Method for partitioning block and decoding device
Disclosed are a method for partitioning a block and a decoding device. A method for decoding an image comprises the steps of: partitioning one prediction unit, in which an intra prediction is performed in a predetermined intra prediction mode, into a plurality of transform units; and performing the intra prediction for each of the plurality of the transform units on the basis of the predetermined intra prediction mode. Thus, the invention increases the image encoding efficiency.
1. A decoding apparatus for decoding an image, the apparatus comprising:
an entropy decoder configured to:
receive information of a size of a minimum coding unit, wherein the minimum coding unit represents a smallest coding unit derivable from a largest coding unit based on a quad-tree structure;
derive, based on the information of the size of the minimum coding unit, a size of a minimum coding unit;
derive a current coding unit, wherein a coding unit is split into multiple coding units in a current picture, wherein the current coding unit is one of the multiple coding units, and wherein an intra prediction is applied on the current coding unit;
check whether a size of the current coding unit is equal to the size of the minimum coding unit; and
derive one or multiple prediction units from the current coding unit based on the result of the checking; and
a predictor configured to perform the intra prediction with respect to the derived one or multiple prediction units,
wherein based on the size of the current coding unit being equal to the size of the minimum coding unit, explicit partition information is received and the one or multiple prediction units are derived from the current coding unit based on the explicit partition information, and
wherein based on the size of the current coding unit being different from the size of the minimum coding unit, the one prediction unit of a same size as the current coding unit is derived without receiving the explicit partition information, and the current coding unit cannot be partitioned into the multiple prediction units.
2. The apparatus of claim 1 , wherein, based on a size of the current coding unit being a 2N×2N size, a size of the multiple prediction units is an N×N size, N is a positive integer.
3. The apparatus of claim 1 , wherein with respect to the current coding unit, a size of a minimum coding unit for performing an intra prediction and a size of a minimum coding unit for performing an inter prediction is determined differently.
4. The apparatus of claim 1 , wherein the one prediction unit is derived from the current coding unit and the prediction process is performed by differently setting a size of a coding unit for performing an intra prediction and a size of a coding unit for performing an inter prediction.
5. The apparatus of claim 1 , wherein prev_intra_luma_pred_flag information is received per a prediction unit in the current coding unit when the prediction type with respect to the current coding unit is the intra prediction, and
wherein a specific intra prediction mode is determined for the prediction unit in the current coding unit based on the received prev_intra_luma_pred_flag information.
6. An encoding apparatus for encoding an image, the apparatus comprising:
a picture partitioner configured to:
derive a size of a minimum coding unit, wherein the minimum coding unit represents a smallest coding unit derivable from a largest coding unit based on a quad-tree structure;
derive a current coding unit, wherein a coding unit is split into multiple coding units in a current picture, wherein the current coding unit is one of the multiple coding units, wherein an intra prediction is applied on the current coding unit;
check whether a size of the current coding unit is equal to the size of the minimum coding unit;
derive one or multiple prediction units from the current coding unit based on the result of the checking; and
an entropy encoder configured to encode image information including information on the size of the minimum coding unit,
wherein based on the size of the current coding unit being equal to the size of the minimum coding unit, the one or multiple prediction units are derived from the current coding unit and explicit partition information indicating whether the one or multiple prediction units are derived from the current coding unit is generated, and the explicit partition information is included in the image information, and
wherein based on the size of the current coding unit being different from the size of the minimum coding unit, the one prediction unit of a same size as the current coding unit is derived from the current coding unit and the current coding unit cannot be partitioned into the multiple prediction units.
7. The apparatus of claim 6 , wherein based on a size of the current coding unit being a 2N×2N size, a size of the multiple prediction units is an N×N size, N is a positive integer.
8. The apparatus of claim 6 , wherein with respect to the current coding unit, a size of a minimum coding unit for performing an intra prediction and a size of a minimum coding unit for performing an inter prediction is determined differently.
9. The apparatus of claim 6 , wherein the one prediction unit is derived from the current coding unit and the prediction process is performed by differently setting a size of a coding unit for performing an intra prediction and a size of a coding unit for performing an inter prediction.
10. The apparatus of claim 6 , wherein prev_intra_lumapred_flag information is encoded as to be comprised in the encoded image information per a prediction unit in the current coding unit when the prediction type with respect to the current coding unit is the intra prediction, and
wherein prev_intra_luma_pred_flag information specifies an intra prediction mode of the prediction unit.
11. A non-transitory decoder-readable storage medium storing encoded image information generated by an encoding method comprising:
deriving a size of a minimum coding unit, wherein the minimum coding unit represents a smallest coding unit derivable from a largest coding unit based on a quad-tree structure;
deriving a current coding unit, wherein a coding unit is split into multiple coding units in a current picture, wherein the current coding unit is one of the multiple coding units, wherein an intra prediction is applied on the current coding unit;
checking whether a size of the current coding unit is equal to the size of the minimum coding unit;
deriving one or multiple prediction units from the current coding unit based on the result of the checking; and
encoding image information including information on the size of the minimum coding unit,
wherein based on the size of the current coding unit being equal to the size of the minimum coding unit, the one or multiple prediction units are derived from the current coding unit and explicit partition information indicating whether the one or multiple prediction units are derived from the current coding unit is generated, and the explicit partition information is included in the image information, and,
wherein based on the size of the current coding unit being different from the size of the minimum coding unit, the one prediction unit of a same size as the current coding unit is derived from the current coding unit and the current coding unit cannot be partitioned into the multiple prediction units.
12. The non-transitory decoder-readable storage medium of claim 11 , wherein based on a size of the current coding unit being a 2N×2N size, a size of each of the multiple prediction units is an N×N size, N is a positive integer.
13. The apparatus of claim 1 , wherein, based on the quad-tree structure, the entropy decoder splits a first coding unit into second coding units,
wherein, based on one of the second coding units being a leaf node, the one of the second coding units is derived as the current coding unit, and
wherein, based on the quad-tree structure indicating that the one of the second coding units is not the leaf node, the one of the second coding units is further split into third coding units.
14. The apparatus of claim 1 , wherein, in response to the multiple prediction units being derived based on the explicit partition information, intra prediction mode information for each of the multiple prediction units is received,
wherein an intra prediction mode for each of the multiple prediction units is determined according to the intra prediction mode information, and
wherein the intra prediction with respect to the derived multiple prediction units is performed according to the intra prediction mode for each of the multiple prediction units.
15. The apparatus of claim 1 , wherein, based on the one prediction unit being derived, intra prediction mode information for the one prediction unit is received,
wherein an intra prediction mode for the one prediction unit is determined according to the intra prediction mode information, and
wherein the intra prediction with respect to the derived one prediction unit is performed according to the intra prediction mode for the one prediction unit.