IP Library Granted Patent US 12707064
Granted Patent B2
US 12707064 · App. 18/622,277 · Granted Aug 11, 2026

Video encoding/decoding method and apparatus applying a merge mode with motion vector difference to combined an inter/intra prediction mode or a geometric partitioning mode

Inventors: Seung Wook Park (Yongin-si, KR); Jin Heo (Yongin-si, KR)
Assignees: HYUNDAI MOTOR COMPANY; KIA CORPORATION
H04N19/159H04N19/139H04N19/176
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 12707064
App. No.
18/622,277
Granted
Aug 11, 2026
Kind
B2
Abstract

A method and an apparatus for encoding/decoding a video are provided. A method for decoding a video according to the present disclosure includes determining whether to apply a combined inter/intra prediction (CIIP) mode or a geometric partitioning mode. The method for decoding the video also includes deriving a motion vector from a merge candidate list. The method for decoding the video also includes correcting the motion vector by using motion vector difference information. The method for decoding the video also includes generating a prediction block of a current block based on a corrected motion vector.

Claims (45)

1 . A video decoding method, comprising:

determining whether to apply a geometric partitioning mode;

deriving a motion vector from a merge candidate list;

correcting the motion vector by using motion vector difference information; and

generating a prediction block of a current block based on a corrected motion vector,

wherein the motion vector difference information includes direction information and distance information, and

wherein the direction information indicates one direction selected from a predetermined direction set, the predetermined direction set including four diagonal directions.

2 . The video decoding method of claim 1 , wherein, when the geometric partitioning mode is applied, deriving the motion vector from the merge candidate list comprises:

deriving a motion vector for each of two partitioned regions from the merge candidate list.

3 . The video decoding method of claim 2 , wherein correcting the motion vector, when the two partitioned regions use different motion vector difference information, comprises:

correcting the motion vector for each of the two partitioned regions by using the different motion vector difference information.

4 . The video decoding method of claim 2 , wherein correcting the motion vector, when the two partitioned regions use same motion vector difference information, comprises:

correcting the motion vector for each of the two partitioned regions by using the same motion vector difference information.

5 . The video decoding method of claim 3 , wherein generating the prediction block of the current block comprises:

generating an inter-prediction block for each of the two partitioned regions by using the corrected motion vector; and

generating the prediction block of the current block by combining or weighted averaging inter-prediction blocks for the two partitioned regions.

6 . The video decoding method of claim 4 , wherein generating the prediction block of the current block comprises:

generating an inter-prediction block for each of the two partitioned regions by using the corrected motion vector; and

generating the prediction block of the current block by combining or weighted averaging inter-prediction blocks for the two partitioned regions.

7 . A video encoding method, comprising:

determining whether to apply a geometric partitioning mode;

determining a motion vector from a merge candidate list;

correcting the motion vector by using motion vector difference information; and

generating a prediction block of a current block based on a corrected motion vector,

wherein the motion vector difference information includes direction information and distance information, and

wherein the direction information indicates one direction selected from a predetermined direction set, the predetermined direction set including four diagonal directions.

8 . The video encoding method of claim 7 , wherein, when the geometric partitioning mode is applied, determining the motion vector from the merge candidate list comprises:

determining a motion vector for each of two partitioned regions from the merge candidate list.

9 . The video encoding method of claim 8 , wherein correcting the motion vector, when the two partitioned regions use different motion vector difference information, comprises:

correcting the motion vector for each of the two partitioned regions by using the different motion vector difference information.

10 . The video encoding method of claim 8 , wherein correcting the motion vector, when the two partitioned regions use same motion vector difference information, comprises:

correcting the motion vector for each of the two partitioned regions by using the same motion vector difference information.

11 . The video encoding method of claim 9 , wherein generating the prediction block of the current block comprises:

generating an inter-prediction block for each of the two partitioned regions by using a corrected motion vector; and

generating the prediction block of the current block by combining or weighted averaging inter-prediction blocks for the two partitioned regions.

12 . A method for transmitting a bitstream associated with video data, the method comprising:

generating the bitstream by encoding the video data; and

transmitting the bitstream to a video decoding apparatus,

wherein generating the bitstream comprises:

determining whether to apply a geometric partitioning mode;

determining a motion vector from a merge candidate list;

correcting the motion vector by using motion vector difference information; and

generating a prediction block of a current block based on a corrected motion vector,

wherein the motion vector difference information includes direction information and distance information, and

wherein the direction information indicates one direction selected from a predetermined direction set, the predetermined direction set including four diagonal directions.