公开(公告)号：US11064192B2

公开(公告)日：2021-07-13

申请号：US16592005

申请日：2019-10-03

Applicant: QUALCOMM Incorporated

Inventor： Yu Han ,  Han Huang ,  Wei-Jung Chien ,  Marta Karczewicz

IPC: H04N19/176 ,  H04N19/159 ,  H04N19/137 ,  H04N19/105

Abstract: A device and method for coding video data determines a spatial-temporal motion vector predictor (STMVP) so that the STMVP does not need to be motion vector scaled. The device may determine a candidate list. The device may then determine which candidates in the candidate list have the same reference picture. The device may then generate the STMVP based on the candidates in the candidate list determined to have the same reference picture. The device may then code a current block of video data using the STMVP.

公开(公告)号：US11057636B2

公开(公告)日：2021-07-06

申请号：US16572387

申请日：2019-09-16

Applicant: QUALCOMM Incorporated

Inventor： Han Huang ,  Wei-Jung Chien ,  Marta Karczewicz

IPC: H04N19/176 ,  H04N19/16 ,  H04N19/105 ,  H04N19/52 ,  H04N19/96 ,  H04N19/46 ,  H04N19/50 ,  H04N19/15 ,  H04N19/169 ,  G06T3/00

Abstract: Systems and techniques for improving costs and efficiency in affine motion prediction of a current block includes storing motion information of previously coded block in a line buffer or one or more local buffers of a coding unit such as coding tree unit (CTU). The line buffer can store motion information of a line of coding blocks located at a boundary of the CTU, while the local buffers can be utilized for storing motion information of coding blocks located anywhere in the CTU. The utilization of the line buffer can be improved by limiting the number of motion vectors of a neighboring block stored in the line buffer and obtaining remaining motion vectors from an additional block when available. The utilization of local buffers can be improved by restricting a size of neighboring blocks whose motion information is stored in the local buffers.

公开(公告)号：US11051034B2

公开(公告)日：2021-06-29

申请号：US16593388

申请日：2019-10-04

Applicant: QUALCOMM Incorporated

Inventor： Yu Han ,  Wei-Jung Chien ,  Han Huang ,  Chao-Hsiung Hung ,  Marta Karczewicz

IPC: H04N7/12 ,  H04N19/52 ,  H04N19/176 ,  H04N19/56 ,  H04N19/132 ,  H04N19/105 ,  H04N19/159 ,  H04N19/172

Abstract: Systems and techniques for processing video data include a history-based motion vector predictor (HMVP) table with two or more HMVP candidates, where a first entry of the HMVP table associated with a first index value includes a first HMVP candidate and a second entry of the HMVP table associated with a second index value includes a second HMVP candidate, the first index value being lower than the second index value. For an advanced motion vector prediction (AMVP) candidate list, one or more HMVP candidates from the HMVP table are selected in a reverse order, where the second HMVP candidate is selected before the first HMVP candidate according to the reverse order. The selected one or more HMVP candidates are added to the AMVP candidate list. The AMVP candidate list can be used to perform AMVP for the one or more blocks of video data.

公开(公告)号：US11019332B2

公开(公告)日：2021-05-25

申请号：US16829626

申请日：2020-03-25

Applicant: QUALCOMM Incorporated

Inventor： Hongtao Wang ,  Han Huang ,  Yu Han ,  Geert Van der Auwera ,  Wei-Jung Chien ,  Marta Karczewicz

IPC: H04N19/11 ,  H04N19/139 ,  H04N19/132 ,  H04N19/159 ,  H04N19/70 ,  H04N19/186 ,  H04N19/176

Abstract: A video coder selects a set of wide-angle intra prediction directions based on a size of a luma block of a picture having a YUV 4:2:2 chroma sampling format. Additionally, the video coder determines an intra prediction direction for the luma block. The intra prediction direction for the luma block is in the set of wide-angle intra prediction directions. The video coder also determines an intra prediction direction for a chroma block. The luma block is collocated in the picture with the chroma block. The chroma block has a different width/height ratio than the luma block. The intra prediction direction for the chroma block is guaranteed to have the intra prediction direction for the luma block. The video coder uses the intra prediction directions for the luma and chroma blocks to generate prediction blocks for the luma and chroma blocks, respectively.

公开(公告)号：US20210152825A1

公开(公告)日：2021-05-20

申请号：US17078710

申请日：2020-10-23

Applicant: QUALCOMM Incorporated

Inventor： Kevin Pascal Andre Reuze ,  Han Huang ,  Chun-Chi Chen ,  Po-Han Lin ,  Wei-Jung Chien ,  Vadim Seregin ,  Marta Karczewicz

IPC: H04N19/119 ,  H04N19/167 ,  G06F1/03

Abstract: An example device for decoding video data includes a memory configured to store the video data and one or more processors implemented in circuitry and coupled to the memory. The one or more processors are configured to determine a first distance index associated with a first geometric partition mode (GEO) angle for a first prediction unit (PU) of the video data to be 4. The one or more processors are configured to determine a first displacement value based on the first distance index, the first displacement value being indicative of a distance from a center of the first PU to a GEO split. The one or more processors are configured to decode the first PU based on the first GEO angle and the first displacement value. The first displacement value is half of a displacement value associated with a distance index of 2.

公开(公告)号：US10999596B2

公开(公告)日：2021-05-04

申请号：US16443130

申请日：2019-06-17

Applicant: QUALCOMM Incorporated

Inventor： Han Huang ,  Wei-Jung Chien ,  Marta Karczewicz

IPC: H04B1/66 ,  H04N7/12 ,  H04N11/02 ,  H04N11/04 ,  H04N19/513 ,  H04N19/44

Abstract: A method of decoding video data includes receiving a plurality of blocks of video data, wherein each block of the plurality of blocks is encoded using a respective affine motion model of a plurality of affine motion models, and decoding the plurality of blocks of video data using the same affine motion field derivation process for each of the plurality of affine motion models.

公开(公告)号：US20210058617A1

公开(公告)日：2021-02-25

申请号：US16998639

申请日：2020-08-20

Applicant: QUALCOMM Incorporated

Inventor： Kevin Pascal Andre Reuze ,  Han Huang ,  Vadim Seregin ,  Marta Karczewicz

IPC: H04N19/119 ,  H04N19/132 ,  H04N19/105 ,  H04N19/176

Abstract: A method of decoding video data includes determining that a geometric partition mode is enabled for a current block of the video data and determining a split line dividing the current block into a first partition and a second partition, where determining the split line comprises selecting an angle for the split line from a plurality of angles, Each angle of the plurality of angles corresponding to an N:M ratio of samples of the current block, where N and M are integers. The split line is not at a corner of the current block. The method further includes determining geometric mode weights for the current block using the angle of the split line, generating a first prediction block using motion information for the first partition, and generating a second prediction block using motion information for the second partition.

公开(公告)号：US20200314443A1

公开(公告)日：2020-10-01

申请号：US16836013

申请日：2020-03-31

Applicant: QUALCOMM Incorporated

Inventor： Han Huang ,  Wei-Jung Chien ,  Marta Karczewicz

IPC: H04N19/513

Abstract: This disclosure describes gradient-based prediction refinement. A video coder (e.g., video encoder or video decoder) determines one or more prediction blocks for inter-predicting a current block (e.g., based on one or more motion vectors for the current block). In gradient-based prediction refinement, the video coder modifies one or more samples of the prediction block based on various factors such as displacement in a horizontal direction, the horizontal gradient, a displacement in the vertical direction, and a vertical gradient. This disclosure provides for gradient-based prediction refinement where a precision level of the displacement (e.g., at least one of the horizontal or vertical displacement) is unified (e.g., the same) for different prediction modes.

公开(公告)号：US20200296405A1

公开(公告)日：2020-09-17

申请号：US16809162

申请日：2020-03-04

Applicant: QUALCOMM Incorporated

Inventor： Han Huang ,  Wei-Jung Chien ,  Marta Karczewicz

IPC: H04N19/513 ,  H04N19/137 ,  H04N19/132 ,  H04N19/117 ,  H04N19/176 ,  G06F17/16

Abstract: A video encoder and/or video decoder may predict a subblock of a block of video data using affine motion compensation and determine spatial gradient information for a sample of the subblock of the block of the video data. The video encoder and/or video decoder may determine a difference motion vector indicating a difference between a pixel motion vector computed for the sample and a subblock motion vector for the subblock. To determine the difference motion vector, the video encoder and/or video decoder may constrain an absolute value of a horizontal component of the difference motion vector to be less than half of a pixel of the video data and to constrain an absolute value of a vertical component of the difference motion vector to be less than half of the pixel of the video data.

公开(公告)号：US20200296367A1

公开(公告)日：2020-09-17

申请号：US16813508

申请日：2020-03-09

Applicant: QUALCOMM Incorporated

Inventor： Luong Pham Van ,  Geert Van der Auwera ,  Adarsh Krishnan Ramasubramonian ,  Han Huang ,  Marta Karczewicz

IPC: H04N19/119 ,  H04N19/96 ,  H04N19/186 ,  H04N19/176

Abstract: A method of decoding video data includes determining, by one or more processors implemented in circuitry, a partition of the video data into a plurality of blocks. The partition of the video data applies a block size restriction to prevent a splitting of a block of the plurality of blocks that would result in a small block comprising a block width and a block height when the block height times the block width is less than a threshold. The method further includes generating, by the one or more processors, prediction information for the block and determining, by the one or more processors, a predicted block for the block based on the prediction information. The method further includes decoding, by the one or more processors, a residual block for the block and combining, by the one or more processors, the predicted block and the residual block to decode the block.