Image encoding/decoding method and device therefor

Patent metadata

Field	Value
Publication number	US-2024114159-A1
Application number	US-202318386135-A
Country	US
Kind code	A1
Filing date	Nov 1, 2023
Priority date	Aug 8, 2018
Publication date	Apr 4, 2024
Grant date	—

Abstract

The present invention relates to a video signal decoding method based on a Multiple Transform Selection (MTS). The method may comprise the steps of: parsing a first syntax element representing whether MTS applies to the inverse transformation of a current block, wherein the MTS represents a transform mode which uses a transform type other than a default transform type predefined for the current block; by performing inverse quantization on the current block, deriving an inverse-quantized transform coefficient array having the width and the height of the current block; determining, on the basis of the first syntax element, a vertical transform type applying to the vertical direction of the current block, and a horizontal transform type applying to the horizontal direction of the current block; and, by performing inverse transformation on the inverse-quantized transform coefficient array by using the vertical transform type and the horizontal transform type, deriving a residual sample array having the width and the height of the current block.

First claim

1 - 12 . (canceled)

13 . A method of decoding a video signal based on a Multiple Transform Selection (MTS) by an apparatus, comprising:
obtaining syntax elements from the video signal, wherein the syntax elements include a first syntax element indicating whether an MTS index could be present in a syntax for an intra coding unit, a second syntax element indicating whether the MTS index could be present in a syntax for an inter coding unit, and a third syntax element corresponding to one of a plurality of transform combinations within a transform configuration group, wherein the first and second syntax elements are included in a sequence parameter set (SPS) of the video signal, and wherein the MTS represents a transform method performed based on a transform combination selected from the plurality of transform combinations; deriving an inverse quantized transform coefficient array with a width and a height of a current block by performing inverse quantization on the current block; determining a vertical transform type applied to a vertical direction and a horizontal transform type applied to a horizontal direction of the current block based on the third syntax element; and deriving a residual sample array with the width and the height of the current block by performing inverse transform on the inverse quantized transform coefficient array based on the vertical transform type and the horizontal transform type, wherein deriving the residual sample array includes: performing, based on the vertical transform type, a one-dimensional transform process in the vertical direction on each of columns of the inverse quantized transform coefficient array; and performing, based on the horizontal transform type, a one-dimensional transform process in the horizontal direction on each of rows of an intermediate sample array output by the one-dimensional transform process for each of the columns, wherein performing the one-dimensional transform process in the horizontal direction includes clipping an intermediate sample value output by the one-dimensional transform process for each of the columns based on a minimum value and a maximum value of a predefined coefficient.

14 . The method of claim 13 ,
wherein the plurality of transform combinations includes at least one of DCT2, DST7, or DCT8.

15 . The method of claim 13 ,
wherein based on the third syntax element being equal to a value of zero, the vertical transform type and the horizontal transform type are determined as DCT2, and wherein based on the third syntax element being equal to a value of non-zero, each of the vertical transform type and the horizontal transform type is determined as one of DST7 or DCT8.

16 . A method of encoding a video signal based on a Multiple Transform Selection (MTS) by an apparatus, comprising:
generating a residual sample array of a current block; determining a vertical transform type applied to a vertical direction of the current block and a horizontal transform type applied to a horizontal direction of the current block; performing transform on the current block based on the vertical transform type and horizontal transform type; deriving a quantized transform coefficient array with a width and a height of the current block by performing quantization on the current block; and entropy-encoding the quantized transform coefficient array of the current block, wherein the MTS represents a transform method performed based on a transform combination selected from a plurality of transform combinations within a transform configuration group, wherein syntax elements for the MTS are encoded into a bitstream, wherein the syntax elements include a first syntax element indicating whether an MTS index could be present in a syntax for an intra coding unit, a second syntax element indicating whether the MTS index could be present in a syntax for an inter coding unit, and a third syntax element corresponding to one of the plurality of transform combinations, wherein the first and second syntax elements are included in a sequence parameter set (SPS) of the bitstream, wherein performing the transform on the current block includes: performing, based on the horizontal transform type, a one-dimensional transform process in the horizontal direction on each of rows of the residual sample array; and performing, based on the vertical transform type, a one-dimensional transform process in the vertical direction on each of columns of an intermediate sample array output by the one-dimensional transform process for each of the rows of the residual sample array.

17 . The method of claim 16 ,
wherein the plurality of transform combinations includes at least one of DCT2, DST7, or DCT8.

18 . The method of claim 16 ,
wherein the third syntax element is encoded as a value of zero based on the vertical transform type and the horizontal transform type being determined as DCT2, and wherein the third syntax element is encoded as a value of non-zero based on each of the vertical transform type and the horizontal transform type being determined as one of DST7 or DCT8.

19 . A transmission method of data for an image, comprising:
obtaining a bitstream for the image, wherein the bitstream is generated by: generating a residual sample array of a current block, determining a vertical transform type applied to a vertical direction of the current block and a horizontal transform type applied to a horizontal direction of the current block, performing transform on the current block based on the vertical transform type and horizontal transform type, deriving a quantized transform coefficient array with a width and a height of the current block by performing quantization on the current block, and entropy-encoding the quantized transform coefficient array of the current block; and transmitting the data comprising the bitstream, wherein the MTS represents a transform method performed based on a transform combination selected from a plurality of transform combinations within a transform configuration group, wherein syntax elements for the MTS are encoded into a bitstream, wherein the syntax elements include a first syntax element indicating whether an MTS index could be present in a syntax for an intra coding unit, a second syntax element indicating whether the MTS index could be present in a syntax for an inter coding unit, and a third syntax element corresponding to one of the plurality of transform combinations, wherein the first and second syntax elements are included in a sequence parameter set (SPS) of the bitstream, wherein performing the transform on the current block includes: performing, based on the horizontal transform type, a one-dimensional transform process in the horizontal direction on each of rows of the residual sample array; and performing, based on the vertical transform type, a one-dimensional transform process in the vertical direction on each of columns of an intermediate sample array output by the one-dimensional transform process for each of the rows of the residual sample array.

Assignees

• Lg Electronics Inc

Inventors

• Koo Moonmo
• Salehifar Mehdi
• Kim Seunghwan
• Lim Jaehyun

Classifications

• H04N19/18

  the unit being a set of transform coefficients · CPC title

• H04N19/61

  in combination with predictive coding · CPC title

• H04N19/12 Primary

  Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264 · CPC title

• H04N19/44 Primary

  Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder · CPC title

• H04N19/132

  Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking · CPC title