Intra prediction in image processing
An intra prediction method and a device using the intra prediction method are provided. The intra prediction method includes the steps of: deriving a current prediction mode as a prediction mode of a current block; constructing neighboring samples of the current block with available reference samples; filtering the available reference samples; and generating predicted samples of the current block on the basis of the filtered available reference samples. The filtering step includes performing the filtering using the available reference sample located in the prediction direction of the current prediction mode and a predetermined number of available reference samples neighboring to the prediction direction of the current prediction mode.
1. An image encoding method comprising:
determining an intra prediction mode for a current block, wherein the intra prediction mode is one of intra prediction modes including a plurality of directional intra prediction modes, and the intra prediction mode for the current block is induced into at least one of a left top diagonal directional intra prediction mode, a vertical intra prediction mode or a horizontal intra prediction mode;
deriving neighboring reference samples of the current block;
filtering the neighboring reference samples;
generating a prediction sample for the current block based on at least one of the filtered neighboring reference samples; and
generating a bitstream including prediction information for the current block,
wherein, in the filtering, filtering is performed by using a first neighboring reference sample located in a prediction direction of the intra prediction mode with regard to the prediction sample and neighboring reference samples adjacent to the first neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the left top diagonal directional intra prediction mode and the first neighboring reference sample is located on a top side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a second neighboring reference sample located on a left side of the first neighboring reference sample and a third neighboring reference sample located on a right side of the first neighboring reference sample,
( r ( x− 1)+2* r ( x )+ r ( x+ 1)+2)>>2
wherein, the r(x) represents a sample value of the first neighboring reference sample, the r(x−1) represents a value of the second neighboring reference sample and the r(x+1) represents a sample value of the third neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the vertical intra prediction mode and the first neighboring reference sample is located on a top side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a fourth neighboring reference sample located on a left side of the first neighboring reference sample and a fifth neighboring reference sample located on a right side of the first neighboring reference sample,
( r ( x− 1)+2* r ( x )+ r ( x+ 1)+2)>>2
wherein, the r(x) represents a sample value of the first neighboring reference sample, the r(x−1) represents a value of the fourth neighboring reference sample and the r(x+1) represents a sample value of the fifth neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the horizontal intra prediction mode and the first neighboring reference sample is located on a left side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a sixth neighboring reference sample located on an upper side of the first neighboring reference sample and a seventh neighboring reference sample located on a lower side of the first neighboring reference sample,
( r ( y− 1)+2* r ( y )+ r ( y+ 1)+2)>>2
wherein, the r(y) represents a sample value of the first neighboring reference sample, the r(y−1) represents a value of the sixth neighboring reference sample and the r(y+1) represents a sample value of the seventh neighboring reference sample.
2. The method of claim 1 , wherein, in the filtering, a neighboring reference sample available in the intra prediction mode is only filtered.
3. The method of claim 1 , wherein it is determined whether the filtering is applied depending on a size and the intra prediction mode of the current block.
4. The method of claim 3 , wherein when it is determined that the filtering is not performed, the prediction sample is generated based on the neighboring reference samples, and
wherein, when the prediction sample is adjacent to at least one of the neighboring reference samples, a smoothing using 2-tap filter is applied to the prediction sample.
5. A digital storage medium, wherein a video bitstream encoded by the image encoding method according to claim 1 is stored.
6. An image decoding method comprising:
deriving an intra prediction mode for a current block, wherein the intra prediction mode is one of intra prediction modes including a plurality of directional intra prediction modes, and the intra prediction mode for the current block is induced into at least one of a left top diagonal directional intra prediction mode, a vertical intra prediction mode or a horizontal intra prediction mode;
deriving neighboring reference samples of the current block;
filtering the neighboring reference samples; and
generating a prediction sample for the current block based on one of the filtered neighboring reference samples,
wherein, in the filtering step, filtering is performed by using a first neighboring reference sample located in a prediction direction of the intra prediction mode with regard to the prediction sample and neighboring reference samples adjacent to the first neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the left top diagonal directional intra prediction mode and the first neighboring reference sample is located on a top side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a second neighboring reference sample located on a left side of the first neighboring reference sample and a third neighboring reference sample located on a right side of the first neighboring reference sample,
( r ( x− 1)+2* r ( x )+ r ( x+ 1)+2)>>2
wherein, the r(x) represents a sample value of the first neighboring reference sample, the r(x−1) represents a value of the second neighboring reference sample and the r(x+1) represents a sample value of the third neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the vertical intra prediction mode and the first neighboring reference sample is located on a top side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a fourth neighboring reference sample located on a left side of the first neighboring reference sample and a fifth neighboring reference sample located on a right side of the first neighboring reference sample,
( r ( x− 1)+2* r ( x )+ r ( x+ 1)+2)>>2
wherein, the r(x) represents a sample value of the first neighboring reference sample, the r(x−1) represents a value of the fourth neighboring reference sample and the r(x+1) represents a sample value of the fifth neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the horizontal intra prediction mode and the first neighboring reference sample is located on a left side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a sixth neighboring reference sample located on an upper side of the first neighboring reference sample and a seventh neighboring reference sample located on a lower side of the first neighboring reference sample,
( r ( y− 1)+2* r ( y )+ r ( y+ 1)+2)>>2
wherein, the r(y) represents a sample value of the first neighboring reference sample, the r(y−1) represents a value of the sixth neighboring reference sample and the r(y+1) represents a sample value of the seventh neighboring reference sample.
7. A transmission method of data for image, the method comprising:
obtaining a bitstream of image information including prediction information for a current block;
transmitting the data comprising the bitstream of the image information including the prediction information,
wherein the prediction information is generated based on an intra prediction process of generating a prediction sample for the current block,
wherein the prediction sample is generated based on at least one of filtered neighboring reference samples derived by determining an intra prediction mode for the current block, deriving neighboring reference samples of the current block, filtering the neighboring reference samples,
wherein the intra prediction mode is one of intra prediction modes including a plurality of directional intra prediction modes, and the intra prediction mode for the current block is induced into at least one of a left top diagonal directional intra prediction mode, a vertical intra prediction mode or a horizontal intra prediction mode,
wherein the filtering is performed by using a first neighboring reference sample located in a prediction direction of the intra prediction mode with regard to the prediction sample and neighboring reference samples adjacent to the first neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the left top diagonal directional intra prediction mode and the first neighboring reference sample is located on a top side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a second neighboring reference sample located on a left side of the first neighboring reference sample and a third neighboring reference sample located on a right side of the first neighboring reference sample,
( r ( x− 1)+2* r ( x )+ r ( x+ 1)+2)>>2
wherein, the r(x) represents a sample value of the first neighboring reference sample, the r(x−1) represents a value of the second neighboring reference sample and the r(x+1) represents a sample value of the third neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the vertical intra prediction mode and the first neighboring reference sample is located on a top side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a fourth neighboring reference sample located on a left side of the first neighboring reference sample and a fifth neighboring reference sample located on a right side of the first neighboring reference sample,
( r ( x− 1)+2* r ( x )+ r ( x+ 1)+2)>>2
wherein, the r(x) represents a sample value of the first neighboring reference sample, the r(x−1) represents a value of the fourth neighboring reference sample and the r(x+1) represents a sample value of the fifth neighboring reference sample,
wherein, when the intra prediction mode for the current block is induced into the horizontal intra prediction mode and the first neighboring reference sample is located on a left side of the current block, a filtered sample value for the first neighboring reference sample is derived based on following equation by using the first neighboring reference sample, a sixth neighboring reference sample located on an upper side of the first neighboring reference sample and a seventh neighboring reference sample located on a lower side of the first neighboring reference sample,
( r ( y− 1)+2* r ( y )+ r ( y+ 1)+2)>>2
wherein, the r(y) represents a sample value of the first neighboring reference sample, the r(y−1) represents a value of the sixth neighboring reference sample and the r(y+1) represents a sample value of the seventh neighboring reference sample.