Video signal processing method and apparatus using multiple transform kernels
A video signal processor is configured to: obtain at least one transform block for a residual signal of a current block from a video signal bitstream, wherein the transform block comprises a plurality of transform coefficients two-dimensionally arranged, determine, on the basis of length information of a first side of the transform block, a horizontal transform kernel for horizontal transformation of the transform block, regardless of a length of a second side of the transform block, which is orthogonal to the first side, determine, on the basis of length information of the second side, a vertical transform kernel for vertical transformation of the transform block, regardless of a length of the first side, obtain the residual signal of the current block by performing, on the transform block, inverse transformation using the horizontal transform kernel and the vertical transform kernel, and reconstruct the current block based on the residual signal.
1. A video decoding method, the method performed by an apparatus and comprising:
obtaining flag information indicating whether index information indicating a horizontal transform kernel and a vertical transform kernel is signaled for an intra predicted block in a bitstream;
obtaining prediction mode information indicating whether a current block is the intra predicted block;
obtaining a plurality of transform coefficients for residual samples of the current block, wherein the plurality of transform coefficients are two-dimensionally arranged;
based on the flag information indicating that the index information is not signaled for the intra predicted block in the bitstream and the prediction mode information indicating that the current block is the intra predicted block:
determining, based on length information of a first side of the current block, the horizontal transform kernel for horizontal transformation of the plurality of transform coefficients, regardless of a length of a second side of the current block, wherein the second side is orthogonal to the first side, and
determining, based on length information of the second side of the current block, the vertical transform kernel for vertical transformation of the plurality of transform coefficients, regardless of a length of the first side of the current block;
obtaining the residual samples of the current block by performing, based on the plurality of transform coefficients, inverse transformation using the horizontal transform kernel and the vertical transform kernel; and
reconstructing the current block based on the obtained residual samples,
wherein, based on the length information of the second side having a value outside a preconfigured range and the length information of the first side having a value within the preconfigured range, the vertical transform kernel is determined to be a first transform kernel and the horizontal transform kernel is determined to be a second transform kernel, and
wherein the first transform kernel is a discrete cosine transform type 2 (DCT-2)-based transform kernel, and the second transform kernel is not the DCT-2-based transform kernel.
2. The method of claim 1 ,
wherein the second transform kernel is a discrete sine transform type 7 (DST-7)-based transform kernel.
3. The method of claim 1 ,
wherein the length of the first side is a width of the current block, and the length of the second side is a height of the current block.
4. The method of claim 1 ,
wherein the inverse transformation using the horizontal transform kernel and the vertical transform kernel comprises first inverse transformation using the vertical transform kernel and second inverse transformation using the horizontal transform kernel.
5. The method of claim 4 ,
wherein the first inverse transformation is performed prior to the second inverse transformation.
6. A video encoding method, the method performed by an apparatus and comprising:
obtaining a bitstream to be decoded by a decoder using a decoding method,
wherein the decoding method comprises:
obtaining flag information indicating whether index information indicating a horizontal transform kernel and a vertical transform kernel is signaled for an intra predicted block in the bitstream;
obtaining prediction mode information indicating whether a current block is the intra predicted block;
obtaining a plurality of transform coefficients for residual samples of the current block, wherein the plurality of transform coefficients are two-dimensionally arranged;
based on the flag information indicating that the index information is not signaled for the intra predicted block in the bitstream and the prediction mode information indicating that the current block is the intra predicted block:
determining, based on length information of a first side of the current block, the horizontal transform kernel for horizontal transformation of the plurality of transform coefficients, regardless of a length of a second side of the current block, wherein the second side is orthogonal to the first side, and
determining, based on length information of the second side of the current block, the vertical transform kernel for vertical transformation of the plurality of transform coefficients, regardless of a length of the first side of the current block;
obtaining the residual samples of the current block by performing, based on the plurality of transform coefficients, inverse transformation using the horizontal transform kernel and the vertical transform kernel; and
reconstructing the current block based on the obtained residual samples,
wherein, based on the length information of the second side having a value outside a preconfigured range and the length information of the first side having a value within the preconfigured range, the vertical transform kernel is determined to be a first transform kernel and the horizontal transform kernel is determined to be a second transform kernel, and
wherein the first transform kernel is a discrete cosine transform type 2 (DCT-2)-based transform kernel, and the second transform kernel is not the DCT-2-based transform kernel.
7. The method of claim 6 ,
wherein the second transform kernel is a discrete sine transform type 7 (DST-7)-based transform kernel.
8. The method of claim 6 ,
wherein the length of the first side is a width of the current block, and the length of the second side is a height of the current block.
9. The method of claim 6 ,
wherein the inverse transformation using the horizontal transform kernel and the vertical transform kernel comprises first inverse transformation using the vertical transform kernel and second inverse transformation using the horizontal transform kernel.
10. The method of claim 9 ,
wherein the first inverse transformation is performed prior to the second inverse transformation.
11. A non-transitory computer-readable medium storing a bitstream, the bitstream being decoded by a decoding method,
wherein the decoding method comprises:
obtaining flag information indicating whether index information indicating a horizontal transform kernel and a vertical transform kernel is signaled for an intra predicted block in the bitstream;
obtaining prediction mode information indicating whether a current block is the intra predicted block;
obtaining a plurality of transform coefficients for residual samples of the current block, wherein the plurality of transform coefficients are two-dimensionally arranged;
based on the flag information indicating that the index information is not signaled for the intra predicted block in the bitstream and the prediction mode information indicating that the current block is the intra predicted block:
determining, based on length information of a first side of the current block, the horizontal transform kernel for horizontal transformation of the plurality of transform coefficients, regardless of a length of a second side of the current block, wherein the second side is orthogonal to the first side, and
determining, based on length information of the second side of the current block, the vertical transform kernel for vertical transformation of the plurality of transform coefficients, regardless of a length of the first side of the current block;
obtaining the residual samples of the current block by performing, based on the plurality of transform coefficients, inverse transformation using the horizontal transform kernel and the vertical transform kernel; and
reconstructing the current block based on the obtained residual samples,
wherein, based on the length information of the second side having a value outside a preconfigured range and the length information of the first side having a value within the preconfigured range, the vertical transform kernel is determined to be a first transform kernel and the horizontal transform kernel is determined to be a second transform kernel, and
wherein the first transform kernel is a discrete cosine transform type 2 (DCT-2)-based transform kernel, and the second transform kernel is not the DCT-2-based transform kernel.