| 摘要 |
<p>4:2:2 or 4:4:4 video data is coded (and subsequently decoded) by predicting luminance and/or chrominance samples of an image from other samples from the same image according to a prediction mode associated with the sample to be predicted. The prediction mode is selected for each of a plurality of blocks of samples from a set of two or more candidate prediction modes. Differences between the samples and the respective predicted samples are detected and a frequency-separation transform (e.g. DCT, Karhunen-Loeve) is selected from two or more candidate frequency separation transforms according to the prediction mode associated with a current block of samples using a mapping between transform and prediction mode. The mapping is different, with respect to chrominance and luminance samples, for at least the 4:4:4: format. The detected differences are encoded by frequency-separating the differences using the selected frequency-separation transform. Also disclosed is encoding 4:2:2 or 4:4:4 video data in which coefficients derived from frequency-separated differences for blocks of samples are encoded in an order defined by a scan pattern selected according to a prediction mode-scan pattern map (fig. 38). The map is different for chroma and luma samples at least for the 4:4:4 format.</p> |
