Method and device for filtering
Disclosed herein are a video decoding method and apparatus and a video encoding method and apparatus, and more particularly, a method and apparatus for performing filtering in video encoding and decoding. An encoding apparatus may perform filtering on a target, and may generate filtering information indicating whether filtering has been performed on the target. Further, the encoding apparatus may generate a bitstream including filtering information. A decoding apparatus may determine, based on filtering information, whether to perform filtering on a target, and may perform filtering on the target. The decoding apparatus may receive filtering information from the encoding apparatus through a bitstream or may derive filtering information using additional information.
1. A video decoding method, comprising:
determining whether to apply a filtering for a prediction block generated by an inter prediction for a target block;
selecting a filter for the prediction block generated by the inter prediction among a plurality of interpolation filters in a case that it is determined to apply the filtering for the inter prediction; and
generating the prediction block using the filter by performing the inter prediction for the target block in a case that it is determined to apply the filtering for the prediction block, wherein:
one of the plurality of the interpolation filters for the inter prediction is selected as the filter based on a size of the target block and based on whether a merge mode is used for the inter prediction for the target block or not,
the filter comprises a horizontal filter and a vertical filter for the inter prediction,
each of the horizontal filter and the vertical filter for the inter prediction is determined based on the size of the target block and whether the merge mode is used for the inter prediction for the target block or not,
a filter tap and filter coefficients of the horizontal filter for the inter prediction is determined based on a plurality of interpolation filters comprising an 8-tap filter based on the size of the target block, and
a filter tap and filter coefficients of the vertical filter for the inter prediction is determined based on a plurality of interpolation filters comprising an 8-tap filter based on the size of the target block.
2. The video decoding method of claim 1 , wherein the filter is used for an interpolation for a 1/16 sub-reference sample for the target block, and
the 1/16 sub-reference sample is generated by the filter applied to one or more integer-reference samples.
3. The video decoding method of claim 1 , wherein filter taps and filter coefficients of the plurality of the interpolation filters are different, and
wherein a filter tap and filter coefficients used for the filtering are determined based on the size of the target block.
4. The video decoding method of claim 1 , wherein whether the horizontal filter is an 8-tap filter or not and whether the vertical filter is an 8-tap filter or not are determined based on the size of the target block.
5. The video decoding method of claim 1 , wherein the horizontal filter and the vertical filter are determined based on whether the size of the target block is 4×4 or not.
6. The video decoding method of claim 1 , wherein all pixels in the target block are generated based on both of the horizontal filter and the vertical filter.
7. A video encoding method, comprising:
determining whether to apply a filtering for a prediction block generated by an inter prediction for a target block;
selecting a filter for the prediction block generated by the inter prediction among a plurality of interpolation filters in a case that it is determined to apply the filtering for the inter prediction; and
generating the prediction block using the filter by performing the inter prediction for the target block in a case that it is determined to apply the filtering for the prediction block, wherein:
one of the plurality of the interpolation filters for the inter prediction is selected as the filter based on a size of the target block and based on whether a merge mode is used for the inter prediction for the target block or not,
the filter comprises a horizontal filter and a vertical filter for the inter prediction,
each of the horizontal filter and the vertical filter for the inter prediction is determined based on the size of the target block and whether the merge mode is used for the inter prediction for the target block or not,
a filter tap and filter coefficients of the horizontal filter for the inter prediction is determined based on a plurality of interpolation filters comprising an 8-tap filter based on the size of the target block, and
a filter tap and filter coefficients of the vertical filter for the inter prediction is determined based on a plurality of interpolation filters comprising an 8-tap filter based on the size of the target block.
8. The video encoding method of claim 7 , wherein the selected filter is used for an interpolation for a 1/16 sub-reference sample for the target block, and
the 1/16 sub-reference sample is generated by the filter applied to one or more integer-reference samples.
9. The video encoding method of claim 7 , wherein filter taps and filter coefficients of the plurality of the interpolation filters are different, and
wherein a filter tap and filter coefficients used for the filtering are determined based on the size of the target block.
10. The video encoding method of claim 7 , wherein whether the horizontal filter is an 8-tap filter or not and whether the vertical filter is an 8-tap filter or not are determined based on the size of the target block.
11. The video encoding method of claim 7 , wherein the horizontal filter and the vertical filter are determined based on whether the size of the target block is 4×4 or not.
12. The video encoding method of claim 7 , wherein all pixels in the target block are generated based on both of the horizontal filter and the vertical filter.
13. A non-transitory computer-readable medium storing a bitstream generated by the video encoding method of claim 7 .
14. A non-transitory computer-readable medium storing a bitstream, the bitstream comprising:
encoded information for a target block;
wherein:
decoding for the target block is performed using the encoded information,
whether to apply a filtering using a filter for a prediction block generated by an inter prediction for the target block is determined,
a filter for the prediction block generated by the inter prediction is selected among a plurality of interpolation filters for the inter prediction in a case that it is determined to apply the filtering for the inter prediction, and
the prediction block using the filter is generated by performing the inter prediction for the target block using the filter in a case that it is determined to apply the filtering for the prediction block,
one of the plurality of the interpolation filters is selected as the filter based on a size of the target block and based on whether a merge mode is used for the inter prediction for the target block or not,
the filter comprises a horizontal filter and a vertical filter for the inter prediction,
each of the horizontal filter and the vertical filter for the inter prediction is determined based on the size of the target block and whether the merge mode is used for the inter prediction for the target block or not,
a filter tap and filter coefficients of the horizontal filter for the inter prediction is determined based on a plurality of interpolation filters comprising an 8-tap filter based on the size of the target block, and
a filter tap and filter coefficients of the vertical filter for the inter prediction is determined based on a plurality of interpolation filters comprising an 8-tap filter based on the size of the target block.
15. The non-transitory computer-readable medium of claim 14 , wherein the filter is used for an interpolation for a 1/16 sub-reference sample for the target block, and
the 1/16 sub-reference sample is generated by the filter applied to one or more integer-reference samples.
16. The non-transitory computer-readable medium of claim 14 , wherein filter taps and filter coefficients of the plurality of the interpolation filters are different, and
a filter tap and filter coefficients used for the filtering are determined based on the size of the target block.
17. The non-transitory computer-readable medium of claim 14 , wherein the filter is determined based on information about an affine motion compensation for the inter prediction.
18. The non-transitory computer-readable medium of claim 14 , wherein whether the horizontal filter is an 8-tap filter or not and whether the vertical filter is an 8-tap filter or not are determined based on the size of the target block.
19. The non-transitory computer-readable medium of claim 14 , wherein the horizontal filter and the vertical filter are determined based on whether the size of the target block is 4×4 or not.
20. The non-transitory computer-readable medium of claim 14 , wherein all pixels in the target block are generated based on both of the horizontal filter and the vertical filter.