| 摘要 |
Video coding / decoding quantises blocks of frequency-transformed luminance and chrominance component video data in a 4:4:4 or a 4:2:2 format according to a selected quantisation parameter which defines a quantisation step size. A quantisation parameter is defined as an association between luminance and chrominance quantisation parameters. The association is such that a maximum chrominance quantisation step size is less than a maximum luminance quantisation step size for the 4:2:2 format, but equal to the maximum luminance quantisation step size for the 4:4:4 format. Each component of the frequency-transformed data is quantised according to a respective value derived from the respective quantisation step size and the result is rounded to an integer value to generate a corresponding block of quantised spatial frequency data. |
