公开(公告)号：US20250150618A1

公开(公告)日：2025-05-08

申请号：US18838079

申请日：2023-02-15

Applicant: LG ELECTRONICS INC.

Inventor： Hyeong Moon JANG ,  Nae Ri PARK

IPC: H04N19/577 ,  H04N19/105 ,  H04N19/159 ,  H04N19/176 ,  H04N19/54

Abstract: An image encoding/decoding method, a bitstream transmission method, and a computer-readable recording medium storing a bitstream are provided. The image decoding method according to the present disclosure, performed by an image decoding apparatus, comprises the steps of: acquiring first information on the basis that the prediction direction of a current block is bidirectional; on the basis that the first information indicates that different prediction types are applied to the respective prediction directions of the current block, deriving, from reference blocks, at least one bidirectional motion information candidate corresponding to the different prediction types; and on the basis of any one of the at least one bidirectional motion information candidate, bidirectionally predicting the current block with the different prediction types, wherein the different prediction types are a sub-block unit (sub PU) prediction type and a non-sub-block unit (non-sub PU) prediction type.

公开(公告)号：US20220256162A1

公开(公告)日：2022-08-11

申请号：US17683203

申请日：2022-02-28

Applicant: LG ELECTRONICS INC.

Inventor： Nae Ri PARK ,  Seung Hwan KIM ,  Jung Hak Nam ,  Hyeong Moon JANG

IPC: H04N19/132 ,  H04N19/137 ,  H04N19/105 ,  H04N19/176

Abstract: An image encoding/decoding method and apparatus are provided. An image decoding method according to the present disclosure is performed by an image decoding apparatus. The image decoding method may comprise deriving a prediction sample of a current block based on motion information of the current block, determining whether prediction refinement with optical flow (PROF) applies to the current block, deriving, based on that the PROF applies to the current block, a difference motion vector for each sample position in the current block, deriving a gradient for each sample position in the current block, deriving a PROF offset based on the difference motion vector and the gradient, and deriving a refined prediction sample for the current block based on the PROF offset.

公开(公告)号：US20250071258A1

公开(公告)日：2025-02-27

申请号：US18723700

申请日：2023-04-12

Applicant: LG Electronics Inc.

Inventor： Hyeong Moon JANG ,  Nae Ri PARK ,  Jung Hak NAM

IPC: H04N19/105 ,  H04N19/139 ,  H04N19/176 ,  H04N19/513

Abstract: An image encoding/decoding method, a bitstream transmission method, and a computer-readable recording medium storing a bitstream are provided. The image decoding method performed by an image decoding apparatus, according to the present disclosure, comprises the steps of: on the basis of whether or not reordering of prediction candidates in a merge mode with motion vector difference (MMVD) mode is activated, determining whether or not to reorder the prediction candidates; identifying a prediction candidate indicated by prediction candidate identification information from the prediction candidates or the reordered prediction candidates; and deriving an offset of the MMVD mode on the basis of the identified prediction candidate, wherein the prediction candidate identification information is acquired from a bitstream.

公开(公告)号：US20230412830A1

公开(公告)日：2023-12-21

申请号：US18242408

申请日：2023-09-05

Applicant: LG Electronics Inc.

Inventor： Nae Ri PARK ,  Jung Hak NAM ,  Hyeong Moon JANG

IPC: H04N19/46 ,  H04N19/105 ,  H04N19/159 ,  H04N19/174 ,  H04N19/176 ,  H04N19/70

CPC classification number: H04N19/46 ,  H04N19/105 ,  H04N19/159 ,  H04N19/174 ,  H04N19/176 ,  H04N19/70

Abstract: An image encoding/decoding method and apparatus are provided. An image decoding method according to the present disclosure is performed by an image decoding apparatus. The image decoding method comprises deriving a first flag specifying whether to perform weighted prediction on a current block and a weight index BcwIdx of bi-prediction with CU-level weight (BCW) for the current block, determining whether to perform default weighted prediction or explicit weighted prediction on the current block based on the first flag and BcwIdx, and generating a prediction block for the current block by performing the determined method.

公开(公告)号：US20230224470A1

公开(公告)日：2023-07-13

申请号：US18008619

申请日：2021-06-11

Applicant: LG Electronics Inc.

Inventor： Hendry HENDRY ,  Hyeong Moon JANG ,  Jaehyun LIM ,  Nae Ri PARK ,  Seung Hwan KIM

IPC: H04N19/136 ,  H04N19/159 ,  H04N19/169 ,  H04N19/172

CPC classification number: H04N19/136 ,  H04N19/159 ,  H04N19/188 ,  H04N19/172

Abstract: An image encoding/decoding method and apparatus is provided. The image decoding method comprises determining whether first information specifying a value of a picture order count (POC) most significant bit (MSB) cycle of a current picture is present in a picture header, determining a POC MSB of the current picture based on whether the first information is present in the picture header, and deriving a POC of the current picture based on the determined POC MSB. Based on the first information being not present in the picture header and the current picture being not a CLVSS (coded layer video sequence start) picture, a POC difference between the current picture and a previous picture preceding the current picture in decoding order may be less than half a maximum value of a POC least significant bit (LSB) of the current picture.

公开(公告)号：US20230080116A1

公开(公告)日：2023-03-16

申请号：US17797953

申请日：2021-02-15

Applicant: LG ELECTRONICS INC.

Inventor： Nae Ri PARK ,  Jung Hak NAM ,  Hyeong Moon JANG

IPC: H04N19/119 ,  H04N19/117 ,  H04N19/172 ,  H04N19/46

Abstract: An image coding/decoding method and device are provided. The image decoding method carried out by the image decoding device, according to the present disclosure, may comprise the steps of: determining the number of tiles in a current picture on the basis that the segmentation of the current picture is not restricted; on the basis that the number of tiles in the current picture is a plurality, acquiring from a bitstream, a first flag representing whether tile boundary filtering is available; and, on the basis of the value of the first flag, determining whether to carry out filtering for the boundaries of the tiles belonging to the current picture.

公开(公告)号：US20220337842A1

公开(公告)日：2022-10-20

申请号：US17639837

申请日：2020-09-09

Applicant: LG Electronics Inc.

Inventor： Nae Ri PARK ,  Jung Hak NAM ,  Hyeong Moon JANG ,  Seung Hwan KIM

IPC: H04N19/132 ,  H04N19/137 ,  H04N19/159 ,  H04N19/176

Abstract: An image encoding/decoding method and apparatus are provided. An image decoding method according to the present disclosure is performed by an image decoding apparatus. The image decoding method may comprise deriving a prediction sample of a current block based on motion information of the current block, determining whether bi-directional optical flow (BDOF) applies to the current block, based on that the BDOF applies to the current block, deriving a gradient for a current subblock in the current block, deriving motion refinement (vx, vy) for the current subblock based on the gradient, deriving a BDOF offset based on the gradient and the motion refinement, and deriving a refined prediction sample for the current block based on the prediction sample of the current block and the BDOF offset.

公开(公告)号：US20220279163A1

公开(公告)日：2022-09-01

申请号：US17636911

申请日：2020-08-20

Applicant: LG ELECTRONICS INC.

Inventor： Nae Ri PARK ,  Jung Hak NAM ,  Hyeong Moon JANG

IPC: H04N19/105 ,  H04N19/137 ,  H04N19/14 ,  H04N19/176

Abstract: An image encoding/decoding method and apparatus are provided. An image decoding method according to the present disclosure is performed by an image decoding apparatus. The image decoding method comprises deriving a prediction mode of a current block, constructing a merge candidate list for the current block, based on the prediction mode of the current block being a triangle partition mode (TPM), deriving motion information of two partitions of the current block based on the merge candidate list and two merge indices of the current block signaled through a bitstream, generating a prediction block of the current block based on the motion information, storing motion information of the current block, and updating a history-based motion vector predictor (HMVP) buffer based on the motion information of the current block.

公开(公告)号：US20250030851A1

公开(公告)日：2025-01-23

申请号：US18775661

申请日：2024-07-17

Applicant: LG ELECTRONICS INC.

Inventor： Jung Hak NAM ,  Hyeong Moon JANG ,  Nae Ri PARK

IPC: H04N19/119 ,  H04N19/137 ,  H04N19/159 ,  H04N19/176 ,  H04N19/186 ,  H04N19/593 ,  H04N19/96

Abstract: An image encoding/decoding method and apparatus are provided. An image decoding method performed by an image decoding apparatus includes determining whether predetermined splitting is allowed for a current block based on a size of the current block, obtaining information on splitting of the current block based on whether the predetermined splitting is allowed, splitting the current block based on the information on splitting of the current block, and decoding the current block. Whether the predetermined splitting is allowed may be determined based on whether a width or height of the current block is a predetermined sample length.

公开(公告)号：US20240244213A1

公开(公告)日：2024-07-18

申请号：US18417903

申请日：2024-01-19

Applicant: LG ELECTRONICS INC.

Inventor： Nae Ri PARK ,  Jung Hak NAM ,  Hyeong Moon JANG

IPC: H04N19/132 ,  H04N19/105 ,  H04N19/137 ,  H04N19/176 ,  H04N19/513

CPC classification number: H04N19/132 ,  H04N19/105 ,  H04N19/137 ,  H04N19/176 ,  H04N19/513

Abstract: An image encoding/decoding method and apparatus are provided. An image decoding method according to the present disclosure is performed by an image decoding apparatus. The image decoding method comprises deriving a prediction sample of a current block based on motion information of the current block, deriving a reference picture resampling (RPR) condition for the current block, determining whether prediction refinement with optical flow (PROF) applies to the current block based on the RPR condition, and deriving a refined prediction sample for the current block by applying PROF to the current block.