Abstract: Disclosed are a follow-up control method and follow-up control system for a forming machine. The follow-up control method comprises: responding to an instruction for a forming drum module to move from a fitting station to a locking ring station, controlling the forming drum module and the stitching and pressing roller module to move towards the ring locking station, in which the moving speed of the stitching and pressing roller module to the ring locking station has a pre-set speed difference from the moving speed of the forming drum module to the ring locking station, controlling a sewing pressure roller module to perform a sewing operation on a compound joint on a molding drum module, and when the molding drum module reaches a pre-set position before a loop locking position of the loop locking station, the sewing pressure roller module completes the sewing operation.

Abstract: Disclosed are a control method and control system for a tire forming machine, and a readable storage medium.

Abstract: The embodiments of the present disclosure disclose gaze estimation cross-scene adaptation method and device based on outlier guidance. A specific implementation of the method comprises: performing pre-training on a source domain based on a given arbitrary gaze estimation model, to obtain a collaborative learning model group; determining an average collaborative model corresponding to each collaborative learning model in the collaborative learning model group, to obtain an average collaborative model group; generating an outlier corresponding to the collaborative learning model group based on a target image, the collaborative learning model group and the average collaborative model group; using an outlier loss function and the outlier to optimize the collaborative learning model group; using any collaborative learning model in the optimized collaborative learning model group to perform gaze estimation.

Abstract: A robot and a robot control method are provided. The robot includes: a robot body, a pose detector being mounted on the robot body; a chassis body provided at a bottom of the robot body; a moving assembly connected to the robot body through a suspension shock absorber, the moving assembly being further connected to a controllable damping rotating shaft, and rotationally connected to the chassis body through the controllable damping rotating shaft; and a controller electrically connected to the pose detector and the controllable damping rotating shaft, respectively, and used to control the controllable damping rotating shaft to change rotational resistance according to pose information of the robot detected by the pose detector.

Abstract: Disclosed is a robot assembly, including: a robot body and a charging pile, wherein a chassis of the robot body is provided with a positioning groove and a guide groove, a first positive electrode contact and a first negative electrode contact are arranged in the positioning groove, the charging pile is provided with a contact seat selectively matching the positioning groove, and the contact seat is provided with a second positive electrode contact and a second negative electrode contact; and the guide groove has a larger cross-sectional area than the positioning groove, the contact seat enters the positioning groove under the guide action of the guide groove, so that the first positive electrode contact makes contact with the second positive electrode contact, and the first negative electrode contact makes contact with the second negative electrode contact.

Abstract: The embodiments of the present disclosure disclose a method for estimating the gaze directions of multiple persons in images, including: a novel method for estimating the gaze directions of multiple persons in images, being able to predict the gaze direction of a single or multiple face regions in an image accurately and in real-time; a novel multi-task learning network structure, which simultaneously predict the gaze directions of multiple face regions in the image through one-time calculation; a self-supervised loss function based on two-dimensional projection, which be used to supervise three-dimensional gaze direction estimation; a novel method for generating gaze direction replacement data of multiple face regions, being able to quickly generate a large amount of labeled realistic data, for training and testing of a deep learning model; the proposed network is trained end-to-end on the constructed data, then the model can predict the gaze direction in real-time during deployment testing.

Abstract: The embodiments of this disclosure disclose an unsupervised image-to-image translation method. A specific implementation of this method comprises: obtaining an initial image, and zooming the initial image to a specific size; performing spatial feature extraction on the initial image to obtain feature information; inputting the feature information to a style-content separation module to obtain content feature information and style feature information; generating reference style feature information of a reference image in response to obtaining the reference image, and setting the reference style feature information as a Gaussian noise in response to not obtaining the reference image; inputting the content feature information and the reference style feature information into a generator to obtain a target image; and zooming the target image to obtain a final target image. This implementation can be applied to a variety of different high-level visual tasks, and improve the expandability of the whole system.

Abstract: This application discloses a network deployment configuration method and apparatus, and belongs to the field of network technologies. The method includes: obtaining user intent, and obtaining first network deployment configuration information of a user network based on the user intent. The user intent is for deploying the user network, the user intent includes scenario information of the user network and service information of the user network, the scenario information indicates a usage scenario corresponding to the user network, the service information indicates at least one network service deployed in the user network, and the first network deployment configuration information includes device information of at least one network device for planning the user network, a network topology of the network device, and a configuration parameter of the network device.

Abstract: The embodiments of the present disclosure relate to the field of display technology, and in particular to a display substrate, a display panel, and a display device. The display substrate includes a first electrode, a first wire, and a second wire; the first electrode is provided with a plurality of slits; the first wire is provided on a side of the first electrode, and a first gap is provided between the first wire and the first electrode, and the first wire being electrically connected to the first electrode; a second wire provided on a side of the first electrode away from the first wire, and a second gap being provided between the second wire and the first electrode; wherein each of the slits is provided with a first end and a second end opposed to each other, and the first end is close to the first wire, and the second end is close to the second wire, and a light-transmitting part of the first end is larger than a light-transmitting part of the second end.

Abstract: Disclosed in the present disclosure is a command information transmission method, applied to a Physical Function (PF) port. The method includes: parsing a received interrupt request to obtain a target Virtual Function (VF) port; determining a target physical-side interrupt handler corresponding to the target VF port; obtaining target command information corresponding to the target physical-side interrupt handler in a device memory overlap region, where the device memory overlap region is an overlap region obtained by performing memory mapping on the PF port and each VF port in a host address space by a device memory overlap mapping technology; and executing switching to the target physical-side interrupt handler, and performing a response operation on the target command information.

Abstract: The disclosure provides a laser surface treatment method and a laser surface treatment station. The laser surface treatment method is applied to a surface to be adhesively bonded of a casting of a vehicle body-in-white part, wherein the laser surface treatment method comprises the following steps: S1: providing the casting of the vehicle body-in-white part, the casting having been demolded and machined; and S2: performing laser surface treatment on the surface to be adhesively bonded of the casting, the surface to be adhesively bonded being used for a subsequent adhesive application process for connection of the casting, wherein the laser surface treatment is carried out by pulse laser, and the laser surface treatment employs the following parameters: an average laser power greater than 500 W; a laser frequency ranging from 5 kHz to 10 kHz; and laser single-point energy greater than 50 mJ.

Abstract: The present application provides a detecting method of an operating state of a power supply and a detecting apparatus, where the power supply includes a primary circuit, a transformer and a secondary circuit. The secondary circuit includes a secondary current detecting unit, a temperature detecting unit and a secondary controlling unit. Firstly the secondary current detecting unit detects a current value of the secondary circuit, then the secondary controlling unit compares the current value of the secondary circuit with a preset current threshold. When the current value of the secondary circuit is less than or equal to the preset current threshold, the temperature detecting unit detects a temperature value of the power supply and the secondary controlling unit determines the operating state of the power supply according to the acquired temperature value of the power supply.

Abstract: Disclosed in the present disclosure is a command information transmission method, applied to a Physical Function (PF) port. The method includes: parsing a received interrupt request to obtain a target Virtual Function (VF) port; determining a target physical-side interrupt handler corresponding to the target VF port; obtaining target command information corresponding to the target physical-side interrupt handler in a device memory overlap region, where the device memory overlap region is an overlap region obtained by performing memory mapping on the PF port and each VF port in a host address space by a device memory overlap mapping technology; and executing switching to the target physical-side interrupt handler, and performing a response operation on the target command information.

Abstract: A display panel having an active area includes: an array substrate and an opposite substrate disposed opposite to the array substrate. The array substrate includes a first substrate and a conductive structure disposed on the first substrate. The conductive structure is located on a side of the active area. On a side of the conductive structure away from the active area, a side surface of the conductive structure is substantially flush with a side surface of the first substrate. On the side where the conductive structure is located, an edge of an orthographic projection of the opposite substrate on a plane where the array substrate is located is located outside an edge of the array substrate.

Abstract: A device area and a marking area neighboring the device area over a dielectric stack are determined. The dielectric stack includes insulating material layers and sacrificial material layers arranged alternatingly over a substrate. The device area and the marking area are patterned using a same etching process to form a marking pattern having a central marking structure in a marking area and a staircase pattern in the device area. The marking pattern and the staircase pattern have a same thickness equal to a thickness of at least one insulating material layer and one sacrificial material layer, and the central marking structure divides the marking area into a first marking sub-area farther from the device area and a second marking sub-area closer to the device area. A first pattern density of the first marking sub-area is greater than or equal to a second pattern density of the second marking sub-area.

Abstract: Embodiments of a marking pattern in forming the staircase structure of a three-dimensional (3D) memory device are provided. In an example, a semiconductor device includes a stack structure having insulating layers and conductor layers arranged alternatingly over a substrate along a vertical direction; and a marking pattern having interleaved layers over the substrate and neighboring the stack structure. The marking pattern includes a central marking structure located in a marking area. The central marking structure consists of interleaved layers and divides the marking area into a first marking sub-area and a second marking sub-area. A first pattern density of the first marking sub-area is higher than or equal to a second pattern density of the second marking sub-area.

Abstract: Embodiments of a marking pattern in forming the staircase structure of a three-dimensional (3D) memory device are provided. In an example, a marking pattern for controlling a trimming rate of a photoresist trimming process includes a plurality of interleaved layers, the plurality of interleaved layers including at least two layers of different materials stacking along a vertical direction over a substrate. In some embodiments, the marking pattern also includes a central marking structure that divides the marking area into a first marking sub-area farther from a device area and a second marking sub-area closer to the device area, a first pattern density of the first marking sub-area being higher than or equal to a second pattern density of the second marking sub-area.

Abstract: The embodiments of the present disclosure disclose gaze estimation cross-scene adaptation method and device based on outlier guidance. A specific implementation of the method comprises: performing pre-training on a source domain based on a given arbitrary gaze estimation model, to obtain a collaborative learning model group; determining an average collaborative model corresponding to each collaborative learning model in the collaborative learning model group, to obtain an average collaborative model group; generating an outlier corresponding to the collaborative learning model group based on a target image, the collaborative learning model group and the average collaborative model group; using an outlier loss function and the outlier to optimize the collaborative learning model group; using any collaborative learning model in the optimized collaborative learning model group to perform gaze estimation.