Method and apparatus for encoding/decoding images considering low frequency components

1. A method performed by an apparatus for encoding a current block, the method comprising:

generating a predicted block by predicting the current block;

generating a residual block of the current block by subtracting the predicted block from the current block;

partitioning the residual block into a plurality of subblocks and transforming each of the subblocks by using a transform unit of a size identical to each of the subblocks, to thereby generate transform blocks of the subblocks;

quantizing the transform blocks;

reconstructing the transform blocks by inversely quantizing the quantized transform blocks;

reconstructing residual subblocks by inversely transforming the reconstructed transform blocks;

reconstructing the current block by adding the predicted block and the reconstructed residual subblocks;

deblocking filtering the reconstructed current block; and

encoding (i) information of a flag indicating whether offset information is encoded and (ii) transform coefficients of each of the quantized transform blocks, wherein the offset information indicates offset values, the offset values indicating pixel values used for compensating for distortion incurred by the difference between original pixels and reconstructed pixels,

wherein, when the flag indicates that the offset information is encoded, the method further comprises

adding the offset values to the reconstructed pixels in the deblocking filtered current block, and

encoding the offset information for indicating the offset values,

wherein the method further comprises

generating a bitstream including the encoded information of the flag and the encoded offset information, the predicted block and the residual subblocks.

2. The method of claim 1 , further comprising:

encoding information related to the partitioning of the residual block into the plurality of subblocks.

3. The method of claim 1 , wherein at least one of the subblocks has a size equal to or larger than 16×16 pixel size.

4. A method performed by an apparatus for decoding a current block, the method comprising:

reconstructing, from a bitstream, information of a flag indicating whether offset information is encoded, wherein the offset information indicates offset values, the offset values indicating pixel values used for compensating for distortion incurred by the difference between original pixels and reconstructed pixels;

reconstructing the offset information from the bitstream, when the flag indicates that the offset information is encoded;

generating a predicted block of the current block to be decoded by predicting the current block;

partitioning the current block into one or more subblocks, and generating transform blocks corresponding to the subblocks, by extracting, from the bitstream, transform coefficients corresponding to each of the subblocks;

inversely quantizing and then inversely transforming each of the transform blocks by using a transform unit of a size identical to a corresponding subblock, and thereby reconstructing residual subblocks corresponding to the subblocks;

reconstructing the current block by adding the predicted block and the reconstructed residual subblocks; and

deblocking filtering the reconstructed current block,

wherein, when the flag indicates that the offset information is encoded, the method further comprises

adding reconstructed pixels in the deblocking filtered current block and the offset values derived from the reconstructed offset information.

5. The method of claim 4 , further comprising:

extracting, from the bitstream, information related to the partition of the current block into the subblocks.

6. The method of claim 5 , wherein at least one of the subblocks has a size equal to or larger than 16×16 pixel size.

7. The method of claim 4 , wherein the offset information is one or more low frequency components of an error block that is differences between original pixels in the current block and reconstructed pixels in the deblocking filtered current block.

8. The method of claim 7 , wherein the offset values are derived by inversely transforming the one or more low frequency components.

9. A video decoding apparatus for decoding a current block, the apparatus comprising:

a decoder configured to

reconstruct, from a bitstream, information of a flag indicating whether offset information is encoded, wherein the offset information indicates offset values, the offset values indicating pixel values used for compensating for distortion incurred by the difference between original pixels and reconstructed pixels,

reconstruct the offset information from the bitstream, when the flag indicates that the offset information is encoded, and

reconstruct quantized transform coefficients from the bitstream,

a predictor configured to generate a predicted block of the current block to be decoded by predicting the current block;

an inverse quantizer and inverse transformer configured to inversely quantize and then inversely transform the quantized transform coefficients to generate a residual block;

an first adder configured to reconstruct the current block by adding the predicted block and the residual block;

a deblocking filter configured to perform deblocking filtering on the reconstructed current block; and

an second adder configured to add reconstructed pixels in the deblocking filtered current block and the offset values derived from the reconstructed offset information, when the flag indicates that the offset information is encoded.

10. The apparatus of claim 9 , wherein the offset information is one or more low frequency components of an error block that is differences between the original pixels in the current block and the reconstructed pixels in the deblocking filtered current block.

11. The apparatus of claim 10 , wherein the offset values are derived by inversely transforming the one or more low frequency components.