Huawei Technologies Co., Limited

Huawei Device Co., Ltd.

Hai Chen

Meng Li

Jianhua Zheng

Xiaozhen Zheng

Hu Liu

Fengwei Zhang

Zhenjiang Dong

Chio In Ieong

Yu Liu

Xiao Xiao

Haibo Liu

Chen-Xiong Zhang

Lingzhi Liu

Quanhe Yu

Yan Wei Zhang

Weisi Lin

Shuiping Li

Zheyu Liu

Chenglin Zheng

Fei Qiao

Qi Wei

Kaige Jia

Meijun Wan

Anmin Liu

Haiqin Yang

A matrix processing method performed by a graphics processing unit (GPU) includes: determining a plurality of non-zero elements in a to-be-processed matrix at a processor in the GPU; generating a distribution matrix of the to-be-processed matrix at the processor, where the distribution matrix comprises identities for indicating positions of the plurality of non-zero elements in the to-be-processed matrix; obtaining a target matrix from another matrix by using the distribution matrix at a logic circuit in the processor, where the target matrix comprises a plurality of target elements from the another matrix; and performing matrix processing on the plurality of non-zero elements and the target matrix to obtain an operation result at the processor.

Matrix processing method and apparatus, and logic circuit

A video coding mechanism is disclosed. The mechanism includes at least one pre-encoder configured to pre-encode a received input video signal, and output a pre-encoded video signal. The mechanism also includes an encoder configured to obtain the pre-encoded video signal from the pre-encoder, and encode the pre-encoded video signal as a bitstream for transmitting the bitstream to decoder for decoding.

Video coding with successive codecs

A matrix processing method includes: determining a quantity of non-zero elements in a to-be-processed matrix, where the to-be-processed matrix is a one-dimensional matrix; generating a distribution matrix of the to-be-processed matrix, where the distribution matrix is used to indicate a position of a non-zero element in the to-be-processed matrix; combining the quantity of non-zero elements, values of all non-zero elements in the to-be-processed matrix arranged sequentially, and the distribution matrix, to obtain a compressed matrix of the to-be-processed matrix.

An image processing method and apparatus, where the method includes determining a maximum value in nonlinear primary color values of all components of each pixel of a first to-be-processed image, determining dynamic parameters of a first transfer function, converting the maximum value of each pixel into a transfer value based on the first transfer function for which the dynamic parameters are determined, calculating a ratio between the transfer value and the maximum value of each pixel, and adjusting a dynamic range for the nonlinear primary color values of all the components of each pixel based on the ratio to obtain nonlinear primary color values of all components of each corresponding pixel of a first target image.

Image processing method and apparatus for adjusting a dynamic range of an image

An image processing method and apparatus are disclosed, to improve quality of an image on which color processing is performed. The method includes: determining color values of N color components of each pixel of a to-be-processed image, where N is a natural number greater than 1; determining N ratios of a luminance value of each pixel to the color values of the N color components; determining N first color adjustment coefficients based on the N ratios; and performing color processing on each pixel based on the N first color adjustment coefficients and the color values of the N color components to obtain a target image.

Image processing method and apparatus

An image processing apparatus, which includes a first physical computing circuit, configured to receive a plurality of first analog signals output by an image sensor, and perform a convolution operation on the plurality of first analog signals to obtain a second analog signal. The plurality of first analog signals are in a one-to-one correspondence with a plurality of pieces of pixel data of a to-be-recognized image. The first physical computing circuit comprises at least one multiplication circuit array and at least one subtraction circuit, the at least one multiplication circuit array is in a one-to-one correspondence with the at least one subtraction circuit, a multiplication circuit in each multiplication circuit array comprises a differential pair transistor, each multiplication circuit array implements the convolution operation on the plurality of first analog signals using a plurality of multiplication circuits and a corresponding subtraction circuit.

Image processing apparatus

A method and an apparatus for processing a high dynamic range (HDR) image, and a terminal device to improve quantization quality, where the method includes obtaining brightness information of an image, processing the brightness information to obtain processed image information, quantizing the processed image information to obtain quantized image information, and encoding the quantized image information to obtain encoded image information.

Method and apparatus for processing high dynamic range image, and terminal device

The present disclosure discloses a method and an apparatus for processing image signal conversion. In one example method, an input primary color signal is obtained. The primary color signal is a numeric value of an optical signal corresponding to an image. The primary color signal is proportional to light intensity. Conversion processing is performed on the primary color signal to obtain processed image information. The image information is a numeric expression value of the image. The conversion processing includes at least the following processing: where a, b, m, and p are rational numbers, L is the input primary color signal, and L' is the processed image information.

Method and apparatus for processing image signal conversion, and terminal device

Embodiments of the present disclosure provide an image encoding method, an image decoding method, an encoding device, and a decoding device. The method includes: determining a conversion function parameter of a current image block; determining a conversion function parameter of a neighboring image block corresponding to the current image block; determining, according to a correlation between the conversion function parameter of the current image block and the conversion function parameter of the neighboring image block, whether to perform merge encoding on the conversion function parameter of the current image block and the conversion function parameter of the neighboring image block, and determining a first indicator, where the first indicator is used to indicate whether to perform merge encoding on the conversion function parameter of the current image block and the conversion function parameter of the neighboring image block; and encoding the first indicator.

Image encoding method, image decoding method, encoding device, and decoding device

Determining of still/movement may be performed with reference to quantization noise of a first section to which a first pixel belongs, and for different results of determining of whether the first pixel is in a movement area or a still area, different frame difference thresholds applicable to the movement area and the still area are separately set, and different frame difference calculation manners are used, different blending coefficients applicable to the movement area and the still area are selected according to the different frame difference thresholds applicable to the movement area and the still area and the frame difference calculation manners, and a noise reduction blending manner is selected according to the different blending coefficients applicable to the movement area and the still area, the frame difference calculation manners, and a pixel value of the first pixel in a current frame.

Temporal noise reduction method for noisy image and related apparatus

An image processing method and apparatus are provided. The image processing method includes collecting at least two exposure frames with different brightness in a same scene during different exposure time; combining, for each exposure frame, a raw data unit arranged repeatedly in the exposure frame to obtain first brightness data after the combining; acquiring a correction parameter of all exposure frames according to all first brightness data and performing weighting processing on all the exposure frames by using the correction parameter to obtain a high dynamic range (HDR) image of corrected raw data. The foregoing method can resolve a problem in the prior art that colors, brightness, and contrast of an image obtained from Raw data are severely distorted.

Forming high-dynamic-range (HDR) images using single-channel data

A video processing method and apparatus, which relate to the field of multimedia, and the method includes acquiring, by an video processing apparatus, two field images of an interlaced frame image in an interlaced video; performing temporal deinterlacing on either field image in the two field images to obtain a temporal virtual field image, and performing spatial deinterlacing on the either field image to obtain a spatial virtual field image; performing directional combing detection on the temporal virtual field image along a preset candidate combing detection direction with reference to the other field image in the interlaced frame image, to determine whether combing exists in the temporal virtual field image; and according to this, selecting the temporal virtual field image or the spatial virtual field image to be fused with the other field image to generate a new image.

Video processing method and apparatus

A method and an apparatus for correcting a multi-exposure motion image are disclosed, where the method includes determining a luminance mapping function, where the luminance mapping function is a luminance mapping relationship between a reference frame and multiple extended frames; mapping a luminance of an extended frame by using the luminance mapping relationship to obtain a virtual frame; calculating a global motion vector between the virtual frame and the reference frame; correcting a pixel of the extended frame according to the global motion vector to obtain a first extended frame after correction; detecting a pixel of the first extended frame according to the reference frame to obtain a local error pixel of the first extended frame; and correcting a luminance of the local error pixel of the first extended frame to obtain a second extended frame after correction.

Method and apparatus for correcting multi-exposure motion image

An image processing method and apparatus, and an imaging device are provided. The method includes: obtaining a first green component value and a first red component value or first blue component value that are of the current pixel and obtained by using an interpolation algorithm, a minimum first-gradient value in horizontal and vertical directions, a minimum second-gradient value in inclined directions, a threshold for a green component, and a threshold for a red component or blue component; obtaining a corrected value of the green component and a corrected value of the red component or blue component of the current pixel; and obtaining a second green component value and a second red component value or second blue component value of the current pixel. The method may be used to reduce crosstalk of a color component in which the current pixel is located on other two color components obtained by interpolation.

Image processing method and apparatus, and imaging device

Embodiments of the present invention provide a method and an apparatus for acquiring a weight coefficient of a digital filter so as to enhance a processing effect of images or videos and reduce the complexity of operations. The method includes: extracting global block features and local block features of image blocks, where the image blocks include a first image block and a second image block; acquiring an image block distance fbetween the first image block and the second image block according to a global block feature and local block features of the first image block as well as a global block feature and local block features of the second image block; and evaluating a value of e f.

Growing community of inventors

Shenzhen, China

Hai Chen