| 摘要 |
The method for decoding intra predictions according to the present invention comprises: de-multiplexing residual signals and intra-prediction information; decoding the intra-prediction mode of the current prediction unit using the intra-prediction mode of a prediction unit adjacent to the current prediction unit and the intra-prediction information; generating reference pixels using one or more usable reference pixels when the current prediction unit has unusable reference pixels; filtering the reference pixels on the basis of the decoded intra-prediction mode, and decoding the demultiplexed residual signals to generate a residual block; and generating a restoration block using the prediction block and the residual block. Accordingly, the method for encoding the prediction mode can be provided, where an additional amount of encoding caused by the increase in the number of the intra-prediction modes can be effectively reduced. Also, a prediction mode close to the original image can be generated to provide a method for decoding corresponding to the method for encoding which decreases the amount of encodings, thereby enhancing compression efficiency and playback efficiency in encoding/decoding video. |
