Method and apparatus for intra prediction with multiple weighted references
A method of performing intra prediction for encoding or decoding uses multiple layers of reference samples. The layers are formed into reference arrays that are used by a function, such as a weighted combination, to form a final prediction. The prediction is used in encoding or decoding a block of video data. The weights can be determined in a number of ways, and for a given prediction mode, the same weights, or different weights can be used for all pixels in a target block. If the weights are varied, they can depend on the distance of the target pixel from reference arrays.
1. A method, comprising:
obtaining a plurality of reference arrays based on a plurality of reconstructed samples neighboring a block of video data, wherein the plurality of reference arrays include a first reference array located one pixel position from the block of video data and one or more second reference arrays respectively located at pixel positions greater than the one pixel position from the block of video data;
obtaining a plurality of predicted samples for a target pixel of the block of video data respectively from the plurality of reference arrays;
obtaining a final intra prediction for the target pixel of the block of video, the final intra prediction being based on a function of the plurality of predicted samples respectively from the plurality of reference arrays, wherein the function comprises a weighted combination of the plurality of predicted samples, and wherein a plurality of weights of the weighted combination decrease as distances between the target pixel and the respective reference arrays increase; and
encoding the block of video using the final intra prediction.
2. A method, comprising:
obtaining a plurality of reference arrays based on a plurality of decoded samples neighboring a block of video data, wherein the plurality of reference arrays include a first reference array located one pixel position from the block of video data and one or more second reference arrays respectively located at pixel positions greater than the one pixel position from the block of video data;
obtaining a plurality of predicted samples for a target pixel of the block of video data respectively from the plurality of reference arrays;
obtaining a final intra prediction for the target pixel of the block of video, the final intra prediction being based on a function of the plurality of predicted samples respectively from the plurality of reference arrays, wherein the function comprises a weighted combination of the plurality of predicted samples, and wherein a plurality of weights of the weighted combination decrease as distances between the target pixel and the respective reference arrays increase; and
decoding the block of video using the final intra prediction.
3. An apparatus, comprising:
a memory, and
a processor, configured to:
obtain a plurality of reference arrays based on a plurality of reconstructed samples neighboring a block of video data, wherein the plurality of reference arrays include a first reference array located one pixel position from the block of video data and one or more second reference arrays respectively located at pixel positions greater than the one pixel position from the block of video data;
obtain a plurality of predicted samples for a target pixel of the block of video data respectively from the plurality of reference arrays;
obtain a final intra prediction for the target pixel of the block of video, the final intra prediction being based on a function of the plurality of predicted samples respectively from the plurality of reference arrays, wherein the function comprises a weighted combination of the plurality of predicted samples, and wherein a plurality of weights of the weighted combination decrease as distances between the target pixel and the respective reference arrays increase; and
encode the block of video using the final intra prediction.
4. An apparatus, comprising:
a memory, and
a processor, configured to:
obtain a plurality of reference arrays based on a plurality of decoded samples neighboring a block of video data, wherein the plurality of reference arrays include a first reference array located one pixel position from the block of video data and one or more second reference arrays respectively located at pixel positions greater than the one pixel position from the block of video data;
obtain a plurality of predicted samples for a target pixel of the block of video data respectively from the plurality of reference arrays;
obtain a final intra prediction for the target pixel of the block of video, the final intra prediction being based on a function of the plurality of predicted samples respectively from the plurality of reference arrays, wherein the function comprises a weighted combination of the plurality of predicted samples, and wherein a plurality of weights of the weighted combination decrease as distances between the target pixel and the respective reference arrays increase; and
decode the block of video using the final intra prediction.
5. The method of claim 1 , wherein each of the plurality of reference arrays respectively comprises a row of pixels above, and a column of pixels to the left of, said block of video data.
6. The method of claim 1 , wherein the block is a luma or chroma block.
7. The method of claim 1 , wherein the obtaining of the plurality of predicted samples for the target pixel of the block of video data respectively from the plurality of reference arrays includes interpolation of two adjacent reconstructed samples of a respective one of the plurality of reference arrays.
8. The method of claim 1 , wherein the weighted combination is a normalized, weighted combination of the plurality of reference arrays.
9. The method of claim 1 , further comprising:
signaling a number of the plurality of reference arrays used to obtain the plurality of predicted samples.
10. A non-transitory computer readable medium comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of claim 1 .
11. A non-transitory computer readable medium comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of claim 2 .
12. The method of claim 2 , wherein each of the plurality of reference arrays respectively comprises a row of pixels above, and a column of pixels to the left of, said block of video data.
13. The method of claim 2 , wherein the block is a luma block.
14. The method of claim 2 , wherein the obtaining of the plurality of predicted samples for the target pixel of the block of video data respectively from the plurality of reference arrays includes interpolation of two adjacent reconstructed samples of a respective one of the plurality of reference arrays.
15. The method of claim 2 , wherein the weighted combination is a normalized, weighted combination.
16. The apparatus of claim 3 , wherein each of the plurality of reference arrays respectively comprises a row of pixels above, and a column of pixels to the left of, said block of video data.
17. The apparatus of claim 3 , wherein the obtaining of the plurality of predicted samples for the target pixel of the block of video data respectively from the plurality of reference arrays includes interpolation of two adjacent reconstructed samples of a respective one of the plurality of reference arrays.
18. The apparatus of claim 4 , wherein each of the plurality of reference arrays respectively comprises a row of pixels above, and a column of pixels to the left of, said block of video data.
19. The apparatus of claim 4 , wherein the obtaining of the plurality of predicted samples for the target pixel of the block of video data respectively from the plurality of reference arrays includes interpolation of two adjacent reconstructed samples of a respective one of the plurality of reference arrays.