IP Library Granted Patent US 10,951,886
Granted Patent B2
US 10,951,886 · App. 17/079,784 · Granted Mar 16, 2021

Intra prediction-based video encoding/decoding method and device

Inventor: Ki Baek Kim (Daejeon, KR)
H04N19/105H04N19/159H04N19/176H04N19/46H04N19/96
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 10,951,886
App. No.
17/079,784
Granted
Mar 16, 2021
Kind
B2
Abstract

A video encoding/decoding method and device according to the present invention may: determine a reference area for the intra prediction of the current block; derive the intra prediction mode of the current block; and decode the current block on the basis of the reference area and the intra prediction mode.

Claims (46)

1. A method of decoding an image based on intra prediction with a decoding apparatus, comprising:

determining a reference region for the intra prediction of a current block in the image,

wherein the reference region is spatially adjacent to the current block and includes at least one of a top reference region, a left reference region, a top-left reference region, a bottom-left reference region, or a top-right reference region;

selecting, based on a flag signaled from an encoding apparatus, one of two candidate groups pre-defined in the decoding apparatus,

wherein a number of intra prediction modes in one of the two candidate groups is different from a number of intra prediction modes in the other one of the two candidate groups,

wherein the one of the two candidate groups includes one non-directional mode while the other one of the two candidate groups includes three or more non-directional modes, and

wherein the flag specifies the one of the two candidate groups;

determining an intra prediction mode of the current block from the selected candidate group,

wherein the selected candidate group is divided into three mode lists,

wherein determining the intra prediction mode of the current block comprises:

selecting a mode list to which an intra prediction mode of the current block belongs among the three mode lists; and

deriving the intra prediction mode of the current block based on the selected mode list,

wherein a first mode list of the three mode lists includes an intra prediction mode of a neighboring block adjacent to the current block, a second mode list of the three mode lists includes an intra prediction mode derived by adding or subtracting a predetermined constant to the intra prediction mode of the neighboring block, and a third mode list of the three mode lists includes remaining modes except for intra prediction modes in the first mode list and the second mode list among intra prediction modes belonging to the selected candidate group, and

wherein the predetermined constant is a natural number greater than or equal to 5.

2. The method of claim 1 , wherein the current block is obtained through a block division based on at least one of a binary tree or a ternary tree, and

wherein the binary tree is representative of a division type of dividing one coding block into two coding blocks, and the ternary tree is representative of a division type of dividing one coding block into three coding blocks.

3. The method of claim 1 , wherein the reference region further includes at least one of a right reference region or a bottom-right reference region according to a predetermined coding order.

4. The method of claim 1 , wherein determining the reference region comprises searching for whether an unavailable pixel exists in the reference region, and

wherein the search on the left reference region is performed from top to bottom.

5. The method of claim 4 , wherein when the unavailable pixel is found during the search on the left reference region, the unavailable pixel is replaced with a pixel above the unavailable pixel.

6. A method of encoding an image based on intra prediction with an encoding apparatus, comprising:

determining a reference region for the intra prediction of a current block in the image,

wherein the reference region is spatially adjacent to the current block and includes at least one of a top reference region, a left reference region, a top-left reference region, a bottom-left reference region, or a top-right reference region;

selecting one of two candidate groups pre-defined in the encoding apparatus,

wherein a number of intra prediction modes in one of the two candidate groups is different from a number of intra prediction modes in the other one of the two candidate groups, and

wherein the one of the two candidate groups includes one non-directional mode while the other one of the two candidate groups includes three or more non-directional modes, and

wherein a flag for specifying the one of the two candidate groups is encoded in the encoding apparatus;

determining an intra prediction mode of the current block from the selected candidate group,

wherein the other one of the two candidate groups is divided into three mode lists,

wherein a mode list to which the intra prediction mode of the current block belongs is selected among the three mode lists,

wherein a first mode list of the three mode lists includes an intra prediction mode of a neighboring block adjacent to the current block, a second mode list of the three mode lists includes an intra prediction mode derived by adding or subtracting a predetermined constant to the intra prediction mode of the neighboring block, and a third mode list of the three mode lists includes remaining modes except for intra prediction modes in the first mode list and the second mode list among intra prediction modes belonging to the selected candidate group, and

wherein the predetermined constant is a natural number greater than or equal to 5.

7. A non-transitory computer-readable storage medium having instructions stored thereon that, when executed, cause one or more processors to:

determining a reference region for intra prediction of a current block in an image,

wherein the reference region is spatially adjacent to the current block and includes at least one of a top reference region, a left reference region, a top-left reference region, a bottom-left reference region, or a top-right reference region;

selecting, based on a first flag signaled from an encoding apparatus, one of two candidate groups pre-defined in the decoding apparatus,

wherein a number of intra prediction modes in one of the two candidate groups is different from a number of intra prediction modes in the other one of the two candidate groups,

wherein the one of the two candidate groups includes one non-directional mode while the other one of the two candidate groups includes three or more non-directional modes, and

wherein the first flag specifies the one of the two candidate groups;

determining an intra prediction mode of the current block from the selected candidate group,

wherein the selected candidate group is divided into three mode lists,

wherein determining the intra prediction mode of the current block comprises:

selecting a mode list to which an intra prediction mode of the current block belongs among the three mode lists; and

deriving the intra prediction mode of the current block based on the selected mode list,

wherein a first mode list of the three mode lists includes an intra prediction mode of a neighboring block adjacent to the current block, a second mode list of the three mode lists includes an intra prediction mode derived by adding or subtracting a predetermined constant to the intra prediction mode of the neighboring block, and a third mode list of the three mode lists includes remaining modes except for intra prediction modes in the first mode list and the second mode list among intra prediction modes belonging to the selected candidate group, and

wherein the predetermined constant is a natural number greater than or equal to 5.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 12, 2021
From: KIM, KI BAEK
To: B1 INSTITUTE OF IMAGE TECHNOLOGY, INC.
Reel/Frame 055898/0420 →
Priority Claims (2)
KR 10-2018-0173164 · Dec 28, 2018 · national
KR 10-2018-0173228 · Dec 29, 2018 · national
Continuity (2)
Continuation PCTKR2019018740 · Dec 30, 2019
Related Publication 20210044798A1 · Feb 11, 2021
Cited By (4)
US 12,256,063 US 12,587,638 US 12,598,295 US 12,621,436