公开(公告)号：US11503329B2

公开(公告)日：2022-11-15

申请号：US17265949

申请日：2019-08-15

Applicant: MEDIATEK INC.

Inventor： Chun-Chia Chen ,  Chih-Wei Hsu ,  Tzu-Der Chuang ,  Ching-Yeh Chen

IPC: H04N19/52 ,  H04N19/96 ,  H04N19/105 ,  H04N19/119

Abstract: A method and apparatus of Inter prediction for video coding are disclosed. According to one method, a sub-block motion vector prediction (MVP) mode is turned off for small size coding units (CUs). In another method, if the neighbouring reference block for a current coding unit (CU) is in a root CU region, the neighbouring reference block is not used to derive a Merge candidate or a modified neighbouring reference block on the shared boundary of the root CU is used to derive the Merge candidate for the current block. In yet another method, a shared sub-block Merge candidate list is derived for sub-CUs within a root CU region or an MER (Merge estimation region). If a neighbouring reference block is within the same MER as a current sub-CU, the neighbouring reference block is not used for deriving a candidate for the shared sub-CU Merge list.

公开(公告)号：US11438611B2

公开(公告)日：2022-09-06

申请号：US17117724

申请日：2020-12-10

Applicant: MEDIATEK INC.

Inventor： Tzu-Der Chuang ,  Chih-Wei Hsu ,  Ching-Yeh Chen ,  Chia-Ming Tsai ,  Chun-Chia Chen ,  Olena Chubach ,  Yu-Wen Huang

IPC: H04N19/33 ,  H04N19/132 ,  H04N19/172 ,  H04N19/184

Abstract: Method and apparatus of coding a video sequence, are disclosed. According to the method, a bitstream corresponding to encoded data of the video sequence is generated at an encoder side or received at a decoder side, where the bitstream complies with a bitstream conformance that one or more constraints are satisfied. The constraints are related to a set of RPR (Reference Picture Resampling) parameters including scaling window width or height of a current picture, scaling window width or height of a reference picture, current picture width or height, and maximum picture width or height specified for the video sequence. Scaling information for the RPR mode is derived using the set of RPR parameters. A target picture of the video sequence is then encoded at the encoder side or decoded at the decoder side by utilizing the scaling information when the RPR mode is enabled for the target picture.

公开(公告)号：US11381838B2

公开(公告)日：2022-07-05

申请号：US17293674

申请日：2019-11-15

Applicant: MEDIATEK INC.

Inventor： Chen-Yen Lai ,  Yu-Ling Hsiao ,  Tzu-Der Chuang ,  Ching-Yeh Chen

IPC: H04N19/56 ,  H04N19/172 ,  H04N19/176

Abstract: A prediction method and apparatus for video coding include a set of Merge candidates having an MMVD (Merge mode with motion vector differences (MVD)) candidate are generated. When the current block has a block size belonging to at least one default block size, whether a target candidate in the set of Merge candidates is bi-prediction is determined after the set of Merge candidates is generated. If the target candidate is bi-prediction: the target candidate is replaced by a uni-prediction candidate; the uni-prediction candidate is included in the set of Merge candidates to generate a modified set of Merge candidates; and current motion information associated with the current block is encoded using the modified set of Merge candidates at the video encoder side or the current motion information associated with the current block is decoded at the video decoder side using the modified set of Merge candidates.

公开(公告)号：US11381834B2

公开(公告)日：2022-07-05

申请号：US17041655

申请日：2019-04-02

Applicant: MEDIATEK INC.

Inventor： Tzu-Der Chuang ,  Ching-Yeh Chen ,  Chen-Yen Lai ,  Chih-Wei Hsu

IPC: H04N19/513 ,  H04N19/105 ,  H04N19/119 ,  H04N19/159 ,  H04N19/176 ,  H04N19/533

Abstract: Video processing methods and apparatuses for coding a current block comprise receiving input data of a current block, partitioning the current block into multiple sub-blocks, deriving sub-block MVs for the current block according to a sub-block motion compensation coding tool, constraining the sub-block MVs to form constrained sub-block MVs, and encoding or decoding the current block using the constrained sub-block MVs. The sub-block MVs may be constrained according to a size, width, or height of the current block or a sub-block, an inter prediction direction of one of control point MVs of the current block, the current block, or current sub-block, the control point MVs, or a combination of the above.

公开(公告)号：US11310526B2

公开(公告)日：2022-04-19

申请号：US16248452

申请日：2019-01-15

Applicant: MEDIATEK INC.

Inventor： Chun-Chia Chen ,  Zhen-Yen Lai ,  Chih-Wei Hsu ,  Tzu-Der Chuang ,  Ching-Yeh Chen

IPC: H04N19/513 ,  H04N19/105 ,  H04N19/139 ,  H04N19/176 ,  H04N19/573 ,  H04N19/147 ,  H04N19/567

Abstract: A video codec that encodes or decodes video sequences using decoder-side motion vector refinement is provided. The video codec identifies a first motion vector and a second motion vector for coding a current block of pixels of a current video frame in the video sequence. The video codec determines whether to perform motion vector refinement for the current block of pixels based on a comparison between a linear dimension of the current block of pixels and a threshold. When motion vector refinement is performed, the video codec refines the first and second motion vectors to minimize a distortion cost and codes the current block of pixels by using the refined first and second motion vectors. When motion vector refinement is not performed, the video codec codes the current block of pixels by using the identified first and second motion vectors.

公开(公告)号：US11310520B2

公开(公告)日：2022-04-19

申请号：US16559966

申请日：2019-09-04

Applicant: MEDIATEK INC.

Inventor： Tzu-Der Chuang ,  Ching-Yeh Chen ,  Zhi-Yi Lin

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

Abstract: Method and apparatus for coding system using Inter coding tools are disclosed. According to one method, a same conversion process is applied to the MVs associated with a current block to generate the converted MVs at the video encoder side or at the video decoder side when the target Inter coding tool belongs to at least two tools of a group, wherein the group comprises at least one of inherited affine Merge candidate derivation and affine sub-block MV derivation. According to another method, a simplified conversion process is applied to the MV to generate a rounded MV, MV_rounded according to MV_rounded=(MV+offset−(MV>=0))>>shift.

公开(公告)号：US11272182B2

公开(公告)日：2022-03-08

申请号：US17048208

申请日：2019-04-17

Applicant: MEDIATEK INC.

Inventor： Chia-Ming Tsai ,  Chih-Wei Hsu ,  Tzu-Der Chuang ,  Ching-Yeh Chen ,  Shih-Ta Hsiang ,  Yu-Wen Huang ,  Zhi-Yi Lin

IPC: H04N19/132 ,  H04N19/119 ,  H04N19/176 ,  H04N19/70 ,  H04N19/96 ,  H04N19/172 ,  H04N19/61 ,  H04N19/46 ,  H04N19/159 ,  H04N19/186

Abstract: A method and apparatus for video coding using block partition are disclosed. According to the present invention, a partition structure corresponding to recursively partitioning a current block into smaller TU (transform unit) blocks until the partition structure reaches a maximum allowed split depth or until a block size of at least one of smaller TU blocks is a supported core transform size, where the current block is partitioned into final smaller TU blocks according to the partition structure. A transform coding process is applied to the current block according to the partition structure, where the transform coding process is skipped for at least one of the final smaller TU blocks. A flag can be signalled for the current block to indicate whether the current block is allowed to skip the transform coding process for said at least one of the final smaller TU blocks.

公开(公告)号：US11245922B2

公开(公告)日：2022-02-08

申请号：US16541627

申请日：2019-08-15

Applicant: MediaTek Inc.

Inventor： Chun-Chia Chen ,  Chih-Wei Hsu ,  Tzu-Der Chuang ,  Ching-Yeh Chen ,  Yu-Wen Huang

IPC: H04N19/56 ,  H04N19/53 ,  H04N19/96 ,  H04N19/583

Abstract: A video coder that uses a shared candidate list to encode or decode multiple blocks of pixels within a shared boundary is provided. The video coder identifies a shared boundary encompassing a plurality of blocks of pixels of a current picture in a video sequence. The video coder identifies one or more prediction candidates as a shared candidate list based on neighbors of a region defined by the shared boundary. The video coder codes one or more blocks of pixels encompassed by the shared boundary by using one or more prediction candidates selected from the shared candidate list.

公开(公告)号：US11196992B2

公开(公告)日：2021-12-07

申请号：US15754694

申请日：2016-08-29

Applicant: MEDIATEK INC.

Inventor： Yu-Wen Huang ,  Yu-Chen Sun ,  Tzu-Der Chuang ,  Jian-Liang Lin ,  Ching-Yeh Chen

IPC: H04N19/895 ,  H04N19/105 ,  H04N19/166 ,  H04N19/174 ,  H04N19/65 ,  H04N19/117 ,  H04N19/82 ,  G06T5/00 ,  H04N19/86 ,  H04N19/182 ,  H04N19/50 ,  G06N3/04 ,  G06N3/08 ,  H04N19/91

Abstract: A method and apparatus of video coding incorporating Deep Neural Network are disclosed. A target signal is processed using DNN (Deep Neural Network), where the target signal provided to DNN input corresponds to the reconstructed residual, output from the prediction process, the reconstruction process, one or more filtering processes, or a combination of them. The output data from DNN output is provided for the encoding process or the decoding process. The DNN can be used to restore pixel values of the target signal or to predict a sign of one or more residual pixels between the target signal and an original signal. An absolute value of one or more residual pixels can be signalled in the video bitstream and used with the sign to reduce residual error of the target signal.

公开(公告)号：US11178397B2

公开(公告)日：2021-11-16

申请号：US17283332

申请日：2019-10-09

Applicant: MEDIATEK INC.

Inventor： Zhi-Yi Lin ,  Tzu-Der Chuang ,  Ching-Yeh Chen

IPC: H04N19/00 ,  H04N19/105 ,  H04N19/132 ,  H04N19/176 ,  H04N19/186 ,  H04N19/159

Abstract: A video processing method comprises receiving input data of a current block, checking if the current block satisfies one or more predefined criteria, setting the current block to be a root block if the current block satisfies the predefined criteria, one or more color components of one or more blocks in the current block are not checked with the predefined criteria if the current block is a root block, encoding or decoding the one or more color components of one or more blocks in the current block using neighboring reconstructed samples of the one or more color components of the current block as reference samples if the current block is a root block. Each block in the current block is encoded or decoded using neighboring reconstructed samples of each block in the current block as reference samples if the current block is not a root block.