Method and device for processing video signal by using reduced secondary transform

1. A method of reconstructing a video signal based on a reduced secondary transform, comprising:

obtaining, based on zero transform coefficients being at specific transform coefficient locations, a secondary transform index from the video signal;

deriving an inverse secondary transform matrix corresponding to the secondary transform index, wherein the inverse secondary transform matrix represents the reduced secondary transform outputting N residual data (N×1 residual vectors) based on inputted L (L<N) transform coefficient data (L×1 transform coefficient vectors);

obtaining non-zero transform coefficients by performing an entropy decoding and a dequantization for a current block;

performing an inverse secondary transform for the non-zero transform coefficients based on the inverse secondary transform matrix;

performing an inverse primary transform for a block to which the inverse secondary transform is applied; and

reconstructing the current block based on a block to which the inverse primary transform is applied,

wherein the specific transform coefficient locations are arranged in a diagonal scan order from a (L+1) th location to a 16 th location in the transform coefficients of the block to which the inverse secondary transform is applied, the block being a 4×4 block, N being 16, and L being 8.

2. The method of claim 1 , wherein the inverse secondary transform is applied to a specific region of the current block, and

wherein the specific region is a left-top region in the current block.

3. The method of claim 1 , wherein based on the inverse secondary transform being performed, a 4×4 inverse secondary transform is applied to each of divided 4×4 blocks in the current block.

4. The method of claim 1 , wherein based on the last non-zero transform coefficient is not located in a specific region, the secondary transform index is obtained, and

wherein the specific region represents a remaining region other than a position where a non-zero transform coefficient may exist based on transform coefficients are arranged according to a scan order if the inverse secondary transform is applied.

5. A method of encoding a video signal based on a reduced secondary transform, comprising:

performing a primary transform on a transform block;

deriving a secondary transform matrix to be applied to the transform block, wherein the secondary transform matrix represents a reduced secondary transform outputting L (L<N) transform coefficient data (L×1 transform coefficient vectors) based on inputted N residual data (N×1 residual vectors);

performing a secondary transform on the transform block based on the secondary transform matrix;

obtaining transform coefficients based on the secondary transform;

performing a quantization and an entropy encoding for the transform coefficients; and

generating a secondary transform index corresponding to the secondary transform matrix,

wherein zero transform coefficients are at specific transform coefficient locations, and

wherein the specific transform coefficient locations are arranged in a diagonal scan order from a (L+1) th location to a 16 th location in the transform coefficients of the block to which the secondary transform is applied, the block being a 4×4 block, N being 16, and L being 8.

6. A non-transitory computer-readable medium storing video information generated by an encoding method, the encoding method comprising:

performing a primary transform on a transform block;

deriving a secondary transform matrix to be applied to the transform block, wherein the secondary transform matrix represents a reduced secondary transform outputting L (L<N) transform coefficient data (L×1 transform coefficient vectors) based on inputted N residual data (N×1 residual vectors);

performing a secondary transform on the transform block based on the secondary transform matrix;

obtaining transform coefficients based on the secondary transform;

performing a quantization and an entropy encoding for the transform coefficients; and

generating a secondary transform index corresponding to the secondary transform matrix,

wherein zero transform coefficients are at specific transform coefficient locations, and

wherein the specific transform coefficient locations are arranged in a diagonal scan order from a (L+1) th location to a 16 th location in the transform coefficients of the block to which the secondary transform is applied, the block being a 4×4 block, N being 16, and L being 8.

7. The method of claim 1 , further comprising:

obtaining a primary transform index of the current block from the video signal, wherein the primary transform index corresponds to any one of a plurality of transform combinations configured by a combination of Discrete Sine Transform type 7 (DST7) and/or Discrete Cosine Transform type 8 (DCT8); and

deriving a transform combination corresponding to the primary transform index,

wherein the transform combination includes a horizontal transform and a vertical transform, and the horizontal transform and the vertical transform correspond to any one of the DST7 or the DCT8, and

wherein the inverse primary transform is performed based on the transform combination.