IP Library › Granted Patent US 11,483,589
Granted Patent B2
US 11,483,589 · App. 17/394,734 · Granted Oct 25, 2022

Image coding method based on quadratic transform, and apparatus therefor

Inventors: Moonmo Koo (Seoul, KR); Jaehyun Lim (Seoul, KR); Seunghwan Kim (Seoul, KR)
Assignee: LG ELECTRONICS INC.
H04N19/61H04N19/119H04N19/124H04N19/132H04N19/159H04N19/176H04N19/18H04N19/46
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,483,589
App. No.
17/394,734
Granted
Oct 25, 2022
Kind
B2
Abstract

An image decoding method according to the present document comprises the steps of: receiving a quantized transform coefficient for a target block and a transform index for non-separable quadratic transform; deriving transform coefficients by inversely quantizing the quantized transform coefficient; deriving corrected transform coefficients on the basis of a transform kernel matrix in a predetermined transform set indicated by the transform index; and deriving residual samples for the target block on the basis of inverse linear transform for the corrected transform coefficients, wherein when the target block is divided into a predetermined number of sub-blocks and is coded by intra prediction, the corrected transform coefficients are derived in units of the sub-blocks, and the transform index is received for the target block.

Claims (43)

1. An image decoding method performed by a decoding apparatus, the method comprising:

receiving quantized transform coefficients and a transform index for a non-separable secondary transform for a coding block;

deriving transform coefficients by dequantizing the quantized transform coefficients;

deriving modified transform coefficients based on a transform kernel matrix related to the transform index in a transform set; and

deriving residual samples based on an inverse primary transform of the modified transform coefficients,

wherein the coding block is partitioned into a plurality of sub-blocks based on an Intra Sub Partitions (ISP) mode including horizontal and vertical mode,

wherein the coding block is partitioned into a specific number of sub-blocks based on a shape of a coding unit,

wherein the transform index is signaled in unit of the coding block and applied to the plurality of sub-blocks,

wherein the same transform kernel matrix is applied to the sub-blocks in the non-separable secondary transform, and

wherein the modified transform coefficients are derived in unit of the sub-block.

2. The image decoding method of claim 1 , wherein each sub-block is a transform unit.

3. The image decoding method of claim 1 , wherein the coding block is partitioned into two sub: blocks when the coding block has a size (width x height) of 8×4, and

wherein the coding block is partitioned into two sub: blocks when the coding block has a size (width x height) of 4×8.

4. The image decoding method of claim 3 , wherein the coding block is horizontally or vertically partitioned into four sub: blocks when the coding block has a size (width x height) greater than 4×8 or 8×4.

5. An image encoding method performed by an image encoding apparatus, the method comprising:

deriving prediction samples based on an intra prediction mode applied to a coding block;

deriving residual samples for the coding block based on the prediction samples;

deriving transform coefficients by applying a primary transform to the residual samples;

deriving modified transform coefficients by performing a non-separable secondary transform on the transform coefficients using a transform set mapped to the intra prediction mode of the coding block and one transform kernel matrix comprised in the transform set;

deriving quantized transform coefficients by performing quantization based on the modified transform coefficients; and

encoding a transform index related to the transform kernel matrix and the quantized transform coefficients,

wherein the coding block is partitioned into a plurality of sub-blocks based on an Intra Sub Partitions (ISP) mode including horizontal and vertical mode,

wherein the coding block is partitioned into a specific number of sub-blocks based on a shape of a coding unit,

wherein the transform index is encoded in a unit of the coding block and applied to the plurality of sub-blocks,

wherein the same transform kernel matrix is applied to the sub-blocks in the non-separable secondary transform, and

wherein the modified transform coefficients are derived in unit of sub-block.

6. The image encoding method of claim 5 , wherein the coding block is partitioned into two sub-blocks when the coding block has a size (width x height) of 8×4, and

wherein the coding block is partitioned into two sub: blocks when the coding block has a size (width x height) of 4×8.

7. The image encoding method of claim 5 , wherein the coding block is horizontally or vertically partitioned into four sub: blocks when the coding block has a size (width x height) greater than 4×8 or 8×4.

8. The image encoding method of claim 5 , further comprising:

encoding flag information indicating whether the ISP mode is applied to the coding block.

9. A non-transitory computer-readable digital storage medium storing a bitstream generated by a method, the method comprising:

deriving prediction samples based on an intra prediction mode applied to a coding block;

deriving residual samples for the coding block based on the prediction samples;

deriving transform coefficients by applying a primary transform to the residual samples;

deriving modified transform coefficients by performing a non-separable secondary transform on the transform coefficients using a transform set mapped to the intra prediction mode of the coding block and one transform kernel matrix comprised in the transform set;

deriving quantized transform coefficients by performing quantization based on the modified transform coefficients; and

encoding image information including a transform index related to the transform kernel matrix and the quantized transform coefficients to generate the bitstream,

wherein the coding block is partitioned into a plurality of sub-blocks based on an Intra Sub Partitions (ISP) mode including horizontal and vertical mode,

wherein the coding block is partitioned into a specific number of sub: blocks based on a shape of a coding unit,

wherein the transform index is encoded in a unit of the coding block and applied to the plurality of sub-blocks,

wherein the same transform kernel matrix is applied to the sub-blocks in the non-separable secondary transform, and

wherein the modified transform coefficients are derived in unit of sub-block.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 5, 2021
From: KOO, MOONMO; LIM, JAEHYUN; KIM, SEUNGHWAN
To: LG ELECTRONICS INC.
Reel/Frame 057092/0412 →
Continuity (3)
Continuation PCTKR2020002625 · Feb 24, 2020
Provisional Application 62809721 · Feb 24, 2019
Related Publication 20210377568A1 · Dec 2, 2021