Method and apparatus for encoding and/or decoding moving pictures

1. A method of decoding moving pictures by adaptively using a plurality of quantization weight matrices not related by a scalar, the method comprising:

receiving an encoded stream of an input image, the encoded stream comprising at least one macroblock and information specifying an respective quantization weight matrix corresponding to a respectively determined spatial characteristic of each of the at least one macroblock among the plurality of quantization weight matrices;

performing variable length decoding on the encoded stream; and

inversely quantizing each variable length decoded macroblock by using the respective quantization weight matrix specified by the information,

wherein the plurality of the quantization weight matrices comprise a plurality of quantization weight matrices for a luminance component of each macroblock that is determined to be an intra macroblock,

wherein the respective quantization weight matrix used in the inverse quantizing is individually selected, for each of plural macroblocks of the input image, from one of the plurality of quantization weight matrices for the intra macroblock, and

wherein, when a result of a determination of a spatial characteristic of a determined first intra block of the input image is different from a result of a determination of a spatial characteristic of a determined second intra block of the input image, a different quantization weight matrix is selected from the plurality of quantization weight matrices for the luminance component for the first block than is selected for the second block.

2. A method of decoding moving pictures by adaptively using a plurality of quantization weight matrices not related by a scalar, the method comprising:

receiving an encoded stream of an input image, the encoded stream comprising at least one macroblock and information specifying a respective quantization weight matrix corresponding to a respectively determined spatial characteristic of each of the at least one macroblock among the plurality of quantization weight matrices;

performing variable length decoding on the encoded stream; and

inversely quantizing each variable length decoded macroblock by using the respective quantization weight matrix specified by the information,

wherein the plurality of the quantization weight matrices comprise a plurality of quantization weight matrices for a luminance component of each macroblock that is determined to be an intra macroblock,

wherein the information is a difference between and index of a quantization weight matrix corresponding to a macroblock and an index of a quantization weight matrix corresponding to an adjacent macroblock, and

wherein, when a result of a determination of a spatial characteristic of a determined first intra block of the input image is different from a result of a determination of a spatial characteristic of a determined second intra block of the input image, a different quantization weight matrix is selected from the plurality of quantization weight matrices for the luminance component for the first block than is selected for the second block.

3. A method of decoding moving pictures by adaptively using a plurality of quantization weight matrices not related by a scalar, the method comprising:

receiving an encoded stream of an input image, the encoded stream comprising at least one macroblock and information specifying a respective quantization weight matrix corresponding to a respectively determined spatial characteristic of each of the at least one macroblock among the plurality of quantization weight matrices;

performing variable length decoding on the encoded stream; and

inversely quantizing each variable length decoded macroblock by using the respective quantization weight matrix specified by the information,

wherein the respective quantization weight matrix specified by the respective index information used in the inverse quantizing is obtained, for a respective inverse quantization of each of plural macroblocks of the input image, based upon a review of the index information as being a determined one of plural indices each identifying one of a plurality of quantization weight matrices for a single color space component of intra macroblocks, among the plurality of quantization weight matrices, and

wherein, when a result of a determination of a spatial characteristic of a determined first intra block of the input image is different from a result of a determination of a spatial characteristic of a determined second intra block of the input image, a different quantization weight matrix is selected from the plurality of quantization weight matrices for the color space component for the first block than is selected for the second block.

4. A method of decoding moving pictures by adaptively using a plurality of quantization weight matrices not related by a scalar, the method comprising:

receiving an encoded stream of an input image, the encoded stream comprising at least one macroblock and information specifying a respective quantization weight matrix corresponding to a respectively determined spatial characteristic of each of the at least one macroblock among the plurality of quantization weight matrices;

performing variable length decoding on the encoded stream; and

inversely quantizing each variable length decoded macroblock by using the respective quantization weight matrix specified by the information,

wherein the respective quantization weight matrix used in the inverse quantizing is individually selected, for each of plural macroblocks of the input image, from the plurality of quantization weight matrices for a single color space component of each macroblock that is determined to be an intra macroblocks, based on the respective information, and

wherein, when a result of a determination of a spatial characteristic of a determined first intra block of the input image is different from a result of a determination of a spatial characteristic of a determined second intra block of the input image, a different quantization weight matrix is selected from the plurality of quantization weight matrices for the color space component for the first block than is selected for the second block.