Method of decoding video data

摘要

Provided is a method derives a reference picture index and a motion vector of a current prediction unit, generates a prediction block of the current prediction unit using the reference picture index and the motion vector, generating a residual block by inverse-scan, inverse-quantization and inverse transform, and generates reconstructed pixels using the prediction block and the residual block. Prediction pixels of the prediction block is generated using an interpolation filter selected based on the motion vector. Accordingly, the coding efficiency of the motion information is improved by including various merge candidates. Also, the computational complexity of an encoder and a decoder is reduced by selecting different filter according to location of the prediction pixels determined by the motion vector.

主权项

1. A method of decoding video data in uni-directional prediction by a decoding apparatus, the method comprising:
deriving, by the decoding apparatus, a reference picture index and a motion vector of a current prediction unit; generating, by the decoding apparatus, a prediction block of the current prediction unit using the reference picture index and the motion vector; generating, by the decoding apparatus, a quantized block by inversely scanning quantized coefficient components; generating, by the decoding apparatus, a transformed block by inversely quantizing the quantized block using a quantization parameter; generating, by the decoding apparatus, a residual block by inversely transforming the transformed block; and generating, by the decoding apparatus, reconstructed pixels using the prediction block and the residual block, wherein prediction pixels of the prediction block are generated using an interpolation filter selected based on the motion vector, the interpolation filter being a 7-tap filter if the motion vector indicates a quarter pixel position, and the interpolation filter being an 8-tap filter if the motion vector indicates a half pixel position, wherein the quantization parameter is derived by adding a differential quantization parameter and a quantization parameter predictor, wherein, when both of a left quantization parameter and an above quantization parameter are available, the quantization parameter predictor is an average of the left quantization parameter and the above quantization parameter, wherein, when only one of the left quantization parameter and the above quantization parameter is available, the quantization parameter predictor is an average of a previous quantization parameter and the available one of the left quantization parameter and the above quantization parameter, wherein, when both of the left quantization parameter and the above quantization parameter are unavailable, the quantization parameter predictor is set as the previous quantization parameter, and wherein, when a size of the quantized block is larger than a predetermined size, a plurality of subsets are generated by inversely scanning the quantized coefficient components and the quantized block is generated by inversely scanning the plurality of subsets using a diagonal scan pattern.