Intra prediction-based video signal processing method and device
Disclosed are a video signal processing method and device whereby a video signal is encoded or decoded. The video signal processing method may comprise the steps of: determining whether an intra sub-partition (ISP) mode is applicable to a current block; if the ISP mode is applicable to the current block, splitting the current block into a plurality of rectangular transform blocks in the horizontal or vertical direction; generating prediction blocks of the transform blocks by carrying out intra prediction on each of the transform blocks; and reconstructing the current block on the basis of residual blocks of the transform blocks, and the prediction blocks.
1. A device for decoding a video signal, the device comprising a processor,
wherein the processor is configured to:
when an intra sub-partition (ISP) mode is applied to a current block, obtain a plurality of transform blocks by dividing the current block based on a horizontal direction or a vertical direction,
wherein the plurality of transform blocks include a first transform block and a second transform block,
generate a first prediction block of the first transform block and a second prediction block of the second transform block, and
reconstruct the current block based on a first residual block, a second residual block, the first prediction block, and the second prediction block,
wherein the first residual block and the second residual block are obtained by performing a low frequency non-separable transform on each of the first transform block and the second transform block,
wherein the first prediction block and the second prediction block are generated by applying position-dependent intra prediction sample filtering.
2. The device of claim 1 ,
wherein whether the position-dependent intra prediction sample filtering is applied based on at least one of a width and a height of each of the transform blocks.
3. The device of claim 2 ,
wherein the width of each of the transform blocks is equal to or larger than a pre-configured reference value, and the height of each of the transform blocks is equal to or larger than the pre-configured reference value.
4. The device of claim 1 ,
wherein the position-dependent intra prediction sample filtering is applied by weighted combination of reference samples of each of the transform blocks.
5. The device of claim 1 ,
wherein a transform kernel for the low frequency non-separable transform is determined based on an intra prediction mode of the current block.
6. A device for encoding a video signal, the device comprising a processor,
wherein the processor is configured to
obtain a bitstream to be decoded by a decoder using a decoding method,
wherein the decoding method comprises:
when an intra sub-partition (ISP) mode is applied to a current block, obtaining a plurality of transform blocks by dividing the current block based on a horizontal direction or a vertical direction,
wherein the plurality of transform blocks include a first transform block and a second transform block;
generating a first prediction block of the first transform block and a second prediction block of the second transform block; and
reconstructing the current block based on a first residual block, a second residual block, the first prediction block, and the second prediction block,
wherein the first residual block and the second residual block are obtained by performing a low frequency non-separable transform on each of the first transform block and the second transform block,
wherein the first prediction block and the second prediction block are generated by applying position-dependent intra prediction sample filtering.
7. The device of claim 6 ,
wherein whether the position-dependent intra prediction sample filtering is applied based on at least one of a width and a height of each of the transform blocks.
8. The device of claim 7 ,
wherein the width of each of the transform blocks is equal to or larger than a pre-configured reference value, and the height of each of the transform blocks is equal to or larger than the pre-configured reference value.
9. The device of claim 6 ,
wherein the position-dependent intra prediction sample filtering is applied by weighted combination of reference samples of each of the transform blocks.
10. The device of claim 6 ,
wherein a transform kernel for the low frequency non-separable transform is determined based on an intra prediction mode of the current block.
11. A method obtaining a bitstream,
wherein the method, comprises:
when an intra sub-partition (ISP) mode is applied to a current block, obtaining a plurality of transform blocks by dividing the current block based on a horizontal direction or a vertical direction,
wherein the plurality of transform blocks include a first transform block and a second transform block;
generating a first prediction block of the first transform block and a second prediction block of the second transform block;
reconstructing the current block based on a first residual block, a second residual block, the first prediction block, and the second prediction block; and
obtaining the bitstream including information for reconstructing the block,
wherein the first residual block and the second residual block are obtained by performing a low frequency non-separable transform on each of the first transform block and the second transform block,
wherein the first prediction block and the second prediction block are generated by applying position-dependent intra prediction sample filtering.
12. The method of claim 11 ,
wherein whether the position-dependent intra prediction sample filtering is applied based on at least one of a width and a height of each of the transform blocks.
13. The method of claim 12 ,
wherein the width of each of the transform blocks is equal to or larger than a pre-configured reference value, and the height of each of the transform blocks is equal to or larger than the pre-configured reference value.
14. The method of claim 11 ,
wherein the position-dependent intra prediction sample filtering is applied by weighted combination of reference samples of each of the transform blocks.
15. The method of claim 11 ,
wherein a transform kernel for the low frequency non-separable transform is determined based on an intra prediction mode of the current block.