Transform-based image coding method and device for same
An image decoding method according to the present disclosure comprises the steps of: deriving residual samples with respect to a target block on the basis of an inverse primary transform with respect to modified transform coefficients; and on the basis of the residual samples, generating a reconstructed picture, wherein a step of deriving the modified transform coefficients may include the steps of: parsing an LFNST index on the basis of the tree-type of a current block not being a dual-tree chroma and a MIP mode not being applicable to the current block; and deriving the modified transform coefficients on the basis of the LFNST index and an LFNST matrix.
1. An image decoding method performed by a decoding apparatus, comprising:
deriving transform coefficients for a current block based on residual information;
deriving modified transform coefficients by applying a low frequency non-separable transform (LFNST) to the transform coefficients;
deriving residual samples for the current block based on an inverse primary transform for the modified transform coefficients; and
generating a reconstructed picture based on the residual samples,
wherein deriving the modified transform coefficients comprises:
obtaining an LFNST index from a bitstream; and
deriving the modified transform coefficients based on the LFNST index and an LFNST matrix,
wherein when a tree type of the current block is a dual-tree chroma, the LFNST index is obtained from the bitstream regardless of whether a matrix-based intra prediction (MIP) mode is applied to the current block and whether an LFNST-applied width and an LFNST-applied height corresponding to the current block are greater than or equal to 16, and
wherein when the tree type of the current block is not the dual-tree chroma, the LFNST index is obtained from the bitstream based on whether the MIP mode is applied to the current block or whether the LFNST-applied width and the LENST-applied height corresponding to the current block are greater than or equal to 16.
2. The image decoding method of claim 1 , wherein when the tree type of the current block is not the dual-tree chroma and the MIP mode is applied to the current block, the LFNST index is obtained from the bitstream based on the LFNST-applied width and the LFNST-applied height corresponding to the current block being greater than or equal to 16.
3. The image decoding method of claim 2 , wherein the LFNST-applied width is set to a width of the current block divided by a number of split sub-partitions when the current block is vertically split in the sub-partitions, and the LFNST-applied width is set to the width of the current block when the current block is not split into the sub-partitions, and
wherein the current block is a coding block.
4. The image decoding method of claim 2 , wherein the LFNST-applied height is set to a height of the current block divided by a number of sub-partitions when the current block is horizontally split into the sub-partitions, and the LFNST-applied height is set to the height of the current block when the current block is not split into the sub-partitions, and
wherein the current block is a coding block.
5. An image encoding method performed by an image encoding apparatus, the method comprising:
deriving prediction samples for a current block;
deriving residual samples for the current block based on the prediction samples;
deriving transform coefficients for the current block based on a primary transform of the residual samples;
deriving modified transform coefficients by applying a low frequency non-separable transform (LFNST) to the transform coefficients;
deriving quantized transform coefficients by performing a quantization on the modified transform coefficients; and
encoding the quantized transform coefficients and an LFNST index indicating an LFNST matrix applied to the LFNST,
wherein when a tree type of the current block is a dual-tree chroma, the LFNST index is encoded into a bitstream regardless of whether a matrix-based intra prediction (MIP) mode is applied to the current block and whether an LFNST-applied width and an LFNST-applied height corresponding to the current block are greater than or equal to 16, and
wherein when the tree type of the current block is not the dual-tree chroma, the LFNST index is encoded into the bitstream based on whether the MIP mode is applied to the current block or whether the LFNST-applied width and the LFNST-applied height corresponding to the current block are greater than or equal to 16.
6. The image encoding method of claim 5 , wherein when the tree type of the current block is not the dual-tree chroma and the MIP mode is applied to the current block, the LFNST index is encoded into the bitstream based on the LFNST-applied width and the LFNST-applied height corresponding to the current block being greater than or equal to 16.
7. The image encoding method of claim 6 , wherein the LFNST-applied width is set to a width of the current block divided by a number of sub-partitions when the current block is vertically split into the sub-partitions, and the LFNST-applied width is set to the width of the current block when the current block is not split into the sub-partitions, and
wherein the current block is a coding block.
8. The image encoding method of claim 6 , wherein the LFNST-applied height is set to a height of the current block divided by a number of sub-partitions when the current block is horizontally split into the sub-partitions, and the LFNST-applied height is set to the height of the current block when the current block is not split into the sub-partitions, and
wherein the current block is a coding block.
9. A non-transitory computer-readable digital storage medium that stores a bitstream generated by a method, the method comprising:
deriving prediction samples for a current block;
deriving residual samples for the current block based on the prediction samples;
deriving transform coefficients for the current block based on a primary transform of the residual samples;
deriving modified transform coefficients by applying a low frequency non-separable transform (LFNST) to the transform coefficients;
deriving quantized transform coefficients by performing a quantization on the modified transform coefficients; and
encoding the quantized transform coefficients and an LFNST index indicating an LFNST matrix applied to the LFNST to generate the bitstream,
wherein when a tree type of the current block is a dual-tree chroma, the LFNST index is encoded into the bitstream regardless of whether a matrix-based intra prediction (MIP) mode is applied to the current block and whether an LFNST-applied width and an LFNST-applied height corresponding to the current block are greater than or equal to 16, and
wherein when the tree type of the current block is not the dual-tree chroma, the LENST index is encoded into the bitstream based on whether the MIP mode is applied to the current block or whether the LFNST-applied width and the LFNST-applied height corresponding to the current block are greater than or equal to 16.