Method and device for processing a video signal by using an adaptive separable graph-based transform

Patent metadata

Field	Value
Publication number	US-10567763-B2
Application number	US-201615577029-A
Country	US
Kind code	B2
Filing date	May 26, 2016
Priority date	May 26, 2015
Publication date	Feb 18, 2020
Grant date	Feb 18, 2020

Abstract

A method and a device for decoding a video signal using an adaptive separable graph-based transform. The method includes: receiving a transform index for a target block from the video signal in which the transform index indicates a graph-based transform to be applied to the target block; deriving a graph-based transform kernel corresponding to the transform index; and decoding the target block based on the graph-based transform kernel. The device includes: a parsing unit configured to receive a transform index for a target block from the video signal; and an inverse-transform unit configured to derive a graph-based transform kernel corresponding to the transform index and decode the target block based on the graph-based transform kernel.

First claim

The invention claimed is:

1. A method for decoding a video signal using an adaptive separable graph-based transform (GBT), the method comprising:
receiving, from the video signal, a transform index for a target block, the transform index indicating GBT to be applied to the target block;
deriving a 2-dimensional (2D) separable GBT kernel corresponding to the transform index; and
decoding the target block based on the 2D separable GBT kernel,
wherein when the target block is comprised of a plurality of subblocks partitioned in a horizontal or vertical direction, the transform index corresponds to each of the subblocks,
wherein the 2D separable GBT kernel is generated based on a combination of at least two 1-dimensional (1D) GBTs,
wherein the 1D GBTs are generated for each line graph corresponding to a row direction and a column direction of each of the subblocks,
wherein transform types of the 1D GBTs are determined based on graph types of the line graphs, edge weights of the line graphs, and self-loop weights of the line graphs,
wherein a direction of energy concentration in the subblocks is determined by a combination of the transform types for each of the row direction and column direction, and
wherein the edge weights of the line graphs are given based on prediction direction of the target block.

2. The method of claim 1 , wherein
the transform types include at least two of DCT, asymmetric discrete sine transform (ADST), and reverse asymmetric discrete sine transform (RADST).

3. The method of claim 1 , wherein
the transform index is received in every unit of at least one of a coding unit, a prediction unit, and a transform unit.

4. A method for encoding a video signal using an adaptive separable graph-based transform (GBT), the method comprising:
partitioning a target block into a plurality of subblocks in a horizontal or vertical direction;
generating a 2-dimensional (2D) separable GBT kernel for each of the plurality of subblocks based on a combination of at least two 1-dimensional (1D) GBTs;
performing a transform on the plurality of subblocks using the 2D separable GBT kernel, and
encoding a transform index corresponding to the 2D separable GBT kernel,
wherein when the target block is comprised of the plurality of subblocks partitioned in a horizontal or vertical direction, the transform index corresponds to each of the subblocks,
wherein the 1D GBTs are generated for each line graph corresponding to a row direction and a column direction of each of the subblocks,
wherein transform types of the 1D GBTs are determined based on graph types of the line graphs, edge weights of the line graphs, and self-loop weights of the line graphs,
wherein a direction of energy concentration in the subblocks is determined by a combination of the transform types for each of the row direction and column direction, and
wherein the edge weights of the line graphs are given based on prediction direction of the target block.

5. The method of claim 4 , wherein
the transform types include at least two of DCT, ADST, and RADST.

6. A device for decoding a video signal using an adaptive separable graph-based transform (GBT), the device comprising:
a processor configured to:
receive, from the video signal, a transform index for a target block; and
derive a 2-dimensional (2D) separable GBT kernel corresponding to the transform index and decode the target block based on the 2D separable GBT kernel,
wherein when the target block is comprised of a plurality of subblocks partitioned in a horizontal or vertical direction, the transform index corresponds to each of the subblocks,
wherein when the target block is comprised of a plurality of subblocks partitioned in a horizontal or vertical direction, the transform index corresponds to each of the subblocks,
wherein the 2D separable GBT kernel is generated based on a combination of at least two 1-dimensional (1D) GBTs,
wherein the 1D GBTs are generated for each line graph corresponding to a row direction and a column direction of each of the subblocks,
wherein transform types of the 1D GBTs are determined based on graph types of the line graphs, edge weights of the line graphs, and self-loop weights of the line graphs,
wherein a direction of energy concentration in the subblocks is determined by a combination of the transform types for each of the row direction and column direction, and
wherein the edge weights of the line graphs are given based on prediction direction of the target block.

7. A device for encoding a video signal using an adaptive separable graph-based transform (GBT), the device comprising:
a processor configured to:
partition a target block into a plurality of subblocks in a horizontal or vertical direction;
generate a 2-dimensional (2D) separable GBT kernel for each of the plurality of subblocks based on a combination of at least two 1-dimensional (1D) GBTs; and
perform a transform on the plurality of subblocks using 2D separable GBT kernel,
wherein when the target block is comprised of the plurality of subblocks partitioned in a horizontal or vertical direction, a transform index corresponds to each of the subblocks,
wherein the 1D GBTs are generated for each line graph corresponding to a row direction and a column direction of each of the subblocks,
wherein transform types of the 1D GBTs are determined based on graph types of the line graphs, edge weights of the line graphs, and self-loop weights of the line graphs,
wherein a direction of energy concentration in the subblocks is determined by a combination of the transform types for each of the row direction and column direction, and
wherein the edge weights of the line graphs are given based on prediction direction of the target block.

Assignees

• Lg Electronics Inc
• Univ Southern California

Inventors

• Koo Moonmo
• Yea Sehoon
• Lee Bumshik
• Said Amir
• Egilmez Hilmi Enes
• Ortega Antonio

Classifications

• H04N19/109

  among a plurality of temporal predictive coding modes · CPC title

• H04N19/61

  in combination with predictive coding · CPC title

• H04N19/124

  Quantisation · CPC title

• H04N19/82

  involving filtering within a prediction loop · CPC title

• H04N19/176 Primary

  the region being a block, e.g. a macroblock · CPC title