting a next temporal segment of said volumetric video; and

determining, according to said information and according to said network bandwidth, a reference view and the at least one atlas associated with a viewpoint corresponding to a virtual camera for rendering said next temporal segment of said volumetric video.

4. The method of claim 1 , wherein selecting the at least one atlas according to a rendering viewpoint of said volumetric video uses at least one of:

predicting at least one future rendering viewpoint; and

determining regions of interest for which parallax is improved with a transmission of a corresponding atlas.

5. The method of claim 1 , wherein said adaptive parallax parameters further comprise at least one parameter belonging to the group:

a parameter quantifying a size of said encoded data stream representative of an atlas associated with a virtual camera or a reference viewpoint;

a group of parameters defining a projection associated with a virtual camera to be used for 2D to 3D conversion; and

a list of at least one zone of interest for which parallax is improved with a transmission of a corresponding atlas.

6. The method of claim 1 , wherein said information further comprises for at least one of said adaptive parallax parameters, an indication of a time when said at least one adaptive parallax parameter may change during rendering of said volumetric video.

7. The method of claim 6 , wherein said at least one adaptative parallax parameter which may change corresponds to the number of virtual cameras associated to a reference viewpoint.

8. The method of claim 1 , wherein said adaptative parallax parameters comprises, for at least one of said virtual cameras and/or one of said at least reference viewpoint, a plurality of uniform resource identifiers, each of said uniform resource identifiers being associated to a bit-rate level.

9. The method of claim 1 , further comprising:

receiving a data stream representative of updating the information, wherein said updating the information comprises at least one value for updating at least one of said adaptative parallax parameters.

10. A method for encoding a volumetric video, comprising:

encoding at least one data stream representative of at least one reference view comprising color and depth reference data acquired from a reference viewpoint in the volumetric video;

encoding at least one data stream representative of at least one atlas comprising color and depth data patches acquired from a viewpoint in said volumetric video, said color and depth data patches being generated with respect to depth and color reference data; and

encoding a data stream comprising information for selecting the at least one atlas, according to a rendering viewpoint in said volumetric video, wherein said information comprises a list of adaptive parallax parameters comprising:

a list of reference viewpoints comprising at least one reference viewpoint, and for each of said reference viewpoints:

associated coordinates of said at least one reference viewpoint in a volumetric cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said reference viewpoint;

for each of said reference viewpoints in the list, a number of virtual cameras, each of said virtual cameras corresponding to a viewpoint in said volumetric video; and

for each of said virtual cameras:

associated coordinates of said viewpoint corresponding to said virtual camera in said 3D cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said virtual camera.

11. The method of claim 10 , wherein said adaptive parallax parameters further comprise at least one parameter belonging to the group:

a parameter quantifying a size of said encoded data stream representative of an atlas associated with a virtual camera or a reference viewpoint;

a group of parameters defining a projection associated with a virtual camera to be used for 2D to 3D conversion; and

a list of at least one zone of interest for which parallax is improved with a transmission of a corresponding atlas.

12. The method of claim 10 , wherein said information further comprises for at least one of said adaptive parallax parameters, an indication of a time when said at least one adaptive parallax parameter may change during rendering of said volumetric video.

13. The method of claim 12 , wherein said at least one adaptative parallax parameter which may change corresponds to the number of virtual cameras associated to a reference viewpoint.

14. The method of claim 10 , wherein said adaptative parallax parameters comprises, for at least one of said virtual cameras and/or one of said at least reference viewpoint, a plurality of uniform resource identifiers, each of said uniform resource identifiers being associated to a bit-rate level.

15. The method of claim 10 , further comprising receiving a data stream representative of updating information, wherein said updating information comprises at least one value for updating at least one of said adaptative parallax parameters.

16. A non-transitory storage medium storing information for selecting, according to rendering viewpoint of a volumetric video, at least one atlas comprising color and depth data patches acquired from a viewpoint in said volumetric video, said color and depth data patches being generated with respect to depth and color reference data acquired from a reference viewpoint in said volumetric video, wherein said information comprises a list of adaptive parallax parameters comprising:

a list of reference viewpoints comprising at least one reference viewpoint, and for each of said reference viewpoints:

associated coordinates of said at least one reference viewpoint in a volumetric cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said reference viewpoint;

for each of said reference viewpoints in the list, a number of virtual cameras, each of said virtual cameras corresponding to a viewpoint in said volumetric video; and

for each of said virtual cameras:

associated coordinates of said viewpoint corresponding to said virtual camera in said 3D cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said virtual camera.

17. The non-transitory storage medium of claim 16 , wherein said adaptive parallax parameters further comprise at least one parameter belonging to the group:

a parameter quantifying a size of said encoded data stream representative of an atlas associated with a virtual camera or a reference viewpoint;

a group of parameters defining a projection associated with a virtual camera to be used for 2D to 3D conversion; and

a list of at least one zone of interest for which parallax is improved with a transmission of a corresponding atlas.

18. The non-transitory storage medium of claim 16 , wherein said information further comprises for at least one of said adaptive parallax parameters, an indication of a time when said at least one adaptive parallax parameter may change during rendering of said volumetric video.

19. The non-transitory storage medium of claim 18 , wherein said at least one adaptative parallax parameter which may change corresponds to the number of virtual cameras associated to a reference viewpoint.

20. The non-transitory storage medium of claim 16 , wherein said adaptative parallax parameters comprises, for at least one of said virtual cameras and/or one of said at least reference viewpoint, a plurality of uniform resource identifiers, each of said uniform resource identifiers being associated to a bit-rate level.

21. An apparatus for decoding a volumetric video, the apparatus comprising a processor configured for:

receiving a data stream comprising information for selecting, according to a rendering viewpoint, at least one atlas comprising color and depth data patches acquired from a viewpoint in said volumetric video, said color and depth data patches being generated with respect to depth and color reference data acquired from a reference viewpoint in said volumetric video, wherein said information comprises a list of adaptive parallax parameters comprising:

a list of reference viewpoints comprising at least one reference viewpoint, and for each of said reference viewpoints:

associated coordinates of said at least one reference viewpoint in a volumetric cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said reference viewpoint;

for each of said reference viewpoints in the list, a number of virtual cameras, each of said virtual cameras corresponding to a viewpoint in said volumetric video; and

for each of said virtual cameras:

associated coordinates of said viewpoint corresponding to said virtual camera in said 3D cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said virtual camera.

22. The apparatus of claim 21 , further comprising:

means for requesting from a server, using said information:

at least one data stream representative of a reference view comprising said color and depth reference data, and

at least one data stream representative of the at least one selected atlas; and

means for decoding said at least one data stream representative of said reference view and said at least one data stream representative of said at least one selected atlas.

23. An apparatus for encoding a volumetric video, the apparatus comprising a processor configured for:

encoding at least one data stream representative of at least one reference view comprising color and depth reference data acquired from a reference viewpoint in said volumetric video;

encoding at least one data stream representative of at least one atlas comprising color and depth data patches acquired from a viewpoint in said volumetric video, said color and depth data patches being generated with respect to depth and color reference data acquired from said reference viewpoint in said volumetric video; and

encoding a data stream comprising information for selecting at least one atlas, according to a rendering viewpoint of said volumetric video, wherein said information comprises a list of adaptive parallax parameters comprising:

a list of reference viewpoints comprising at least one reference viewpoint, and for each of said reference viewpoints:

associated coordinates of said at least one reference viewpoint in a volumetric cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said reference viewpoint;

for each of said reference viewpoints in the list, a number of virtual cameras, each of said virtual cameras corresponding to a viewpoint in said volumetric video; and

for each of said virtual cameras:

associated coordinates of said viewpoint corresponding to said virtual camera in said 3D cartesian coordinates system, and

at least one uniform resource identifier allowing to download an encoded data stream representative of an atlas associated with said virtual camera.