Method and device for generating a prediction block to encode and decode an image
A method for decoding an image according to the present invention comprises the steps of: restoring a residual block by performing inverse quantization and inverse transformation for the entropy-decoded residual block; generating a prediction block by performing intra prediction for a current block; and restoring an image by adding the restored residual block to the prediction block, wherein the step of generating the prediction block further comprises a step for generating a final prediction value of a pixel to be predicted, on the basis of a first prediction value of the pixel to be predicted, which is included in the current block, and of a final correction value that is calculated by performing an arithmetic right shift by a binary digit 1 for a two's complement integer representation with respect to an initial correction value of the pixel to be predicted. Thus, the operational complexity during image encoding/decoding can be reduced.
1. A video decoding method comprising:
generating a prediction block by performing inter prediction for a current block,
determining whether to perform scaling on a motion vector of a reference block based on picture order count (POC) values of a first reference picture of the current block and a second reference picture of the reference block used for inter prediction for the current block, and
scaling the motion vector of the reference block by calculating a first value and a second value based on the POC values,
calculating an offset value by performing an arithmetic right shift on an absolute value of the first value by a binary digit of 1 and calculating an inverse-proportional value of the first value using the offset value, and
calculating a scaling factor based on the inverse-proportional value and the second value, in response to scaling of the motion vector of the reference block being performed,
wherein the first value is a POC difference between a picture including the reference block and the second reference picture,
the second value is a POC difference between a current picture and the first reference picture,
the POC indicates the order of display output,
the motion vector of the reference block is scaled in response to the first reference picture not being the same as the second reference picture when a spatial motion vector is induced,
the motion vector of the reference block is scaled in response to the POC difference between the current picture and the first reference picture not being the same as the POC difference between the picture comprising the reference block and the second reference picture when a temporal motion vector or a temporal merge candidate is induced, and
the scaling factor is derived by performing an addition operation and an arithmetic right shift operation based on multiplication of the second value and the inverse-proportional value of the first value.
2. A video decoding method comprising:
generating a prediction block by performing inter prediction for a current block,
determining whether to perform scaling on a motion vector of a reference block based on picture order count (POC) values of a first reference picture of the current block and a second reference picture of the reference block used for inter prediction for the current block, and
scaling the motion vector of the reference block by calculating a first value and a second value based on the POC values,
calculating an offset value by performing an arithmetic right shift on an absolute value of the first value by a binary digit of 1 and calculating an inverse-proportional value of the first value using the offset value, and
calculating a scaling factor based on the inverse-proportional value and the second value, in response to scaling of the motion vector of the reference block being performed,
wherein the first value is a POC difference between a picture including the reference block and the second reference picture,
the second value is a POC difference between a current picture and the first reference picture,
the POC indicates the order of display output,
the motion vector of the reference block is scaled in response to the first reference picture not being the same as the second reference picture when a spatial motion vector is induced,
the motion vector of the reference block is scaled in response to the POC difference between the current picture and the first reference picture not being the same as the POC difference between the picture comprising the reference block and the second reference picture when a temporal motion vector or a temporal merge candidate is induced, and
the scaling factor is adjusted to be included in a predetermined range.