Video encoding and decoding method and apparatus involving boundary filtering

1. A video decoding method comprising:

obtaining at least two prediction units split from a coding unit of an image;

determining a filtering condition of a prediction unit boundary filter, and performing filtering by using the prediction unit boundary filter on a predicted value of a pixel adjacent to a boundary between the at least two prediction units based on the filtering condition; and

reconstructing a value of the pixel by using the filtered predicted value,

wherein the determining of the filtering condition of the prediction unit boundary filter comprises:

when a transformation coefficient does not exist in the coding unit, determining to perform a different type of the filtering from a type of the filtering performed when the transformation coefficient exists in the coding unit; and

determining an intensity of the filtering by adjusting filtering coefficients and/or a number of filtering taps, based on information indicating whether a residual value exists in the coding unit or not, and

wherein the filtering condition of the boundary filter is shared with a deblocking filter.

2. The video decoding method of claim 1 , wherein the coding unit is split from the image based on coding unit split information obtained from a bitstream,

wherein the at least two prediction units comprise a first prediction unit and a second prediction unit obtained by splitting at least one of a height and a width of the coding unit, based on a partition mode obtained from the bitstream, and

wherein the boundary is a boundary between the first and second prediction units.

3. The video decoding method of claim 2 , further comprising obtaining at least one transformation unit from the coding unit based on transformation unit split information obtained from the bitstream,

wherein the first and second prediction units are comprised in the transformation unit.

4. The video decoding method of claim 2 , wherein the pixel is comprised in the first prediction unit, and

wherein the filtering is performed by using a predicted value of a neighboring pixel comprised in the second prediction unit and adjacent to the pixel.

5. The video decoding method of claim 1 , wherein the determining of the filtering condition of the boundary filter further comprises obtaining a filtering flag indicating whether to filter the predicted value, from a bitstream.

6. The video decoding method of claim 5 , wherein the filtering flag is obtained when the block is in an inter prediction mode or in an intra prediction mode.

7. The video decoding method of claim 1 , wherein the boundary has a linear shape or a nonlinear shape.

8. The video decoding method of claim 1 , wherein the determining of the filtering condition of the boundary filter further comprises determining at least one of intensity of the filtering and a size of the filtering based on at least one of a distance from the pixel to the boundary, the predicted value, the residual values, a predicted value of at least one neighboring pixel adjacent to the pixel, motion information of the partitions, and mode information of the partitions.

9. The video decoding method of claim 1 , wherein the reconstructing of the value of the pixel comprises reconstructing the value of the pixel by using the filtered predicted value and the residual value of the pixel.

10. A video decoding apparatus comprising at least one processor configured to:

obtain at least two prediction units split from a coding unit of an image;

determine a filtering condition of a prediction unit boundary filter, and perform filtering by using the prediction unit boundary filter on a predicted value of a pixel adjacent to a boundary between the at least two prediction units based on the filtering condition; and

reconstruct a value of the pixel by using the filtered predicted value,

wherein the at least one processor is further configured to:

determine, when a transformation coefficient does not exist in the coding unit, to perform a different type of the filtering from a type of the filtering performed when the transformation coefficient exists in the coding unit, and

determine an intensity of the filtering by adjusting filtering coefficients and/or a number of filtering taps, based on information indicating whether a residual value exists in the coding unit or not, and

wherein the filtering condition of the boundary filter is shared with a deblocking filter.

11. A video encoding method comprising:

obtaining at least two prediction units split from a coding unit of an image;

determining a filtering condition of a prediction unit boundary filter, and performing filtering by using the prediction unit boundary filter on a predicted value of a pixel adjacent to a boundary between the at least two prediction units based on the filtering condition; and

encoding the residual value by using the filtered predicted value and a value of the pixel if the predicted value is filtered,

wherein the determining of the filtering condition of the prediction unit boundary filter comprises:

when a transformation coefficient does not exist in the coding unit, determining to perform a different type of the filtering from a type of the filtering performed when the transformation coefficient exists in the coding unit; and

determining an intensity of the filtering by adjusting filtering coefficients and/or a number of filtering taps, based on information indicating whether a residual value exists in the coding unit or not, and

wherein the filtering condition of the boundary filter is shared with a deblocking filter.

12. The video encoding method of claim 11 , wherein the coding unit is split from the image based on coding unit split information,

wherein the at least two prediction units comprise a first prediction unit and a second prediction unit obtained by splitting at least one of a height and a width of the coding unit, based on a partition mode, and

wherein the boundary is a boundary between the first and second prediction units.

13. The video encoding method of claim 12 , further comprising obtaining at least one transformation unit from the coding unit based on transformation unit split information,

wherein the first and second prediction units are comprised in the transformation unit.

14. The video encoding method of claim 12 , wherein the pixel is comprised in the first prediction unit, and

wherein the filtering is performed by using a predicted value of a neighboring pixel comprised in the second prediction unit and adjacent to the pixel.

15. The video encoding method of claim 11 , wherein the determining of the filtering condition of the boundary filter further comprises generating a filtering flag indicating whether to filter the predicted value.

16. The video encoding method of claim 15 , wherein the filtering flag is generated when the block is in an inter prediction mode or in an intra prediction mode.

17. The video encoding method of claim 11 , wherein the boundary has a linear shape or a nonlinear shape.

18. The video encoding method of claim 11 , wherein the determining of the filtering condition of the boundary filter further comprises determining at least one of intensity of the filtering and a size of the filtering based on at least one of a distance from the pixel to the boundary, the predicted value, the residual values, a predicted value of at least one neighboring pixel adjacent to the pixel, motion information of the partitions, and mode information of the partitions.