Boe Technology Group Co., Ltd.

Hefei Xinsheng Optoelectronics Technology Co., Ltd.

International Business Machines Corporation

Fujitsu Corporation

Beijing Xiaomi Mobile Software Co., Ltd.

Rui Yin

David Louis Kaminsky

Xi Ning Wang

Zhiming Tan

Ying Liu

Xiaoye Ma

Donghui Zhang

Xiuli Si

Shuangtao Lv

Jian Li

Jian Ma

Wei Wu

Zhou Rui

Yunlong Zhang

Haibin Yin

Qun Fang

Anxin Dong

Yun Li Li

Sang Man Yuk

Xing Su

Ya Yu

Guoqiang Zhong

Na Dong

Xiantan Zhu

Wenlong Mao

Television

Embodiments of this disclosure provide an action recognition method and apparatus and an electronic equipment. The action recognition apparatus includes: a key-point feature extracting unit configured to process information on key points in a target human body in an image frame, to calculate key-point features of the target human body, the key-point features including a position of a predetermined key point and an angle of a predetermined joint in the target human body; and a first recognizing unit configured to recognize an action of the target human body according to the key-point features, and output a first recognition result.

Action recognition method and apparatus and electronic equipment

Embodiments of this disclosure provide a deep learning model used for image recognition and apparatus and method thereof. The model includes a determination layer configured to determine whether features in feature maps are features of positions where objects of attention are located, and different weights are granted for the positions where the objects of attention are located and other features in performing weight and composition processing on the features. Hence, the model may be guided to be focused on attention features and make correct determination, thereby improving performance and precision of the model.

Deep learning model used for image recognition and training apparatus of the model and method thereof

Embodiments of this disclosure provide a deep learning model used for driving behavior recognition and training apparatus and method thereof. In recognition, the model performs feature extraction by using a plurality of consecutive input images captured from a driving angle of view of a vehicle, and performs temporal and spatial fusion on extracted features by using a recursive neural network. In training, as images captured from a driving angle of view of a vehicle are acquired, the model may be trained. Hence, the model may accurately recognize various classes of driving behaviors.

Deep learning model used for driving behavior recognition and training apparatus and method thereof

A color filter substrate, a method of manufacturing the color filter substrate and a display device. The color filter substrate includes a base substrate, and a plurality of color resists on the base substrate, the plurality of color resists having at least two different heights. Roughness of top surfaces of the plurality of color resists is inversely proportional to the heights of the plurality of color resists.

Color filter substrate, method of manufacturing the same and display device

Rendering a web element in a web page in an integrated development environment is provided. A reference address is extracted from code of a web page. The reference address references a web element. The web element is obtained based on the reference address and in response to the reference address indicating that the web element is an embedded web element. The web element is added to a resource pool. The web element of the resource pool is rendered in the web page.

Rendering a web element in a web page

The present disclosure provides a touch control component, a method for manufacturing the same, a touch display device, and a method for preventing mistaken touch caused by liquid. The touch control component includes a touch electrode layer configured to sense a touch area within a touch surface, and a light sensing film superposed with the touch electrode layer. A light receiving surface of the light sensing film faces the touch surface. The touch control component has a simple structure and is easy to implement. By sensing the intensity of the reflected light from the touch surface using the light sensing film, the wetted area of the water layer can be identified, thereby effectively avoiding mistaken touch caused by liquid.

Touch control component, manufacturing method thereof, touch display device and method for preventing mistaken touch caused by liquid

A pixel structure and a driving method thereof, a display panel and a display device are provided. The pixel structure includes a pixel electrode, a gate line, a data line, a first thin film transistor and a second thin film transistor. A gate electrode of the first thin film transistor is electrically connected with the gate line, a first electrode of the first thin film transistor is electrically connected with the data line; a gate electrode of the second thin film transistor is electrically connected with a first electrode of the second thin film transistor, the first electrode of the second thin film transistor is electrically connected with the pixel electrode, and a second electrode of the second thin film transistor is electrically connected with a second electrode of the first thin film transistor.

Pixel structure, drive method thereof, display panel, and display device

A method normalizes a page flow into a portlet. One or more processors generates a portlet based on webpages in the page flow. The webpages are linked together through a navigation control in the portlet to execute a specific business logic. The webpages in the page flow include a first webpage and a second webpage, which are linked via a Uniform Resource Locator (URL). The processor(s) build an intra-portlet action based on a linking relationship between the first and second webpages by the portlet. The intra-portlet action is based on the URL that links the first webpage to the second webpage. The URL is located in a JavaScript file for the portlet and indicates a relative path for the portlet. The processor(s) convert the URL indicating the relative path for the portlet into a URL indicating an absolute path, which includes an attached root path for the portlet.

Normalizing a page flow

The present disclosure relates to a method for manufacturing a color filter substrate, a color substrate and a display panel in the field of display technology. The method includes: forming a color resistance layer on a basal substrate; forming a conversion layer on the basal substrate with the color resistance layer, wherein the thickness of the conversion layer is greater than the thickness of the color resistance layer and the conversion layer is convertible in color under the irradiation of preset light rays; and irradiating the conversion layer with the preset light rays so that the conversion layer is converted to a two-layer structure consisting of a black matrix (BM) pattern and a transparent over cover, wherein the BM pattern is disposed at the side of the conversion layer close to the basal substrate. In the present disclosure, it is not necessary to form the BM pattern and the over cover layer in separate two steps, because the conversion layer is formed on the basal substrate, and then is directly converted to the two-layer structure consisting of the BM pattern and the over cover layer by the preset light rays, thereby solving the problem that the method for manufacturing a color filter substrate is relatively complicated in the related art, and achieving the effect of simplifying the method for manufacturing the color filter substrate.

Method for manufacturing color filter substrate, color filter substrate and display panel

A method and/or an apparatus normalizes a page flow into a portlet. The method comprises: generating at least one portlet based on one or more pages in the page flow; building at least one of an intra-portlet action and an inter-portlet action for the least one portlet; and modifying a Uniform Resource Locator URL in the at least one portlet. With the method and the apparatus, the page flow developed by a developer can be automatically normalized into the portlet which is easier to be shared between difference applications and integrated with an existing portal application, so that the developer can focus on development of page flow while achieving user experience and advantages specific to the portlet.

Growing community of inventors

Beijing, China

Rui Yin