Huawei Technologies Co., Limited

Tencent Technology (shenzhen) Company Limited

Shenzhen Goodix Technology Co., Ltd.

Guangdong Oppo Mobile Telecommunications Corp., Ltd.

Cirrus Data Solutions, Inc.

Guangdong University of Technology

Zhejiang University

Shenzhen Royole Technologies Co., Ltd.

Sz Dji Technology Co., Ltd.

Vertiv Tech Co., Ltd.

Xi'an Zte New Software Company Limited

Shenzhen Snapmaker Technologies Co., Ltd.

Ningbo Opr Offshore Engineering Equipment Co., Ltd.

Shenzhen Xft Medical Limited

Zte Corporation

Chang Liu

Wai T Lam

Wayne K Lam

Wangwang Yang

Haixiang Wang

Yi Fang

Renquan Lu

Yong Xu

Jianxiong Wu

Hongxia Rao

Guojun Han

Zhiyao Liu

Hongliang Duan

Hongxing Wang

Xiang Gao

Yu Zhang

Mingxi Wang

Kai Feng Li

Lei Wang

Huimin Zhang

You Li

Xin Xu

Hui Liu

Tao Wang

Rui Li

Tao Xue

Haijun Wu

Zhefeng Yan

Zhaohui Li

Libin Zhang

Gengchun Deng

Lei Mao

Lei Zhang

Yunfei Zhang

Chao Weng

Jun Cai

Lei Zhao

Yinfeng Zhang

Xiaodong Chen

Tingqiu Yuan

Chenghang Zheng

Yang Zhang

Weihong Wu

Zefei Li

Hexing Liu

Tao Liu

Jiang Rui

Yongxin Zhang

Chao Wan

Yiwei Tao

ZiJie Chen

Shuxin Zhang

Zirong Zhu

Mo Zhao

Jie Tao

Huiqin Yang

Xiuling Li

Can Zhou

Xuedong Chen

Yongqiang Zou

Chaohui Ding

Xindong Shi

Xiao Zhang

Hongmin Wei

Chong Li

Junchao Du

Yi Zhu

Zhijie Xie

Junhua Li

Shipeng Zhang

Jie Guo

Yonglin Luo

Hui Peng

Guangyu Yang

Jianrong Qiu

Hua Mao

Zhaohui Ren

Xiaofeng Liu

Wenge Xiao

Weiguo Weng

Qingrui Han

Zongzheng Zhang

Chunlong Zhang

Haobo Zhang

Jiazhi Ni

Keji Wang

Yongmin Luo

Fengli Ruan

Hui Chen

Songrong Bai

Yuru Guo

Ye Kuang

Xiaodong Yan

Shengyu Dong

Jiachao Chen

Taifu Que

Shaopeng Yang

Dayong Zhao

Kehui Li

A device receives a computation graph and transforms the computation graph into a dataflow graph comprising recursive subgraphs. Each recursive subgraph comprises a tuple of another recursive subgraph and an operator node, or an empty graph. The device determines a number of partitioning recursions based on a number of parallel computing devices. For each partitioning recursion, the device determines costs corresponding to operator nodes, determines a processing order of the recursive subgraphs, and processes the recursive subgraphs. To process a recursive subgraph, the device selects a partitioning axis for tensors associated with an operator node of the recursive subgraph. The device outputs a partitioning scheme comprising partitioning axes for each tensor associated with the operator nodes.

Parallel computing scheme generation for neural networks

A signal processing apparatus is configured to preprocess a sound wave signal and output a processed audio signal through an electromagnetic wave. The signal processing apparatus includes: a receiving unit, configured to receive at least one sound wave signal; a conversion unit, configured to convert the at least one sound wave signal to at least one audio signal; a positioning unit, configured to determine position information related to the at least one sound wave signal; a processing unit, configured to determine a sending time point of at least one audio signal based on the position information and a first time point, where the first time point is a time point at which the receiving unit receives the at least one sound wave signal; and a sending unit, configured to send the at least one audio signal through the electromagnetic wave.

Signal processing apparatus, method, and system

An image rendering method and a related device thereof are disclosed. The method includes: performing ray tracing based on rendering information of a reflecting object, and storing an obtained image of a reflection in the reflecting object in a first frame buffer, where the first frame buffer is used for storing an image that does not need to be displayed temporarily; performing rasterization separately based on the rendering information of the reflecting object and rendering information of a reflected object, and storing an obtained image of the reflecting object and an obtained image of the reflected object in a second frame buffer, where the second frame buffer is used for storing an image to be displayed; and obtaining, from the first frame buffer, the image of the reflection in the reflecting object, and storing the image of the reflection in the reflecting object in the second frame buffer.

Image rendering method and related device thereof

A camera invocation method includes preview interfaces corresponding to a plurality of applications in an electronic device that simultaneously display an image captured in real time by a camera of the electronic device, where the applications share the camera.

Camera invocation method and system, and electronic device

In an image processing method, for each video image of a plurality of video images of an original video, a first image generated by format conversion of the respective video image is acquired. The respective video image is represented by a first color representation. The first image is represented by a second color representation. An image dynamic range supported by the second color representation is wider than an image dynamic range supported by the first color representation. An image dynamic range of the first image is the same as an image dynamic range of the respective video image. A second image that is based on the respective video image is acquired. The second image is represented by the second color representation. An image dynamic range of the second image is wider than the image dynamic range of the respective video image. A fusion weight value of each pixel of a plurality of pixels in a target image is determined according to a display order of the respective video image in the original video, the target image being one of the first image and the second image. A fusion weight map is generated based on the fusion weight value of each pixel of the plurality of pixels in the target image. A fusion image corresponding to the respective video image is generated based on the target image, a base image, and the fusion weight map, the base image being another one of the first image and the second image. A fusion video is generated based on the fusion images corresponding to the plurality of video images.

Image processing method and apparatus, computer device, and storage medium

Disclosed is an identity authentication system for distributed Internet of vehicles (IoV), including a core cloud, a plurality of edge clouds, a plurality of road side units (RSUs) and a plurality of terminal vehicles. The core cloud stores registration information about the terminal vehicles and the RSUs; the edge cloud performs identity verification on the RSUs according to the registration information, and after the verification is passed, the edge cloud generates a temporary shared session key and sends the same to the RSU and the terminal vehicle, and the RSU and the terminal vehicle establish encrypted communication according to the temporary shared session key, to provide a network communication service for the terminal vehicle. In the present disclosure, a vehicle identity authentication efficiency in a scene with a large traffic density can be effectively improved.

Identity authentication system for distributed Internet of vehicles

A preparation method and use of a yellow fluorescent glass ceramic are disclosed. The preparation method includes: mixing a monomer, a cross-linking agent and a filling solvent evenly, then adding fumed silica and stirring evenly, further adding an ultraviolet (UV) photoinitiator and an UV absorber, and stirring thoroughly; adding a yellow phosphor (Y,Gd)AG:Ce, stirring thoroughly and defoaming to obtain a slurry; introducing the slurry into a mold, and curing by UV irradiation or three-dimensional (3D) printing to obtain a body; putting the body into a high-temperature furnace for heating to obtain a phosphor-embedded porous silica glass; putting the porous silica glass into a high-temperature vacuum furnace for densification and sintering to obtain a densified fluorescent glass ceramic; and finally cutting and surface-polishing.

Preparation method and use of yellow fluorescent glass ceramic

3D printer

A differential chaos shift keying (DCSK) communication method based on three-dimensional constellation is provided. Target information bit is mapped to a three-dimensional constellation symbol. An initial chaotic signal generated by a chaotic generator is processed by cyclic shift to generate a chaotic signal group including the initial chaotic signal, a first shifted chaotic signal and a second shifted chaotic signal. Information modulation is performed based on the three-dimensional constellation symbol and the chaotic signal group to obtain an information-bearing signal. A transmission signal is generated based on the information-bearing signal and the initial chaotic signal, and sent to a receiver to generate a received signal, which is subjected to information recovery analysis in the receiver to obtain an estimated information bit. A DCSK communication device based on three-dimensional constellation is also provided.

Differential chaos shift keying (DCSK) communication method and device based on three-dimensional constellation

An air conditioning unit includes: a first refrigeration system including a first evaporator, a first compressor, a first condenser, a first one-way valve, and a first throttling element connected in sequence in a loop as well as a second one-way valve connected in parallel with the first compressor, and a first fluorine pump connected in parallel with the first one-way valve; and a second refrigeration system including a second evaporator, a second compressor, a second condenser, a third one-way valve, and a second throttling element connected in sequence in a loop as well as a fourth one-way valve connected in parallel with the second compressor, and a second fluorine pump connected in parallel with the third one-way valve, where the first evaporator and the second evaporator are arranged front and rear in sequence along a return air cooling duct.

Air conditioning unit for an accurate control of supply air temperature, and operation control method and operation control device for air conditioning unit

The present invention relates to a method for carbon dioxide capture and concentration by partitioned multistage circulation based on mass transfer-reaction regulation. In the present invention, multiple means such as multistage circulating absorption, intelligent multi-factor regulation, pre-washing and cooling, inter-stage cooling, post-stage washing, slurry cleaning, cooling water waste heat utilization, small-particle-size and high-density spraying, external strengthening field such as a thermal field/ultrasonic field/electric field, and catalysis by composite catalyst are adopted, so that the target for low cost, low energy consumption, stability and high efficiency is realized. The secondary pollutants are effectively inhibited while carbon dioxide is efficiently captured; meanwhile, high-efficiency capture, low-energy desorption, and high-purity concentration of carbon dioxide are implemented. From top to bottom in sequence, the multistage circulation is used to remove aerosols, improves carbon capture efficiency, maintains absorption rate, concentrates solution, which reduces the carbon emission reduction cost.

Method for carbon dioxide capture and concentration by partitioned multistage circulation based on mass transfer-reaction regulation

Oil spout lid

In an embodiment, a terminal device includes a display unit having a display screen, and a light sensing unit located below the display unit and configured to sense incident light transmitted through the display screen. The display unit includes a transparent electrode, an opaque electrode, and an organic layer sandwiched between the two electrodes. The organic layer spontaneously emits light when a voltage difference is applied between the two electrodes. The second electrode has a semi-transparent area disposed corresponding to the light sensing unit and a remaining area which is an area of the second electrode except for the semi-transparent area. Also, an image capturing method is provided.

Terminal device and image capturing method

A terminal device includes a display unit, and an outer case, a first light sensing unit, and a second light sensing unit. The display unit has a display screen. The display unit displays images on the display screen. The outer case receives the display unit. The first light sensing unit is located below the display unit and configured to sense incident light transmitted through the display screen. The second light sensing unit is placed inside the outer case when deactivated, and extends outside the outer case when activated. Also, a dual front-facing camera operating method is provided.

Terminal device and dual front-facing camera operating method

A method for audio signal noise cancellation is provided. In response to current noise cancellation coefficients being required to be updated to new noise cancellation coefficients, the digital signal processor calculates the new noise cancellation coefficients and writes the new noise cancellation coefficients into an idle storage module in the at least two storage modules, and the digital signal processor sends an update request for updating the noise cancellation coefficients to the active noise cancellation module. The update request carries position information configured to indicate a position of the storage module to which the new noise cancellation coefficients is written. The active noise cancellation module reads the new noise cancellation coefficients in the storage module indicated by the position information based on the position information carried in the update request, and performs noise cancellation processing according to the new noise cancellation coefficients after a current noise cancellation processing cycle ends.

Method for audio signal noise cancellation, apparatus for audio signal processing, and electronic device

Growing community of inventors

Shenzhen, China

Chang Liu