Processing a point cloud
At least one embodiment relates to methods for encoding, signaling and decoding a 3D point cloud with different layers associated with Point Local Reconstruction information and modes. A layer may be associated with its own information and modes or associated metadata may comprise indexes to information or modes encoded or decoded in relation with a different layer.
1. A method for decoding a 3D point cloud, the method comprising:
decoding a first layered image and associated first metadata from a data stream, said first metadata comprising a first information indicating whether said first layered image is to be decoded according to a point local reconstruction method; and
on condition that said first information is enabled:
retrieving first point local reconstruction information and first point local reconstruction modes in said first metadata, and
applying said first point local reconstruction modes to said first layered image according to said first point local reconstruction information to reconstruct said 3D point cloud; and
decoding a second layered image and associated second metadata from said data stream, said second metadata comprising a second information indicating whether said second layer image is to be decoded according to a point local reconstruction method and according to an index that points to said first point local reconstruction information; and
on condition that said second information is enabled:
retrieving second point local reconstruction modes in said second metadata, and
applying said second point local reconstruction modes to said second layered image according to said first point local reconstruction information to reconstruct said 3D point cloud.
2. The method of claim 1 , further comprising:
decoding a third layered image and associated third metadata from said data stream, said third metadata comprising a third information indicating whether said third layer image is to be decoded according to a point local reconstruction method and according to an index that points to said first or to said second point local reconstruction modes; and
on condition that said third information is enabled:
retrieve third point local reconstruction information in said third metadata, and
applying said first or said second point local reconstruction modes to said third layered image according to said third point local reconstruction information to reconstruct said 3D point cloud.
3. The method of claim 2 , further comprising:
decoding a fourth layered image and associated fourth metadata from said data stream, said fourth metadata comprising a fourth information indicating whether said fourth layer image is to be decoded according to a point local reconstruction method, according to an index that points to said first or to said second or to said third point local reconstruction information and according to an index that points to said first or to said second or to said third point local reconstruction modes; and
on condition that said fourth information is enabled, applying said first or said second or said third point local reconstruction modes to said fourth layered image according to said first or said second or said third point local reconstruction information to reconstruct said 3D point cloud.
4. A device for decoding a 3D point cloud, the device comprising a processor configured for:
decoding a first layered image and associated first metadata from a data stream, said first metadata comprising a first information indicating whether said first layered image is to be decoded according to a point local reconstruction method; and
on condition that said first information is enabled:
retrieving first point local reconstruction information and first point local reconstruction modes in said first metadata, and
applying said first point local reconstruction modes to said first layered image according to said first point local reconstruction information to reconstruct said 3D point cloud and decoding a second layered image and associated second metadata from said data stream, said second metadata comprising a second information indicating whether said second layer image is to be decoded according to a point local reconstruction method and according to an index that points to said first point local reconstruction information; and
on condition that said second information is enabled:
retrieve second point local reconstruction modes in said second metadata, and
applying said second point local reconstruction modes to said second layered image according to said first point local reconstruction information to reconstruct said 3D point cloud.
5. The device of claim 4 , said processor further configured for:
decoding a third layered image and associated third metadata from said data stream, said third metadata comprising a third information indicating whether said third layer image is to be decoded according to a point local reconstruction method and according to an index that points to said first or to said second point local reconstruction modes; and
on condition that said third information is enabled:
retrieve third point local reconstruction information in said third metadata, and
applying said first or said second point local reconstruction modes to said third layered image according to said third point local reconstruction information to reconstruct said 3D point cloud.
6. The device of claim 5 , said processor further configured for:
decoding a fourth layered image and associated fourth metadata from said data stream, said fourth metadata comprising a fourth information indicating whether said fourth layer image is to be decoded according to a point local reconstruction method, and according to an index that points to said first or to said second or to said third point local reconstruction information and according to an index that points to said first or to said second or to said third point local reconstruction modes; and
on condition that said fourth information is enabled, applying said first or said second or said third point local reconstruction modes to said fourth layered image according to said first or said second or said third point local reconstruction information to reconstruct said 3D point cloud.
7. A method for encoding a 3D point cloud, the method comprising:
encoding a first layered image and associated first metadata in a data stream, said metadata comprising:
a first information indicating whether said first layered image is to be decoded according to a point local reconstruction method; and
on condition that said first information is enabled, a first point local reconstruction information and first point local reconstruction modes; and
encoding a second layered image and associated second metadata in said data stream, said second metadata comprising:
a second information indicating whether said second laver image is to be decoded according to a point local reconstruction method;
an index that points to said first point local reconstruction information; and
on condition that said second information is enabled, second point local reconstruction modes.
8. The method of claim 7 , further comprising:
encoding a third layered image and associated third metadata in said data stream, said third metadata comprising:
a third information indicating whether said third layer image is to be decoded according to a point local reconstruction method;
an index that points to said first or to said second point local reconstruction modes; and
on condition that said third information is enabled, a third point local reconstruction information.
9. The method of claim 8 , further comprising:
encoding a fourth layered image and associated fourth metadata in said data stream, said fourth metadata comprising:
a fourth information indicating whether said fourth layer image is to be decoded according to a point local reconstruction method;
an index that points to said first or to said second or to said third point local reconstruction information; and
an index that points to said first or to said second or to said third point local reconstruction modes.
10. A non-transitory computer readable medium containing data content generated according to the method of claim 7 , for playback using a processor.
11. A device for encoding a 3D point cloud, the device comprising a processor configured for:
encoding a first layered image and associated first metadata in a data stream, said metadata comprising:
a first information indicating whether said first layered image is to be decoded according to a point local reconstruction method; and
on condition that said first information is enabled, a first point local reconstruction information and first point local reconstruction modes; and
encoding a second layered image and associated second metadata in said data stream, said second metadata comprising:
a second information indicating whether said second layer image is to be decoded according to a point local reconstruction method;
an index that points to said first point local reconstruction information; and
on condition that said second information is enabled, second point local reconstruction modes.
12. The device of claim 11 , said processor further configured for:
encoding a third layered image and associated third metadata in said data stream, said third metadata comprising:
a third information indicating whether said third layer image is to be decoded according to a point local reconstruction method;
an index that points to said first or to said second point local reconstruction modes; and
on condition that said third information is enabled, a third point local reconstruction information.
13. The device of claim 12 , said processor further configured for:
encoding a fourth layered image and associated fourth metadata in said data stream, said fourth metadata comprising:
a fourth information indicating whether said fourth layer image is to be decoded according to a point local reconstruction method;
an index that points to said first or to said second or to said third point local reconstruction information; and
an index that points to said first or to said second or to said third point local reconstruction modes.