Image prediction method and apparatus for performing intra prediction
An image decoding method according to the present document may comprise the steps of: deriving an intra prediction mode for a current block; deriving reference samples located around the current block; generating a prediction sample for the current block on the basis of the intra prediction mode and the reference samples; and generating residual samples for the current block on the basis of the prediction sample, wherein the step of generating the prediction sample for the current block may comprise the steps of: deriving an interpolation filter, which is to be applied to the reference sample, on the basis of the prediction mode of the current block or the size of the current block; and generating the prediction sample by applying the derived interpolation filter to the reference sample.
1. An image decoding method performed by a decoding apparatus, the method comprising:
deriving an intra prediction mode for a current block;
deriving reference samples contiguous to the current block;
generating a prediction sample for the current block based on the intra prediction mode and the reference samples; and
generating a reconstructed picture for the current block based on the prediction sample,
wherein the generating of the prediction sample for the current block includes:
deriving an interpolation filter to be applied to the reference sample based on a prediction mode of the current block or a size of the current block; and
generating the prediction sample by applying the derived interpolation filter to the reference sample,
wherein the deriving of the interpolation filter includes:
comparing a distance between the prediction sample and the reference sample with a predetermined threshold value; and
deriving the interpolation filter as a Gaussian filter based on the distance exceeding the threshold value or as a cubic filter based on the distance being less than or equal to the threshold value,
wherein the distance is derived based on the prediction mode of the current block, and the threshold value is derived based on the size of the current block.
2. The image decoding method of claim 1 , wherein the deriving of the interpolation filter includes:
comparing a prediction angle between a direction of the prediction mode of the current block and a vertical direction or a horizontal direction with a preset threshold value; and
deriving the interpolation filter as either a first interpolation filter or the second interpolation filter depending on a comparison result of the prediction angle and the preset threshold value.
3. The image decoding method of claim 2 , wherein the preset threshold value is set differently depending on the size of the current block.
4. The image decoding method of claim 2 , wherein:
the interpolation filter is derived as the first interpolation filter based on the prediction angle being exceed the threshold value;
the interpolation filter is derived as the second interpolation filter based on the prediction angle being less than or equal to the threshold value; and
the first interpolation filter is a Gaussian filter, and the second interpolation filter is a cubic filter.
5. An image encoding method performed by an encoding apparatus, the method comprising:
deriving an intra prediction mode for a current block;
deriving reference samples contiguous to the current block;
generating a prediction sample for the current block based on the intra prediction mode and the reference samples;
generating residual samples for the current block based on the prediction sample; and
encoding image information including information on the intra prediction mode and the residual sample,
wherein the generating of the prediction sample for the current block includes:
deriving an interpolation filter to be applied to the reference sample based on a prediction mode of the current block or a size of the current block; and
generating the prediction sample by applying the derived interpolation filter to the reference sample,
wherein the deriving of the interpolation filter includes:
comparing a distance between the prediction sample and the reference sample with a predetermined threshold value; and
deriving the interpolation filter as a Gaussian filter based on the distance exceeding the threshold value or as a cubic filter based on the distance being less than or equal to the threshold value,
wherein the distance is derived based on the prediction mode of the current block, and the threshold value is derived based on the size of the current block.
6. The image encoding method of claim 5 , wherein the deriving of the interpolation filter includes:
comparing a prediction angle between a direction of the prediction mode of the current block and a vertical direction or a horizontal direction with a preset threshold value; and
deriving the interpolation filter as either a first interpolation filter or the second interpolation filter depending on a comparison result of the prediction angle and the preset threshold value.
7. The image encoding method of claim 6 , wherein the preset threshold value is set differently depending on the size of the current block.
8. The image encoding method of claim 6 , wherein:
the interpolation filter is derived as the first interpolation filter based on the prediction angle being exceed the threshold value;
the interpolation filter is derived as the second interpolation filter based on the prediction angle being less than or equal to the threshold value; and
the first interpolation filter is a Gaussian filter, and the second interpolation filter is a cubic filter.
9. A non-transitory computer-readable storage medium storing a bitstream generated by a method, the method comprising:
deriving an intra prediction mode for a current block;
deriving reference samples contiguous to the current block;
generating a prediction sample for the current block based on the intra prediction mode and the reference samples;
generating residual samples for the current block based on the prediction sample; and
encoding image information including information on the intra prediction mode and the residual sample,
wherein the generating of the prediction sample for the current block includes:
deriving an interpolation filter to be applied to the reference sample based on a prediction mode of the current block or a size of the current block; and
generating the prediction sample by applying the derived interpolation filter to the reference sample,
wherein the deriving of the interpolation filter includes:
comparing a distance between the prediction sample and the reference sample with a predetermined threshold value; and
deriving the interpolation filter as a Gaussian filter based on the distance exceeding the threshold value or as a cubic filter based on the distance being less than or equal to the threshold value,
wherein the distance is derived based on the prediction mode of the current block, and the threshold value is derived based on the size of the current block.