公开(公告)号：US20230199199A1

公开(公告)日：2023-06-22

申请号：US17978530

申请日：2022-11-01

Applicant: MediaTek Inc.

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

IPC: H04N19/176 ,  H04N19/119

CPC classification number: H04N19/176 ,  H04N19/119

Abstract: Various schemes pertaining to video coding parallelization techniques are described. An apparatus receives video data. The apparatus subsequently calculates a plurality of figures of merits (FOMs), each of the FOM representing how well a particular coding tool may perform in encoding the video data. The apparatus further determines a coding tool that may be suitable for encoding the video data by comparing the FOMs. In determining the coding tool, the apparatus utilizes time-interleaving techniques to parallelly process the video data. The video data may include an array of coding blocks, and the apparatus may receive the video data using a snake-like processing order scanning through the array of coding blocks.

公开(公告)号：US20230199170A1

公开(公告)日：2023-06-22

申请号：US17977466

申请日：2022-10-31

Applicant: MediaTek Inc.

Inventor： Cheng-Yen Chuang ,  Man-Shu Chiang ,  Chun-Chia Chen ,  Chih-Wei Hsu ,  Tzu-Der Chuang ,  Ching-Yeh Chen ,  Yu-Wen Huang

IPC: H04N19/105 ,  H04N19/176 ,  H04N19/117 ,  H04N19/423 ,  H04N19/52

CPC classification number: H04N19/105 ,  H04N19/52 ,  H04N19/117 ,  H04N19/176 ,  H04N19/423

Abstract: A video coding system generating candidates for Merge Mode with Motion Vector Difference (MMVD) with reduced resource usage is provided. The system receives data to be encoded or decoded as a current block of a current picture of a video. The system identifies multiple MMVD candidates for different offset positions based on a merge candidate of the current block. The system generates reference samples for the identified MMVD candidates. The system reconstructs the current block or encodes the current block into a bitstream by using the generated reference samples. The system processes the MMVD candidates in separate groups: a first group of vertical MMVD candidates and a second group of horizontal MMVD candidates. The system generates the reference samples for the identified MMVD candidates by applying a vertical filter to source reference samples of horizontal MMVD candidates and then applying a horizontal filter to outputs of the vertical filter.

公开(公告)号：US11539940B2

公开(公告)日：2022-12-27

申请号：US17292184

申请日：2019-11-12

Applicant: MEDIATEK INC.

Inventor： Man-Shu Chiang ,  Chih-Wei Hsu ,  Tzu-Der Chuang ,  Ching-Yeh Chen

IPC: H04N19/105 ,  H04N19/119 ,  H04N19/172 ,  H04N19/176 ,  H04N19/51

Abstract: A method and apparatus of prediction for video coding using MH (Multiple Hypothesis) mode are disclosed. According to this method, a block is partitioned into a first partition and a second partition. A first candidate and a second candidate are derived for the first and second partitions respectively. At least one of the first candidate and the second candidate is derived using a candidate list derived for a regular Merge mode (or also called as normal Merge mode). An MH prediction generated by blending a first prediction corresponding to a first candidate and a second prediction corresponding to a second candidate, and the MH prediction is applied to a part of the current block.

公开(公告)号：US11516513B2

公开(公告)日：2022-11-29

申请号：US16977158

申请日：2019-03-13

Applicant: MEDIATEK INC.

Inventor： Chun-Chia Chen ,  Chia-Ming Tsai ,  Yu-Chi Su ,  Chen-Yen Lai ,  Chih-Wei Hsu ,  Ching-Yeh Chen ,  Tzu-Der Chuang ,  Yu-Wen Huang ,  Han Huang

IPC: H04N19/70 ,  H04N19/174 ,  H04N19/64 ,  H04N19/66 ,  H04N19/89

Abstract: In one method, the current block is partitioned into multiple final sub-blocks using one or more stages of sub-tree partition comprising ternary tree partition and at least one other-type partition, where ternary partition tree is excluded from the sub-tree partition if a current sub-tree depth associated with a current sub-block is greater than a first threshold and the first threshold is an integer greater than or equal to 1. In another method, if a test condition is satisfied, the current block is encoded or decoded using a current Inter mode selected from a modified group of Inter tools, where the modified group of Inter tools is derived from an initial group of Inter tools by removing one or more first Inter tools from the initial group of Inter tools, replacing one or more second Inter tools with one or more complexity-reduced Inter tools, or both.

公开(公告)号：US11509934B2

公开(公告)日：2022-11-22

申请号：US16494586

申请日：2018-03-13

Applicant: MEDIATEK INC.

Inventor： Ching-Yeh Chen ,  Tzu-Der Chuang ,  Yu-Wen Huang

IPC: H04N19/122 ,  H04N19/60 ,  H04N19/176 ,  H04N19/18 ,  H04N19/96

Abstract: A method and apparatus for video coding incorporating enhanced multiple transform (EMT) are disclosed. According to one method, the number of candidates for EMT in the horizontal direction or vertical direction is reduced depending on the current block size. According to another method, an EMT flag, one or more EMT indices or both are signalled only if the total number of non-zero coefficients in at least a part of the 2D coefficient block is greater than a threshold, where the threshold is equal to 1 or larger. A method and apparatus for video coding using non-separable secondary transform (NSST) are disclosed. According to this method, a total number of non-zero first coefficients in a partial block of the 2D coefficient block is determined and used to determine whether to apply the NSST process.

公开(公告)号：US11483575B2

公开(公告)日：2022-10-25

申请号：US16548340

申请日：2019-08-22

Applicant: MediaTek Inc.

Inventor： Shih-Ta Hsiang ,  Tzu-Der Chuang ,  Ching-Yeh Chen

IPC: H04N19/18 ,  H04N19/146 ,  H04N19/70 ,  H04N19/13

Abstract: A video coder that constrains the total number of regular bins used for entropy coding syntax elements of a current block is provided. The video coder entropy encodes or decodes the syntax elements selectively as either regular bins using context modeling or as bypass bins without context modeling. A constraint is specified to limit a total number of regular bins used for entropy coding the syntax elements of the current block. There may be no constraint limiting a number of regular bins specific to an individual syntax element of the current block.

公开(公告)号：US11470356B2

公开(公告)日：2022-10-11

申请号：US17047244

申请日：2019-04-16

Applicant: MEDIATEK INC.

Inventor： Yu-Ling Hsiao ,  Yu-Chi Su ,  Jan Klopp ,  Ching-Yeh Chen ,  Tzu-Der Chuang ,  Chih-Wei Hsu ,  Yu-Wen Huang

IPC: H04N19/70 ,  H04N19/82 ,  G06N3/04

Abstract: A method and apparatus of video encoding video coding for a video encoder or decoder using Neural Network (NN) are disclosed. According to this method, the multiple frames in a video sequence comprises multiple segments, where each of the multiple segments comprises a set of frames. The NN (Neural Network) processing is applied to a target signal in one or more encoded frames of a target segment in the encoder side or to the target signal in one or more decoded frames of the target segment in the decoder side using one NN parameter set for the target segment. The target signal may correspond to reconstructed residual, reconstructed output, de-blocked output, SAO (sample adaptive offset) output, ALF (adaptive loop filter) output, or a combination thereof. In another embodiment, the NN processing is applied to a target signal only in one or more specific encoded or decoded frames.

公开(公告)号：US11445176B2

公开(公告)日：2022-09-13

申请号：US17117757

申请日：2020-12-10

Applicant: MEDIATEK INC.

Inventor： Tzu-Der Chuang ,  Ching-Yeh Chen

IPC: H04N19/105 ,  H04N19/132 ,  H04N19/176 ,  H04N19/186 ,  H04N19/70

Abstract: Method and apparatus of coding a video sequence are disclosed, wherein Reference Picture Resampling (RPR) mode is included as a coding tool. According to the method, a bitstream corresponding to encoded data of the video sequence is generated or received, where the bitstream includes one or more syntaxes related to the RPR mode when the RPR mode is enabled, and scaling information for the RPR mode is derived using first information including said one or more syntaxes, and wherein said one or more syntaxes are constrained by taking into account a value corresponding to Max(8, (minimum coding block size for a luma component)). A target picture of the video sequence is encoded or decoded utilizing the scaling information when the RPR mode is enabled for the target picture.

公开(公告)号：US11438590B2

公开(公告)日：2022-09-06

申请号：US17265967

申请日：2019-08-16

Applicant: MEDIATEK INC.

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

IPC: H04N19/119 ,  H04N19/124 ,  H04N19/167 ,  H04N19/176 ,  H04N19/186 ,  H04N19/70 ,  H04N19/196 ,  H04N19/96

Abstract: Video data processing methods and apparatuses comprise receiving input data associated with a current chroma block, determine a luma Quantization Parameter (QP) of a collocated luma block, reuse the luma QP of the collocated luma block to derive a chroma QP for the current chroma block, and encode or decode one or more Transform Units (TUs) associated with the current chroma block using the chroma QP. The collocated luma block is a block that covers a collocated sample of one predefined sample of the current chroma block, and an example of the predefined sample is a center sample of the current chroma block. The input video data is partitioned according to two separate Coding Unit (CU) partitioning structures for luma and chroma components when dual tree coding is enabled.

公开(公告)号：US11425379B2

公开(公告)日：2022-08-23

申请号：US17614703

申请日：2020-05-29

Applicant: MEDIATEK INC.

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

IPC: H04N19/119 ,  H04N19/184 ,  H04N19/146 ,  H04N19/132

Abstract: A method and apparatus of video coding are disclosed. A piece-wise linear mapping is derived to map a target luma sample from a first signal domain to a second signal domain, or vice versa. The piece-wise linear mapping process comprises determining first pivot points for performing video encoding or decoding of the luma samples using a piece-wise linear mapping. The first pivot points are determined so that: if a target mapped pivot point corresponding to one first pivot point in the second signal domain has a multiple of the target interval value, the target mapped pivot point and a next mapped pivot point can be located in a same second segment; and otherwise, the target mapped pivot point and the next mapped pivot point cannot be located in the same second segment.