摘要 |
A component of an image processor receives at least a portion of one or more reconstructed versions of an image including a causal context of an object pixel of the image, and provides a prediction for the object pixel computed as a weighted sum according to at least one measure of correlation and a weighting policy. A component receives a first prediction of an object pixel and reconstructed versions of an image, and provides a second prediction for the object pixel in accordance with the first prediction for the object pixel and a substantially mean error for the context. A component receives one or more prediction differences versions of the image including a causal context prediction difference of an object prediction difference of the image, and provides a prediction for the object prediction difference computed as a weighted sum according to at least one measure of correlation and a weighting policy.</PTEXT>
|