Abstract: A photoelectric sensor and a substrate are disclosed. The photoelectric sensor includes a photoelectric conversion layer, a first electrode and a second electrode, wherein the first electrode is arranged on a side of the photoelectric conversion layer, and the second electrode is arranged on a side of the photoelectric conversion layer and is spaced apart from the first electrode; wherein the first electrode and the second electrode are configured to drive the photoelectric conversion layer; and in a direction perpendicular to a surface of the photoelectric conversion layer, the first electrode and the second electrode are overlapped with the photoelectric conversion layer respectively, and the photoelectric conversion layer includes an oxide semiconductor material.

Abstract: Provided is a shift register unit. The shift register unit includes an input circuit, a compensation control circuit, and an output circuit; wherein input circuit coupled to an input signal terminal, an input control terminal, a first power supply terminal, a reference node, and a first node; the compensation control circuit coupled to a first clock signal terminal, the reference node, and the first node; and the output circuit coupled to the first node, a second clock signal terminal, and an output terminal.

Abstract: A thin film transistor includes a first active layer, a second active layer, a first electrode, a second electrode and a third electrode. The first active layer includes a first surface away from a substrate. The second active layer includes a second surface in contact with the first surface. The first electrode, the first active layer and the second active layer have an overlapping region. The second electrode, the first active layer and the second active layer have an overlapping region. The third electrode, the first active layer and the second active layer have an overlapping region, and the third electrode is opposite to the second electrode. The second surface is located within the first surface, and a distance between at least part of a border of the second surface and a border of the first surface is less than or equal to 0.5 ?m.

Abstract: A thin film transistor including: a base substrate, and an active layer and a gate on the base substrate, where the active layer includes a first part and a second part, a conductivity of the second part is greater than a conductivity of the first part; an orthographic projection of the gate on the base substrate covers an orthographic projection of the first part on the base substrate, and the orthographic projection of the gate on the base substrate does not overlap an orthographic projection of the second part on the base substrate; and the first part includes a plurality of first sub-parts, and two sides of any one first sub-part in a trend direction of the active layer are each connected to the second part.

Abstract: The present disclosure provides a shift register unit, and driving method therefor, a gate drive circuit, and a display device, belonging to the field of display technologies. The shift register unit includes an input circuit, a compensation control circuit, and an output circuit. The input circuit can control a potential of a first node under control of an input signal provided by an input signal terminal and control a potential of a reference node under control of the input signal and an input control signal provided by an input control terminal. The compensation control circuit can adjust the potential of the first node based on the potential of the reference node under control of a first clock signal provided by a first clock signal terminal. In this way, the flexibility of controlling the first node is improved. Thus, the output circuit can flexibly output a drive signal to an output terminal coupled to a gate line under control of the first node.

Abstract: Methods, systems, and computer storage media for providing an automotive data analytics service associated with an automotive data analytics engine in an automotive data analytics system. The automotive data analytics service is provided via an automotive data analytics engine (i.e., an automotive market intelligence engine) that includes a conversion rate engine to generate conversion rates for different car dealerships. A conversion rate is based on a number of visitors to a car dealership location who are inferred to have bought cars from the car dealership location. The automotive data analytics engine supports: aggregating different types of automotive data analytics data; using an automotive data computation engine—having an automotive data analytics model—to process the data; calculating a conversion rate for different types of car dealerships; and generating automotive data analytics results data—including the conversion rate—that is provided for display on a graphical user interface.

Abstract: A method for manufacturing a thin film transistor, and a thin film transistor are provided. The method includes: forming an active layer on a substrate by a single patterning process; forming a gate insulating layer by deposition on a side of the active layer away from the substrate; forming a first via hole and a second via hole penetrating through the gate insulating layer by a single patterning process, the first and second via holes being located at two ends of the active layer respectively; and forming a first electrode, a gate electrode, and a second electrode on the gate insulating layer by a single patterning process, the first and second electrodes being connected to the active layer through the first and second via holes, respectively, and an orthographic projection of the gate electrode on the substrate at least partially overlapping that of the active layer on the substrate.

Abstract: The present disclosure describes a computer-implemented method for image landmark detection. The method includes receiving an input image for the image landmark detection, generating a feature map for the input image via a convolutional neural network, initializing an initial graph based on the generated feature map, the initial graph representing initial landmarks of the input image, performing a global graph convolution of the initial graph to generate a global graph, where landmarks in the global graph move closer to target locations associated with the input image, and iteratively performing a local graph convolution of the global graph to generate a series of local graphs, where landmarks in the series of local graphs iteratively move further towards the target locations associated with the input image.

Abstract: The present disclosure describes a computer-implemented method for image landmark detection. The method includes receiving an input image for the image landmark detection, generating a feature map for the input image via a convolutional neural network, initializing an initial graph based on the generated feature map, the initial graph representing initial landmarks of the input image, performing a global graph convolution of the initial graph to generate a global graph, where landmarks in the global graph move closer to target locations associated with the input image, and iteratively performing a local graph convolution of the global graph to generate a series of local graphs, where landmarks in the series of local graphs iteratively move further towards the target locations associated with the input image.

Abstract: An image segmentation method includes generating a CTN (contour transformer network) model for image segmentation, where generating the CTN model includes providing an annotated image, the annotated image including an annotated contour, providing a plurality of unannotated images, pairing the annotated image to each of the plurality of unannotated images to obtain a plurality of image pairs, feeding the plurality of image pairs to an image encoder to obtain a plurality of first-processed image pairs, and feeding the plurality of first-processed image pairs to a contour tuner to obtain a plurality of second-processed image pairs.

Abstract: An image segmentation method includes generating a CTN (contour transformer network) model for image segmentation, where generating the CTN model includes providing an annotated image, the annotated image including an annotated contour, providing a plurality of unannotated images, pairing the annotated image to each of the plurality of unannotated images to obtain a plurality of image pairs, feeding the plurality of image pairs to an image encoder to obtain a plurality of first-processed image pairs, and feeding the plurality of first-processed image pairs to a contour tuner to obtain a plurality of second-processed image pairs.