IP Library Granted Patent US 11,632,546
Granted Patent B2
US 11,632,546 · App. 17/260,948 · Granted Apr 18, 2023

Method and device for effective video encoding/decoding via local lighting compensation

Inventors: Gun Bang (Daejeon, KR); Hui Yong Kim (Daejeon, KR); Dong Gyu Sim (Seoul, KR); Seoung Jun Oh (Seongnam-si, KR); Sea Nae Park (Seoul, KR); Jun Taek Park (Seoul, KR); Jong Seok Lee (Seoul, KR)
Assignee: Electronics and Telecommunications Research Institute
H04N19/117H04N19/105H04N19/159H04N19/167H04N19/176H04N19/186
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 11,632,546
App. No.
17/260,948
Granted
Apr 18, 2023
Kind
B2
Abstract

There is provided an image decoding method comprising: deriving a first prediction value of a current block by using at least one sample included in a reference block, obtaining an illumination compensation parameter on the basis of a predetermined reference region, deriving a second prediction value of the current block by applying the illumination compensation parameter to the first prediction value and reconstructing the current block on the basis of the second prediction value.

Claims (41)

1. An image decoding method comprising:

deriving a first prediction value of a current block by inter prediction at least based on one sample included in a reference block;

obtaining an illumination compensation parameter based on a linear relationship between a first reference region adjacent to the reference block and a second reference region adjacent to the current block;

deriving a second prediction value of the current block by applying the illumination compensation parameter to the first prediction value; and

reconstructing the current block based on the second prediction value,

wherein the first prediction value is one among a value of a reference sample included in the reference block corresponding to a position of at least one sample included in the current block and a value obtained by performing predetermined filtering on the reference sample, and

wherein the filtering is performed based on a predetermined mapping table.

2. The image decoding method of claim 1 , wherein information on the mapping table is signaled in an adaptation parameter set.

3. The image decoding method of claim 2 , wherein the predetermined mapping table includes a first mapping table,

the first mapping table is derived based on the information on the mapping table, and

a second mapping table corresponding to the first mapping table is derived using the first mapping table.

4. The image decoding method of claim 3 , wherein when the current block is in an intra prediction mode, the filtering is performed using the first mapping table, and

when the current block is in an inter prediction mode, the filtering is performed using the first mapping table and the second mapping table.

5. The image decoding method of claim 1 , wherein the illumination compensation parameter includes at least one among a weighting factor, an offset, and a filter coefficient.

6. The image decoding method of claim 5 , wherein the second prediction value is obtained by applying at least one among the weighting factor and the offset to the first prediction value.

7. The image decoding method of claim 1 , further comprising:

obtaining, when the current block is a luma block, a residual block for a chroma block corresponding to the luma block, based on an illumination compensation parameter for the luma block.

8. An image encoding method comprising:

deriving a first prediction value of a current block by inter prediction at least based on one sample included in a reference block;

obtaining an illumination compensation parameter based on a linear relationship between a first reference region adjacent to the reference block and a second reference region adjacent to the current block;

deriving a second prediction value of the current block by applying the illumination compensation parameter to the first prediction value; and

encoding information on the illumination compensation parameter,

wherein the first prediction value is one among a value of a reference sample included in the reference block corresponding to a position of at least one sample included in the current block and a value obtained by performing predetermined filtering on the reference sample, and

wherein the filtering is performed based on a predetermined mapping table.

9. The image encoding method of claim 8 , wherein information on the mapping table is signaled in an adaptation parameter set.

10. The image encoding method of claim 9 , wherein the predetermined mapping table includes a first mapping table,

the first mapping table is derived based on the information on the mapping table, and

a second mapping table corresponding to the first mapping table is derived using the first mapping table.

11. The image encoding method of claim 10 , wherein when the current block is in an intra prediction mode, the filtering is performed using the first mapping table, and

when the current block is in an inter prediction mode, the filtering is performed using the first mapping table and the second mapping table.

12. The image encoding method of claim 8 , wherein the illumination compensation parameter includes at least one among a weighting factor, an offset, and a filter coefficient.

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

obtaining, when the current block is a luma block, a residual block for a chroma block corresponding to the luma block, based on an illumination compensation parameter for the luma block.

14. A computer-readable non-transitory recording medium storing image data that is received and decoded by an image decoding apparatus to be used for reconstructing an image,

wherein the image data includes information on an illumination compensation parameter,

the information on the illumination compensation parameter is used to obtain an illumination compensation parameter for a current block based on a linear relationship between a first reference region adjacent to the reference block and a second reference region adjacent to the current block,

a first prediction value of the current block is derived using at least one sample included in a reference block,

a second prediction value of the current block is derived by applying the illumination compensation parameter to the first prediction value, and

the current block is reconstructed based on the second prediction value,

wherein the first prediction value is one among a value of a reference sample included in the reference block corresponding to a position of at least one sample included in the current block and a value obtained by performing predetermined filtering on the reference sample, and

wherein the filtering is performed based on a predetermined mapping table.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 16, 2021
From: KWANGWOON UNIVERSITY INDUSTRY-ACADEMIC COLLABORATION FOUNDATION
To: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE
Reel/Frame 055277/0187 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 15, 2021
From: BANG, GUN; KIM, HUI YONG; SIM, DONG GYU; OH, SEOUNG JUN; PARK, SEA NAE; PARK, JUN TAEK; LEE, JONG SEOK
To: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTE; KWANGWOON UNIVERSITY INDUSTRY- ACADEMIC COLLABORATION FOUNDATION
Reel/Frame 055015/0203 →
Priority Claims (2)
KR 10-2018-0083641 · Jul 18, 2018 · national
KR 10-2018-0115944 · Sep 28, 2018 · national
Continuity (1)
Related Publication 20210289201A1 · Sep 16, 2021
Cited By (2)
US 12,368,889 US 12,684,157