公开(公告)号：US10097826B2

公开(公告)日：2018-10-09

申请号：US15864493

申请日：2018-01-08

Applicant: Huawei Technologies Co., Ltd.

Inventor： Sixin Lin ,  Mingyuan Yang ,  Haoping Yu ,  Jiantong Zhou ,  Cunbin Pan

IPC: H04N7/12 ,  H04N19/109 ,  G06T9/00 ,  H04N19/56 ,  H04N19/149 ,  H04N19/117 ,  H04N19/53 ,  H04N19/523 ,  H04N19/577

Abstract: A method and device for generating a predicted value of image to generate a predicted value of a current block during image encoding or decoding is disclosed, where the method includes determining a searching scope, wherein a plurality of motion vectors are included in the searching scope, performing up-sampling interpolations on first reference blocks corresponding to the motion vector in the searching scope, in a reference image of the current block using a first filter to obtain up-sampled first reference blocks, obtaining, using the up-sampled first reference blocks, at least one candidate motion vector corresponding to the current block, performing up-sampling interpolations on second reference blocks, corresponding to the at least one candidate motion vector, in the reference image of the current block using a second filter to obtain up-sampled second reference blocks, combining the up-sampled second reference blocks to obtain a predicted value of the current block.

公开(公告)号：US09888245B2

公开(公告)日：2018-02-06

申请号：US15165517

申请日：2016-05-26

Applicant: Huawei Technologies Co., Ltd.

Inventor： Jiali Fu ,  Jiantong Zhou ,  Sixin Lin

IPC: G06K9/36 ,  H04N19/154 ,  H04N19/176 ,  H04N19/132 ,  H04N19/14 ,  H04N19/48 ,  H04N19/124 ,  H04N19/18 ,  H04N19/40

CPC classification number: H04N19/154 ,  H04N19/124 ,  H04N19/132 ,  H04N19/14 ,  H04N19/176 ,  H04N19/18 ,  H04N19/40 ,  H04N19/48

Abstract: The present invention relates to an image compression method and apparatus, where the image compression method includes a step of performing amplitude decreasing processing on a frequency domain coefficient or a quantization coefficient of a to-be-processed image. The image compression method includes: determining a texture direction of the to-be-processed image; and performing amplitude decreasing processing on the frequency domain coefficient or the quantization coefficient of the to-be-processed image according to the texture direction, where the frequency domain coefficient is a coefficient obtained after the image is transformed, and the quantization coefficient is a coefficient obtained after the frequency domain coefficient is quantized. According to embodiments of the present invention, amplitude decreasing processing is performed on a frequency domain coefficient of a to-be-processed image according to a texture direction of the to-be-processed image, which can improve the compression efficiency without affecting subjective quality of the to-be-processed image.

公开(公告)号：US09743107B2

公开(公告)日：2017-08-22

申请号：US15186180

申请日：2016-06-17

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Sixin Lin ,  Lianhuan Xiong ,  Yingjia Liu ,  Steffen Kamp ,  Mathias Wien

IPC: H04N19/56 ,  H04N19/513 ,  H04N19/176 ,  H04N19/567

CPC classification number: H04N19/56 ,  H04N19/176 ,  H04N19/513 ,  H04N19/567

Abstract: A method, an apparatus and a system for a rapid motion search applied in template matching are disclosed. The method includes: selecting motion vectors of blocks related to a current block as candidate motion vectors of the current block; after the uniqueness of a series of the candidate motion vectors of the current block is maintained, calculating the cost function of the candidate motion vectors in a corresponding template area of a reference frame, and obtaining the motion vector of the best matching template from the candidate motion vectors of the current block. In the embodiments of the present invention, there is no need to determine a large search range and no need to determine the corresponding search path template, and it is only necessary to perform a search in a smaller range.

公开(公告)号：US20160277746A1

公开(公告)日：2016-09-22

申请号：US15165517

申请日：2016-05-26

Applicant: Huawei Technologies Co., Ltd.

Inventor： Jiali Fu ,  Jiantong Zhou ,  Sixin Lin

IPC: H04N19/154 ,  H04N19/18 ,  H04N19/124

CPC classification number: H04N19/154 ,  H04N19/124 ,  H04N19/132 ,  H04N19/14 ,  H04N19/176 ,  H04N19/18 ,  H04N19/40 ,  H04N19/48

Abstract: The present invention relates to an image compression method and apparatus, where the image compression method includes a step of performing amplitude decreasing processing on a frequency domain coefficient or a quantization coefficient of a to-be-processed image. The image compression method includes: determining a texture direction of the to-be-processed image; and performing amplitude decreasing processing on the frequency domain coefficient or the quantization coefficient of the to-be-processed image according to the texture direction, where the frequency domain coefficient is a coefficient obtained after the image is transformed, and the quantization coefficient is a coefficient obtained after the frequency domain coefficient is quantized. According to embodiments of the present invention, amplitude decreasing processing is performed on a frequency domain coefficient of a to-be-processed image according to a texture direction of the to-be-processed image, which can improve the compression efficiency without affecting subjective quality of the to-be-processed image.

Abstract translation: 图像压缩方法和装置技术领域本发明涉及图像压缩方法和装置，其中图像压缩方法包括对频域系数或待处理图像的量化系数执行幅度减小处理的步骤。 图像压缩方法包括：确定待处理图像的纹理方向; 并且根据纹理方向对频域系数或待处理图像的量化系数进行幅度减小处理，其中频域系数是在图像被变换之后获得的系数，并且量化系数是系数 在频域系数被量化之后获得。 根据本发明的实施例，根据待处理图像的纹理方向对被处理图像的频域系数进行幅度降低处理，这可以提高压缩效率而不影响主观质量 的待处理图像。

公开(公告)号：US09324165B2

公开(公告)日：2016-04-26

申请号：US13738773

申请日：2013-01-10

Applicant: Huawei Technologies Co., Ltd.

Inventor： Sixin Lin ,  Mingyuan Yang ,  Haoping Yu ,  Jiantong Zhou ,  Cunbin Pan

IPC: H04N7/12 ,  G06T9/00 ,  H04N19/56 ,  H04N19/149 ,  H04N19/109 ,  H04N19/117 ,  H04N19/53 ,  H04N19/523 ,  H04N19/577

CPC classification number: H04N19/109 ,  G06T9/004 ,  H04N19/117 ,  H04N19/149 ,  H04N19/523 ,  H04N19/53 ,  H04N19/56 ,  H04N19/577

Abstract: A method and device for generating a predicted value of image that are mostly used to generate a predicted value of a current block during image encoding or decoding. The method includes: determining a searching scope, wherein multiple motion vectors are included in the searching scope; performing up-sampling interpolations on first reference blocks, corresponding to the motion vector in the searching scope, in a reference image of the current block by using a first filter to obtain up-sampled first reference blocks; by using the up-sampled first reference blocks, obtaining at least one candidate motion vector corresponding to the current block; performing up-sampling interpolations on second reference blocks, corresponding to the at least one candidate motion vector, in the reference image of the current block by using a second filter to obtain up-sampled second reference blocks; combining the up-sampled second reference blocks to obtain a predicted value of the current block.

Abstract translation: 一种用于在图像编码或解码期间生成主要用于生成当前块的预测值的图像的预测值的方法和装置。 该方法包括：确定搜索范围，其中多个运动矢量包括在搜索范围内; 通过使用第一滤波器来获取上采样的第一参考块，在当前块的参考图像中对与搜索范围中的运动矢量相对应的第一参考块执行向上采样内插; 通过使用上采样的第一参考块，获得与当前块相对应的至少一个候选运动矢量; 通过使用第二滤波器来获取上采样的第二参考块，在当前块的参考图像中对应于至少一个候选运动矢量的第二参考块上执行上取样内插; 组合上采样的第二参考块以获得当前块的预测值。

公开(公告)号：US09288492B2

公开(公告)日：2016-03-15

申请号：US13947681

申请日：2013-07-22

Applicant: Huawei Technologies Co., Ltd.

Inventor： Sixin Lin ,  Mingyuan Yang ,  Jiantong Zhou ,  Haoping Yu

IPC: H04N19/52 ,  H04N19/533 ,  H04N19/56 ,  H04N19/523

CPC classification number: H04N19/533 ,  H04N19/52 ,  H04N19/523 ,  H04N19/56

Abstract: Embodiments of the present invention provide a motion prediction or compensation method during a video coding and decoding process. A motion prediction or compensation method provided in the embodiments of the present invention includes: acquiring a candidate motion vector set; conducting a refined search for an optimal motion vector based on information of the candidate motion vector set; and performing motion prediction or compensation by using the motion vector acquired by the refined search. This improves coding performance and meanwhile maintains reasonable complexity.

Abstract translation: 本发明的实施例在视频编码和解码过程中提供运动预测或补偿方法。 本发明实施例提供的运动预测或补偿方法包括：获取候选运动矢量集; 基于候选运动矢量集的信息进行精确搜索最优运动矢量; 以及通过使用由精细搜索获得的运动矢量来执行运动预测或补偿。 这提高了编码性能，同时保持了合理的复杂性。