Method for decoding chroma image using luma quantization
Provided is a method that derives a chroma intra prediction mode of a prediction unit, determines a size of a current chroma block using luma transform size information, generates a chroma prediction block of the current chroma block using the chroma intra prediction mode, generates a chroma residual block of the current chroma block using the chroma intra prediction mode and a chroma quantization parameter, generates a chroma reconstructed block adding the chroma prediction block and the chroma residual block, and the chroma quantization parameter is generated using a luma quantization parameter and information indicating the relationship between the luma quantization parameter and the chroma quantization parameter. Therefore, the coding efficiency is improved by adjusting the chroma quantization parameter per picture. Also, the amount of bits for transmitting the luma and chroma quantization parameters is reduced by encoding the luma quantization parameter using neighboring luma quantization parameters.
1. A method for decoding chroma image, comprising:
deriving a chroma intra prediction mode of a prediction unit;
determining a size of a chroma prediction block using luma transform size information;
generating the chroma prediction block using the chroma intra prediction mode;
generating a chroma residual block using the chroma intra prediction mode and a chroma quantization parameter; and
generating a chroma reconstructed block adding the chroma prediction block and the chroma residual block,
wherein the chroma quantization parameter is generated using a luma quantization parameter and information indicating the relationship between the luma quantization parameter and the chroma quantization parameter, and
the luma quantization parameter is generated using a luma quantization parameter predictor and a luma differential quantization parameter, the luma quantization parameter predictor is generated using two quantization parameters which are determined in a predetermined order among a left luma quantization parameter, an above luma quantization parameter and a previous luma quantization parameter when two or three luma quantization parameters are available.
2. The method of claim 1 , wherein if both of the left luma quantization parameter and the above luma quantization parameter are available, an average of the left luma quantization parameter and the above luma quantization parameter is set as the luma quantization parameter predictor.
3. The method of claim 1 , wherein if only one of the left luma quantization parameter and the above luma quantization parameter is available, an average of the available luma quantization parameter and the luma previous quantization parameter is set as the luma quantization parameter predictor.
4. The method of claim 1 , wherein the luma quantization parameter is generated per quantization unit and a minimum size of the quantization unit is adjusted per picture.
5. The method of claim 4 , wherein the minimum size of the quantization unit is derived using a parameter specifying the depth of a minimum size of the quantization unit and the size of the largest coding unit.
6. The method of claim 1 , wherein the differential luma quantization parameter is restored by arithmetic-decoding an encoded differential luma quantization parameter to generate a bin string and by applying inverse binarization to the bin string.
7. The method of claim 6 , wherein the bin string includes a bin string indicating an absolute value of the differential luma quantization parameter and a bin indicating a sign of the differential luma quantization parameter.