-
公开(公告)号:US11228780B2
公开(公告)日:2022-01-18
申请号:US17032416
申请日:2020-09-25
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
IPC: H04N19/513 , H04N19/176 , H04N19/182 , H04N19/169 , H04N19/523 , H04N19/80
Abstract: A method for inter prediction of a sample value of a current pixel of a plurality of pixels of a current block of a current frame of a video signal is proposed. The method (800) comprises: determining (801) a plurality of block-wise motion vectors related one-to-one to a plurality of blocks of the current frame; determining (803) a pixel-wise motion vector of the current pixel based on the plurality of block-wise motion vectors; determining one or more reference pixels in the reference frame based on the pixel-wise motion vector of the current pixel; and determining (805) an inter predicted sample value of the current pixel based on one or more sample values of the one or more reference pixels in the reference frame. Coding efficiency is thus improved.
-
公开(公告)号:US10834416B2
公开(公告)日:2020-11-10
申请号:US15934043
申请日:2018-03-23
Applicant: Huawei Technologies Co., Ltd.
IPC: H04N19/51 , H04N19/11 , H04N19/82 , H04N19/50 , H04N19/523 , H04N19/117 , H04N19/176 , H04N19/44 , G06T5/00
Abstract: The present disclosure relates to a video coder for predictive coding a video stream of subsequent frames according to motion compensation into an encoded video bit stream, comprising a frame buffer adapted to store at least one reference frame of the video stream, the reference frame being different from a current frame of the video stream, an inter prediction unit adapted to generate a prediction block of a current block of the current frame from a reference block of the reference frame, and an adaptive sharpening filter configured to adaptively filter the prediction block.
-
公开(公告)号:US20250008128A1
公开(公告)日:2025-01-02
申请号:US18884321
申请日:2024-09-13
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor: Timofey Mikhailovich Solovyev , Esin Koyuncu , Alexander Alexandrovich Karabutov , Maxim Borisovitch Sychev , Mikhail Vyacheslavovich Sosulnikov , Sergey Yurievich Ikonin , Elena Alexandrovna Alshina
IPC: H04N19/189 , H04N19/13 , H04N19/42
Abstract: A neural network including at least one neural network layer and an activation function connected to an output of the at least one neural network layer. The activation function is implemented as an approximation function of a mathematically defined real valued non-linear activation function, wherein the approximation function allows for integer-only processing of fixed-point representations of input values of the approximation function.
-
公开(公告)号:US20250005331A1
公开(公告)日:2025-01-02
申请号:US18885411
申请日:2024-09-13
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor: Timofey Mikhailovich Solovyev , Sergey Yurievich Ikonin , Elena Alexandrovna Alshina , Johannes Sauer , Esin Koyuncu , Maxim Borisovitch Sychev , Alexander Alexandrovich Karabutov , Mikhail Vyacheslavovich Sosulnikov , Kirill Igorevich SOLODSKIKH , Vladimir Mikhailovich Kryzhanovskiy , Alexander Nikolaevich Filippov
IPC: G06N3/0455 , G06T9/00
Abstract: The present disclosure relates to a method of operating a neural network with clipped input data. The method includes defining lower and upper threshold values for integer numbers in data entities of input data for at least one neural network layer. If a value of an integer number in a data entity of the input data is smaller than the defined lower threshold value, the method includes clipping the value of the integer number comprised in the data entity of the input data to the defined lower threshold value. If a value of an integer number in a data entity of the input data is larger than the defined upper threshold value, the method includes clipping the value of the integer number comprised in the data entity of the input data to the defined upper threshold value. Integer overflow of an accumulator register is thereby avoided.
-
公开(公告)号:US20250005330A1
公开(公告)日:2025-01-02
申请号:US18883907
申请日:2024-09-12
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor: Timofey Mikhailovich Solovyev , Sergey Yurievich Ikonin , Elena Alexandrovna Alshina , Esin Koyuncu , Maxim Borisovitch Sychev , Alexander Alexandrovich Karabutov , Mikhail Vyacheslavovich Sosulnikov
IPC: G06N3/0455
Abstract: A method of operating a neural network based on conditioned weights includes defining integer lower and upper threshold values for values of integer numbers comprised in data entities of input data for the neural network layer. If a value of an integer numbers comprised in a data entity of the input data is smaller than the lower threshold value, the value of the integer number comprised in the data entity of the input data is clipped to the lower threshold value, or if a value of an integer number comprised in a data entity of the input data is larger than the upper threshold value, the value of the integer number comprised in the data entity of the input data is clipped to the upper threshold value. Integer valued weights are determined based on the lower threshold value, the upper threshold value, and a pre-defined accumulator register size, such that integer overflow of the accumulator register can be avoided.
-
16.
公开(公告)号:US12063369B2
公开(公告)日:2024-08-13
申请号:US17701834
申请日:2022-03-23
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor: Maxim Borisovitch Sychev
IPC: H04N19/146 , H04N19/172 , H04N19/184 , H04N19/46 , H04N19/70
CPC classification number: H04N19/146 , H04N19/172 , H04N19/184 , H04N19/46 , H04N19/70
Abstract: A method of coding implemented by an encoding device for encoding video data is disclosed. The method comprises reducing at least part of Least Significant Bits of Picture Order Counter (POC_LSB) in the video data by omitting one bit of the POC_LSB or by omitting a number of bits of the POC_LSB depending on the layer IDs of the video data. The method further includes coding the reduced POC_LSB.
-
公开(公告)号:US12047609B2
公开(公告)日:2024-07-23
申请号:US17707202
申请日:2022-03-29
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor: Timofey Mikhailovich Solovyev , Maxim Borisovitch Sychev , Huanbang Chen , Alexander Alexandrovich Karabutov , Roman Igorevich Chernyak , Sergey Yurievich Ikonin , Haitao Yang , Elena Alexandrovna Alshina
IPC: H04N19/82 , H04N19/105 , H04N19/109 , H04N19/117 , H04N19/132 , H04N19/139 , H04N19/176 , H04N19/182 , H04N19/196 , H04N19/52 , H04N19/80
CPC classification number: H04N19/80 , H04N19/105 , H04N19/109 , H04N19/117 , H04N19/132 , H04N19/139 , H04N19/176 , H04N19/182 , H04N19/197 , H04N19/52
Abstract: An apparatus for inter prediction of a sample value of a current full-integer pixel of a plurality of pixels associated with a current block of a current frame of a video signal is disclosed. The apparatus is configured to: determine a motion vector of the current full-integer pixel based on a motion compensation model; determine for the current full-integer pixel a corresponding sub-integer pixel in a reference frame; determine positions of a set of corresponding filter support pixels in the reference frame; obtain a respective sample value of the corresponding sub-integer pixel of the current full-integer pixel and the corresponding filter support pixels in the reference frame by applying a bilinear interpolation; and determine the inter predicted sample value of the current full-integer pixel.
-
公开(公告)号:US12034954B2
公开(公告)日:2024-07-09
申请号:US17701388
申请日:2022-03-22
Applicant: HUAWEI TECHNOLOGIES CO., LTD.
Inventor: Maxim Borisovitch Sychev
IPC: H04N19/146 , H04N19/172 , H04N19/46 , H04N19/70
CPC classification number: H04N19/46 , H04N19/146 , H04N19/172 , H04N19/70
Abstract: A coding method implemented by a decoding/encoding device for coding video data comprises coding, in a supplemental enhancement information (SEI) message, a set of sequence parameters. The set of sequence parameters is associated with temporal IDs of the sub-layers of a bitstream or coded layer video sequence (C(L)VS). The coding method further comprises processing the bitstream or C(L)VS by using the set of sequence parameters included in the SEI message. The set of sequence parameters is used for deriving or representing a picture rate of the sequence.
-
公开(公告)号:US20230209069A1
公开(公告)日:2023-06-29
申请号:US18178196
申请日:2023-03-03
Applicant: Huawei Technologies Co., Ltd.
Inventor: Ye-Kui Wang , FNU Hendry , Maxim Borisovitch Sychev
IPC: H04N19/174 , H04N19/119 , H04N19/172 , H04N19/30 , H04N19/70 , H04N19/44 , H04N19/66
CPC classification number: H04N19/174 , H04N19/119 , H04N19/172 , H04N19/30 , H04N19/70 , H04N19/44 , H04N19/66 , H04N19/176
Abstract: A video coding mechanism is disclosed. The mechanism includes partitioning a picture into a plurality of first level tiles. A subset of the first level tiles is partitioned into a plurality of second level tiles. The first level tiles and the second level tiles are encoded into a bitstream. A split indication is encoded into the bitstream. The split indication indicates that at least one of the first level tiles is split into the second level tiles. The bitstream is stored for communication toward a decoder.
-
公开(公告)号:US20210329264A1
公开(公告)日:2021-10-21
申请号:US17357724
申请日:2021-06-24
Applicant: Huawei Technologies Co., Ltd.
Inventor: Yei-Kui Wang , FNU Hendry , Maxim Borisovitch Sychev
IPC: H04N19/174 , H04N19/119 , H04N19/30 , H04N19/172
Abstract: A video coding mechanism is disclosed. The mechanism includes partitioning a picture into a plurality of first level tiles. A subset of the first level tiles is partitioned into a plurality of second level tiles. Each second level tile contains a single rectangular slice of picture data. The first level tiles and the second level tiles are encoded into a bitstream. The bitstream is stored for communication toward a decoder.
-
-
-
-
-
-
-
-
-