| 摘要 |
A video encoding method for subjecting an input video having a luminance and two color differences to prediction encoding, the method comprising: obtaining, for a to-be-encoded block of the input video, a plurality of index informations each indicating a reference image and a combination of a plurality of combinations each comprising a weighting factor for each of the luminance and the two color differences and an offset for each of the luminance and the two color differences; generating a prediction image for the to-be-encoded block by multiplying a given number of reference images by the weighting factor corresponding to each reference image and adding the offset; generating a prediction error signal for the to-be-encoded block by calculating an error between the input video and the prediction image; generating a quantized orthogonal transform coefficient for the to-be-encoded block by subjecting the prediction error signal to orthogonal transform and quantization; and encoding, for the to-be-encoded , block, the quantized orthogonal transform coefficient and the given number of index informations and encoding, for one or more of the to-be-encoded blocks, the plurality of combinations. |
