Shenzhen Mindray Bio-medical Electronics Co., Ltd.

Canaan Bright Sight Co., Ltd.

Boe Technology Group Co., Ltd.

Chengdu Boe Optoelectronics Technology Co., Ltd.

Shenzhen China Star Optoelectronics Semiconductor Display Technology Co., Ltd.

Xiaomi Inc.

Shenzhen Planck Innovation Technology Co., Ltd.

The Procter Gamble Company

Corestone Biosciences (beijing) Co., Ltd.

Research Institute of Petroleum Processing, Sinopec

China Petroleum & Chemical Corporation

Bing Xu

Nangeng Zhang

Xiaowei Sun

Jianhui Shao

Hongcheng Yang

Lixuan Chen

Li Zhang

Ying Han

Kai Wang

Ting Lei

Dongze Li

Xinghua Li

Miao Zhou

Zhijie Li

Junping Bao

Baohua Zhang

Yumei Zhao

Yuji Kuang

Yang Zhao

Pai Liu

Pengfei Zhang

Ming Tang

Xiang Li

Xiaojun Peng

Chunyuan Qiu

Jiangli Fan

Qing Chen

Tao Huang

Shiguo Sun

Xiaojin Tang

Qinghua Duan

Kai Xiao

Fan Fang

Tong Wu

Zuoxin Huang

Dongwang Jia

Lichuan Xiao

Zuoliang Wen

Tao Hou

Liyuan Niu

Shuo Su

Xin Cheng

Bingshuai Wang

Yizun Liu

Weiping Yang

Fei Cao

Jiayao Yao

A dilated convolution acceleration calculation method and apparatus. The method comprises: decomposing a dilated convolution computation of R×S into S sub-dilated convolution computations of R×1 (); for each sub-dilated convolution computation, caching a plurality of weight values in parallel to a plurality of calculation units in a calculation unit array (); determining, from input image data, a plurality of input data streams respectively corresponding to the plurality of weight values, and inputting the plurality of input data streams in parallel into the plurality of calculation units (); within the plurality of calculation units, executing a sliding window operation and a multiplication operation on the basis of the cached weight values and the input data streams, and executing an accumulation operation between the plurality of calculation units, so as to output an intermediate result of the sub-dilated convolution computation (); and superimposing intermediate results of the S sub-dilated convolution computations of R×1, so as to obtain a convolution result of the dilated convolution computation (). By using the method, a dilated convolution operation is accelerated with a relatively low complexity, and the function of Im2col does not need to be separately realized, thereby reducing the complexity.

Dilated convolution acceleration calculation method and apparatus

The present application discloses a manufacturing method of an optical film and the optical film. The manufacturing method includes: step S, mixing titanium source precursors and a barium source and adding an alkaline agent for a reaction to obtain nanoparticles; and step S, mixing quantum dots, an organic adhesive, and the nanoparticles followed by coating to obtain the optical film.

Manufacturing method of optical film and optical film

A method and apparatus for accelerating a convolutional neural network. The method comprises: splitting, according to rows, a weight matrix of a convolutional layer into a plurality of weight segments, and respectively caching the plurality of weight segments to a plurality of calculation units in a calculation unit array (step); reading a plurality of input data streams respectively corresponding to the plurality of weight segments, and inputting the plurality of input data streams in parallel into the plurality of calculation units (step), wherein the input data streams are formed by means of splicing a plurality of rows of data in an input feature map of the convolutional layer; and within each calculation unit, performing a sliding window operation and a multiply-accumulate computation on the input data streams on the basis of the cached weight segments, so as to obtain an output feature map of the convolutional layer (step). By means of the method, data that enters each row of calculation units is continuously read according to the whole row of data, and does not need to be read in a cross-row or cross-column manner, such that a special design is not needed for the arrangement of a memory, convolutions of different sizes are supported, and the function of Im2col does not need to be realized, thereby reducing the complexity.

Method and apparatus for accelerating convolutional neural network

A data compression method, comprising: obtaining a plurality of values of a parameter and an occurrence probability of each of the plurality of values (S); comparing the occurrence probability with a predetermined threshold, wherein values with the occurrence probability less than the predetermined threshold are first set of values, and values with the occurrence probability greater than or equal to the predetermined threshold are second set of values (S); performing pretreatment on the first set of values (S); and encoding the second set of values and the pretreated first set of values (S). By means of the data compression method, the maximum codeword length can be effectively reduced, so as to reduce the requirements of a code table to the storage space.

Data compression method and apparatus, and computer readable storage medium

Provided are a quantum dot material, a preparation method and use thereof. The quantum material includes a quantum dot, and a first cladding layer and a second cladding clad outside of the quantum dot, wherein the first cladding layer is located between the quantum dot and the second cladding layer. The quantum dot material provided herein has good water and oxygen barrier properties and good stability.

Quantum dot material, and preparation method and use thereof

A data compression method, comprising: obtaining a plurality of values of a parameter and an occurrence probability of each of the plurality of values (S) comparing the occurrence probability with a predetermined threshold, wherein values with the occurrence probability less than the predetermined threshold are first set of values, and values with the occurrence probability greater than or equal to the predetermined threshold are second set of values (S), performing pretreatment on the first set of values (S), and encoding the second set of values and the pretreated first set of values (S). By means of the data compression method, the maximum codeword length can be effectively reduced, so as to reduce the requirements of a code table to the storage space.

Provided are a backlight module, a display device, a method for driving a display device, an electronic apparatus, and a computer-readable storage medium. The backlight module includes: a base substrate; multiple light emitting modules disposed on a side of the base substrate, where each of the multiple light emitting modules includes a first light emitting region, a second light emitting region and a third first light emitting region; and a light exiting layer disposed on a side of the light emitting modules facing away from the base substrate. Vertical projections of the first light emitting region, the second light emitting region and the third light emitting region on the base substrate overlap with vertical projections of a first light exiting region, a second light exiting region and a third light exiting region on the base substrate respectively. Each backlight driving module is connected to at least one light emitting module.

Backlight module, display device, method for driving display device, electronic apparatus, and computer-readable storage medium

Barium titanium particles, a method of preparing the same, and a display panel are disclosed. The method includes steps of mixing a barium source precursor and a titanium source precursor with an alkaline solution to obtain a to-be-reacted mixture solution, wherein the barium source precursor is selected from barium enoate; and reacting the to-be-reacted mixture solution under a first condition for a first time, and then isolating and purifying to obtain the barium titanium particles.

Barium titanium particles, method of preparing the same, and display panel

A backlight module, display apparatus and method for manufacturing the display apparatus are provided. The backlight module includes: a positioning frame and a buffer adhesive tape. A display panel is arranged on a bearing platform of the positioning frame, and distance from a platform surface in an inner region of the bearing platform to the display panel is greater than distance from a platform surface in an edge region of the bearing platform to the display panel. The buffer adhesive tape is arranged on the bearing platform, a side of the buffer adhesive tape in the inner region is in contact with the display panel and the other side thereof is attached to the platform surface, a back adhesive is arranged on the side of the buffer adhesive tape attached to the platform surface of the bearing platform. The backlight module improves the display effect of the black-state pictures.

Backlight module, display apparatus and method for manufacturing display apparatus

A printed circuit board, a method for determining an engagement state between the printed circuit board and a flexible printed circuit, and a display device are provided. The printed circuit board includes a substrate, a connector fixed on the substrate, and configured to be connected with the flexible printed circuit, and a determining circuit connected with the connector. The secondary pin is added to the connector, and can determine together with the determining circuit, added to the substrate, connected with the secondary pin, an engagement state between the first pins of the connector, and the second pins of the flexible printed circuit.

Printed circuit board, method for determining engagement state between printed circuit board and flexible printed circuit and display device

The present disclosure provides a display device which includes a liquid crystal display panel and a backlight assembly. The backlight assembly includes a light diffusion component and a light emitting component optically coupled to the light diffusion component. The display device further includes a light guide component configured to direct ambient light to the light diffusion component.

Display device

An endotracheal tube (ETT) system and methods are provided. In at least one example, an ETT system may comprise, a tube, inflatable cuff coupled to the tube, and a restrictor. In at least one example the restrictor may comprise one more cavities. In a further example, a restrictor drainage assembly may be coupled to the restrictor, the restrictor drainage assembly configured to drain secretions that may be collected at a collection point formed by the restrictor when the ETT system is positioned in an airway of a patient and in an inflated state.

Endotracheal tube drainage system and methods

The present disclosure provides a method for controlling a terminal device by using a headset wire connected to the terminal device. The method includes: recognizing a user's gesture based on a current detected in a specific region of the headset wire; acquiring control instruction corresponding to the user's gesture; and executing the control instruction. According to the present disclosure, when a current is detected in the specific region of the headset wire, a user's gesture is recognized based on the current, control instruction corresponding to the user's gesture is acquired, and the control instruction is executed to control the terminal. In the whole procedure, the user only needs to touch the specific region of the headset wire without using pressure button to produce current signal, thereby the operation of controlling the terminal device is convenient and fast.

Method for controlling terminal device by using headset wire and the terminal device thereof

The present invention relates to a stable liquid laundry detergent composition containing low levels of a C-Clinear alkyl benzene sulphonate surfactant in combination with high levels of a water-soluble alkali metal carbonate for achieving improved suds profile.

Low surfactant, high carbonate liquid laundry detergent compositions with improved suds profile

A battery for electronic equipment includes: a rechargeable power supply; and a battery chip, wherein: the rechargeable power supply is configured to supply power to the electronic equipment; and the battery chip is configured to detect whether the rechargeable power supply has started to supply power to the electronic equipment and, if a detection result is that the rechargeable power supply has started to supply power to the electronic equipment, transmit a customized signal to the electronic equipment through a predetermined transmitting pin.

Battery, and method, device, and system for battery protection

Growing community of inventors

Beijing, China

Bing Xu