Methods and systems for chroma residual data prediction
Several methods and systems for chroma residual data prediction for encoding blocks corresponding to video data are disclosed. In an embodiment, at least one coefficient correlating reconstructed luma residual samples and corresponding reconstructed chroma residual samples is computed for one or more encoded blocks of video data. Predicted chroma residual samples are generated for encoding a block of video data based on corresponding reconstructed luma residual samples and the at least one coefficient.
1. A method, comprising:
obtaining, by a processor, a correlation between a reconstructed luma residual sample and each of a C b component and a C r component corresponding to a reconstructed chroma residual sample;
determining, by the processor, a coefficient correlating the reconstructed luma residual sample and the reconstructed chroma residual sample for an encoded block of video data based on the correlation;
determining, by the processor, a predicted chroma residual sample for inter-prediction encoding a block of the video data based on a corresponding reconstructed luma residual sample and the coefficient; and
selecting, by the processor, a chroma residual prediction encoding mode or a chroma prediction encoding mode based on a cost associated with chroma residual samples in the encoded block.
2. The method of claim 1 , wherein:
the encoded block of the video data corresponds to a neighboring encoded block of a currently encoded block of the video data associated with the encoded block of the video data.
3. The method of claim 1 , wherein:
the encoded block of the video data corresponds to one of a top neighboring block and a left neighboring block neighboring a currently encoded block of the video data associated with the encoded block of the video data.
4. The method of claim 1 , wherein:
determining the coefficient is based on a least mean square error approximation corresponding to a linear model.
5. The method of claim 1 , wherein:
the obtaining of the correlation between the reconstructed luma residual sample and each of the C b component and the C r component corresponding to the reconstructed chroma residual sample includes:
determining a first and second coefficient for the C b component based on least mean square error approximations; and
determining a third and fourth coefficient for the C r component based on least mean square error approximations.
6. The method of claim 1 , further comprising determining a weighted sum of the reconstructed luma residual sample and the reconstructed chroma residual sample.
7. The method of claim 1 , wherein:
determining the coefficient is based on a presence of at least one block of the video data encoded based on an inter-prediction coding scheme neighboring the block of the video data and comprising motion vectors in correlation with motion vectors corresponding to the block of the video data.
8. The method of claim 1 , further comprising determining a subsampling of the reconstructed luma residual sample corresponding to the encoded block of video data prior to determining the coefficient in response to a number of reconstructed luma residual samples being greater than a number of reconstructed chroma residual samples.
9. The method of claim 8 , wherein:
determining the subsampling is performed based on a chroma subsampling metric.
10. A system, comprising:
a processor configured to:
obtain a correlation between a reconstructed luma residual sample and each of a C b component and a C r component corresponding to a reconstructed chroma residual sample;
determine a coefficient correlating the reconstructed luma residual sample and the reconstructed chroma residual sample for an encoded block of video data based on the correlation;
determine a predicted chroma residual sample for inter-prediction encoding a block of the video data based on a corresponding reconstructed luma residual sample and the coefficient; and
select a chroma residual prediction encoding mode or a chroma prediction encoding mode based on a cost associated with chroma residual samples in the encoded block; and
an encoding unit configured to inter-prediction encode the block of the video data based on the selection.
11. The system of claim 10 , wherein:
the encoded block of the video data corresponds to a neighboring encoded block of a currently encoded block of the video data associated with the encoded block of the video data.
12. The system of claim 10 , wherein:
the encoded block of the video data corresponds to one of a top neighboring block and a left neighboring block neighboring a currently encoded block of the video data associated with the encoded block of the video data.
13. The system of claim 10 , wherein:
the processor is configured to determine the coefficient based on a least mean square error approximation corresponding to a linear model.
14. The system of claim 10 , wherein:
obtaining the correlation between the reconstructed luma residual sample and each of the C b component and the C r component corresponding to the reconstructed chroma residual sample includes:
determining a first and second coefficient for the C b component based on least mean square error approximations; and
determining a third and fourth coefficient for the C r component based on least mean square error approximations.
15. The system of claim 10 , wherein:
the processor is configured to determine a weighted sum of the reconstructed luma residual sample and the reconstructed chroma residual sample.
16. The system of claim 10 , wherein:
the processor is configured to determine the coefficient based on a presence of at least one block of the video data encoded based on an inter-prediction coding scheme neighboring the block of the video data and comprising motion vectors in correlation with motion vectors corresponding to the block of the video data.
17. The system of claim 10 , wherein:
the processor is configured to determine a subsampling of the reconstructed luma residual sample corresponding to the encoded block of video data prior to determining the coefficient in response to a number of reconstructed luma residual samples being greater than a number of reconstructed chroma residual samples.
18. The system of claim 17 , wherein:
the processor is configured to perform the subsampling based on a chroma subsampling metric.
19. A system, comprising:
a processor configured to:
determine a coefficient correlating a reconstructed luma residual sample and a reconstructed chroma residual sample for an encoded block of video data; and
determine a predicted chroma residual sample for inter-prediction encoding a block of the video data based on the corresponding reconstructed luma residual sample and the coefficient; and
select a chroma residual prediction encoding mode or a chroma prediction encoding mode based on a cost associated with chroma residual samples in the encoded block; and
an encoding unit configured to inter-prediction encode the block of the video data based on the selection.
20. The system of claim 19 ,
wherein to determine the coefficient, the processor is configured to determine a first coefficient and a second coefficient for a C b component based on least mean square error approximations, and
wherein the processor is configured to determine the predicted chroma residual sample for inter-prediction encoding the block of the video data based on the corresponding reconstructed luma residual sample, further based on the first coefficient, and further based on the second coefficient.