Coding distortion removal method, video encoding method, video decoding method, and apparatus and program for the same
View Patent ↗In an interlaced picture having a mixture of field structure macroblocks and frame structure macroblocks, a coding distortion removal method removes coding distortion in blocks composing the macroblock. The method removes a coding distortion in a block block which is included in a macroblock of a frame structure, and which is not adjacent to a boundary of a macroblock of field structure. A coding distortion in a block which is included in a macroblock of the field structure, and which is not adjacent to a boundary of a macroblock of the frame structure is then removed, and a coding distortion in a block which is adjacent to both a macroblock of the frame structure and a macroblock of the field structure is removed.
1. A receiver for receiving multiplexed data in which audio data and encoded video data are multiplexed, said receiver comprising:
a demultiplexer that demultiplexes the audio data and the encoded video data from said multiplexed data;
an audio processor that decodes the demultiplexed audio data; and
a video decoder that decodes the demultiplexed encoded video data;
wherein said video decoder includes a picture decoder that decodes a coded picture contained in the encoded video data, said coded picture being obtained by coding a picture on a macroblock basis, said macroblock being formed by dividing the picture into a plurality of macroblocks, and
wherein said picture decoder comprises:
a decoder that decodes a coded motion compensation error to output a motion compensation error on a macroblock basis;
a motion compensator that outputs a motion compensated macroblock from a reference picture;
an adder that adds the motion compensation error and the motion compensated macroblock to output a reconstructed macroblock; and
a coding distortion remover that removes a coding distortion in the reconstructed macroblock to output a decoded macroblock,
wherein the coding distortion remover comprises:
a motion compensation block boundary detector that receives a motion compensation block size signal and the motion compensation error, determines whether the motion compensation error of two adjacent coding blocks located inside the macroblock is zero, and determines whether a boundary of two adjacent coding blocks is also a boundary of two adjacent motion compensation blocks; and
a selector that selects whether the coding distortion remover removes the coding distortion or not,
wherein the coding distortion remover removes a coding distortion at a boundary between two adjacent coding blocks, one of the two adjacent coding blocks being located inside one of the two adjacent motion compensation blocks and the other coding block being located inside the other of the two adjacent motion compensation blocks, when a respective motion compensation error of the two adjacent motion compensation blocks located inside the macroblock is zero, and does not remove a coding distortion at a boundary between two adjacent coding blocks, both of the two adjacent coding blocks being located inside one of the two adjacent motion compensation blocks, when a respective coded motion compensation error of the two adjacent motion compensation blocks located inside the macroblock is zero, wherein a block size of each of the two adjacent motion compensation blocks is smaller than a block size of the macroblock and is larger than a block size of the coding block, the motion compensation block being a unit of motion compensation and the coding block being a unit of frequency transformation.
2. The receiver according to claim 1 , wherein
a block size of a macroblock is a size of 16×16 pixels, and a block size of the coding block is a size of 4×4 pixels.
3. A transmitter for transmitting multiplexed data in which audio data and encoded video data are multiplexed, said transmitter comprising:
an audio data processor that generates the audio data through encoding of an audio signal;
a video encoder that generates encoded video data through encoding of a video signal; and
a multiplexer that multiplexes the audio data and the encoded video data,
wherein said video encoder includes a picture coder that generates a coded picture by coding a picture divided into a plurality of macroblocks on a macroblock basis, and
wherein said picture coder comprises:
a motion compensator that outputs a motion compensated macroblock from a reference picture;
a motion compensation error generator that generates a motion compensation error between the motion compensated macroblock and an input picture to be coded;
a coder that codes the motion compensation error to generate a coded motion compensation error;
a decoder that decodes the coded motion compensation error to output a decoded motion compensation error on a macroblock basis;
an adder that adds the decoded motion compensation error and the motion compensated macroblock to output a reconstructed macroblock; and
a coding distortion remover that removes a coding distortion in the reconstructed macroblock to output a decoded macroblock,
wherein the coding distortion remover comprises:
a motion compensation block boundary detector that receives a motion compensation block size signal and the motion compensation error, determines whether the motion compensation error of two adjacent coding blocks located inside the macroblock is zero, and determines whether a boundary of two adjacent coding blocks is also a boundary of two adjacent motion compensation blocks; and
a selector that selects whether the coding distortion remover removes the coding distortion or not,
wherein the coding distortion remover removes a coding distortion at a boundary between two adjacent coding blocks, one of the two adjacent coding blocks being located inside one of the two adjacent motion compensation blocks and the other coding block being located inside the other of the two adjacent motion compensation blocks, when a respective motion compensation error of the two adjacent motion compensation blocks located inside the macroblock is zero, and does not remove a coding distortion at a boundary between two adjacent coding blocks, both of the two adjacent coding blocks being located inside one of the two adjacent motion compensation blocks, when a respective motion compensation error of the two adjacent motion compensation blocks located inside the macroblock is zero, wherein a block size of each of the two adjacent motion compensation block is smaller than a block size of the macroblock and is larger than a block size of the coding block, the motion compensation block being a unit of motion compensation and the coding block being a unit of frequency transformation.
4. The transmitter according to claim 3 , wherein
a block size of the macroblock is a size of 16×16 pixels, and a block size of the coding block is a size of 4×4 pixels.