Single loop decoding based interlayer prediction
View Patent ↗Methods, systems, and instrumentalities are described to implement single loop decoding, A video decoding device (VDD) may receive a video bitstream, which may include one or more bitstreams (e.g., base layer bitstream, enhanced layer bitstream, etc.) The VDD may generate a hybrid inter-layer reference (H-ILR) picture. The H-ILR picture may be generated based on one or more inputs including, for example, a first layer motion information, a processed (e.g., upsampled) first layer residue signal, a second layer texture information, etc. The first layer motion information may be derived from a compressed first layer motion field or an uncompressed first layer motion field. An H-ILR block of the H-ILR picture may be reconstructed by deriving motion information of the H-ILR block from one or more neighboring H-ILR blocks in the H-ILR picture, for example, when the neighboring blocks are inter-coded.
1. A decoding method comprising:
receiving a video bitstream comprising a first layer and a second layer;
processing first-layer residue information and first-layer texture information associated with the first layer of the video bitstream;
generating a hybrid inter-layer reference (H-ILR) picture based on first-layer motion information, the processed first-layer residue information, and second-layer texture information, wherein when a first H-ILR block of the H-ILR picture corresponds to an intra-coded first layer block, the first H-ILR block is reconstructed based on the processed first-layer texture information associated with the first layer, wherein a second H-ILR block is reconstructed based on the processed first-layer residue information, the first-layer motion information, and the second-layer texture information associated with a second-layer reference picture; and
storing the generated H-ILR picture in a second layer decoded picture buffer (DPB).
2. The method of claim 1 , wherein the first layer comprises a base layer and the second layer comprises an enhancement layer.
3. The method of claim 1 , wherein the first-layer motion information comprises at least one of a variable indicating one or more reference picture lists on which a prediction of the first-layer motion information is based, one or more motion vectors, or one or more reference picture indices.
4. The method of claim 1 , wherein the first-layer motion information is derived from at least one of a compressed first-layer motion field or an uncompressed first-layer motion field.
5. The method of claim 1 , wherein the second-layer reference picture is associated with one or more H-ILR blocks in the H-ILR picture.
6. The method of claim 1 , further comprising:
partially reconstructing the first layer; and
decoding the second layer based on the partially reconstructed first layer.
7. The method of claim 1 , wherein the H-ILR picture is generated based on a weighted prediction based on at least one of second-layer motion compensated prediction information or the processed first-layer residue information.
8. The method of claim 1 , further comprising:
parsing a first layer bitstream;
de-quantizing a first layer residual by applying an inverse quantization to the first layer bitstream; and
applying an inverse transform to the de-quantized first layer residual.
9. The method of claim 5 , wherein the second H-ILR block of the H-ILR picture is reconstructed by:
deriving motion information for the second H-ILR block from the first-layer motion information associated with one or more neighboring H-ILR blocks in the H-ILR picture, wherein the one or more neighboring H-ILR blocks are inter-coded; and
performing motion compensation for the second H-ILR block based on the derived motion information for the second H-ILR block from the first-layer motion information, the corresponding second-layer texture information associated with the second-layer reference picture, and the corresponding processed first-layer residue information.
10. A single loop decoding method comprising:
receiving a video bitstream comprising a first layer and a second layer;
processing first-layer residue information and first-layer texture information associated with the first layer of the video bitstream;
generating a hybrid inter-layer reference (H-ILR) picture based on first-layer motion information, the processed first-layer residue information, and second-layer texture information, wherein when a first H-ILR block of the H-ILR picture corresponds to an intra-coded first layer block, the first H-ILR block is reconstructed based on the processed first-layer texture information associated with the first layer, wherein a second H-ILR block is reconstructed based on the processed first-layer residue information, the first-layer motion information, and the second-layer texture information associated with a second-layer reference picture; and
storing the generated H-ILR picture in a second layer decoded picture buffer (DPB).
11. A decoder comprising:
a processor; and
a memory comprising instructions that, when executed by the processor, cause the decoder to
receive a video bitstream comprising a first layer and a second layer;
process first-layer residue information and first-layer texture information associated with the first layer of the video bitstream;
generate a hybrid inter-layer reference (H-ILR) picture based on first-layer motion information, the processed first-layer residue information, and second-layer texture information, wherein when a first H-ILR block of the H-ILR picture corresponds to an intra-coded first layer block, the first H-ILR block is reconstructed based on the processed first-layer texture information associated with the first layer, wherein a second H-ILR block is reconstructed based on the processed first-layer residue information, the first-layer motion information, and the second-layer texture information associated with a second-layer reference picture; and
store the generated H-ILR picture in a second layer decoded picture buffer (DPB).
12. The decoder of claim 11 , wherein the first layer comprises a base layer and the second layer comprises an enhancement layer.
13. The decoder of claim 11 , wherein the first-layer motion information comprises at least one of a variable indicating one or more reference picture lists on which a prediction of the first-layer motion information is based, one or more motion vectors, or one or more reference picture indices.
14. The decoder of claim 11 , wherein the first-layer motion information is derived from at least one of a compressed first-layer motion field or an uncompressed first-layer motion field.
15. The decoder of claim 11 , wherein the second-layer reference picture is associated with one or more H-ILR blocks in the H-ILR picture.
16. The decoder of claim 11 , the memory storing further instructions for:
partially reconstructing the first layer; and
decoding the second layer based on the partially reconstructed first layer.
17. The decoder of claim 11 , wherein the H-ILR picture is generated based on a weighted prediction based on at least one of second-layer motion compensated prediction information or the processed first-layer residue information.
18. The decoder of claim 11 , the memory storing further instructions for:
parsing a first layer bitstream;
de-quantizing a first layer residual by applying an inverse quantization to the first layer bitstream; and
applying an inverse transform to the de-quantized first layer residual.
19. The decoder of claim 15 , wherein the memory storing further instructions to reconstruct the second H-ILR block of the H-ILR picture by:
deriving motion information for the second H-ILR block from the first-layer motion information associated with one or more neighboring H-ILR blocks in the H-ILR picture, wherein the one or more neighboring H-ILR blocks are inter-coded; and
performing motion compensation for the second H-ILR block based on the derived motion information for the second H-ILR block from the first-layer motion information, the corresponding second-layer texture information associated with the second-layer reference picture, and the corresponding processed first-layer residue information.
20. A single loop decoder comprising:
a processor; and
a memory comprising instructions that, when executed by the processor, cause the single loop decoder to
receive a video bitstream comprising a first layer and a second layer;
process first-layer residue information and first-layer texture information associated with the first layer of the video bitstream;
generate a hybrid inter-layer reference (H-ILR) picture based on first-layer motion information, the processed first-layer residue information, and second-layer texture information, wherein when a first H-ILR block of the H-ILR picture corresponds to an intra-coded first layer block, the first H-ILR block is reconstructed based on the processed first-layer texture information associated with the first layer, wherein a second H-ILR block is reconstructed based on the processed first-layer residue information, the first-layer motion information, and the second-layer texture information associated with a second-layer reference picture; and
store the generated H-ILR picture in a second layer decoded picture buffer (DPB).