Abstract: An automatic vehicle window cleaning method and apparatus includes obtaining a vehicle window image; determining a dark channel image corresponding to a single frame of vehicle window image; further determining, based on a quantity of pixels whose gray values exceed a preset gray threshold in the dark channel image and/or definition of the dark channel image, that a first-type foreign object exists on a vehicle window, and/or determining, based on RGB values of pixels in i consecutive frames of vehicle window images, that a second-type foreign object exists on the vehicle window, and controlling a cleaning tool to remove a foreign object. The foreign object includes the first-type foreign object and/or the second-type foreign object.

Abstract: Embodiments of the disclosure relate to unsupervised learning with synthetic data and attention masks. According to example embodiments of the disclosure, a plurality of sample images are generated by providing a plurality of text prompts into a trained generative model, respectively. For a sample image of the plurality of sample images, at least one attention map is obtained from a generative model, the at least one attention map being determined by the generative model for generating the sample image, an attention map indicating visual elements of an object within the sample image. Training of a target model is performed according to unsupervised learning at least based on the plurality of sample images and attention maps for the plurality of sample images, the target model being configured to perform an image processing task.

Abstract: A method for presenting a face in a video call includes: obtaining a key feature point of a facial expression of a user based on a face image of the user; driving a three-dimensional (3D) head image of the user by using the key feature point of the facial expression of the user, to obtain a target 3D avatar of the user, where the target 3D avatar of the user has an expression of the user; rotating the target 3D avatar based on a preset presentation angle, to obtain a target 3D avatar at the preset presentation angle; and transmitting the target 3D avatar at the preset presentation angle to a peer video call device. A user can see a 3D avatar that is of a user and that is at a preset presentation angle in real time.

Abstract: Embodiments of this application provide a data collection method and system, and an apparatus, to improve effect of collecting information about an environment around a vehicle. The system includes: an image collection apparatus, including a camera and a drive component, where the drive component is configured to drive adjustment of a photographing angle of the camera, so that in an image collected by the camera, an included angle between a lower side line of a body side of the vehicle and a vertical direction is less than a first angle; and a control system, electrically connected to the image collection apparatus, and configured to send a first control instruction to the image collection apparatus, where the first control instruction is used to control the photographing angle of the camera.

Abstract: An adaptive rearview mirror adjustment method includes obtaining a spatial position of a human eye of a driver of a current vehicle and a spatial position of a target rearview mirror, obtaining a horizontal field of view of the driver in the target rearview mirror based on the spatial position of the human eye of the driver and a spatial position of a target rearview mirror, obtaining a next-vehicle image, obtaining a first auxiliary angle of a target next vehicle based on the next-vehicle image, calculating a second auxiliary angle based on the horizontal field of view, the spatial position of the human eye, and the position of the target rearview mirror, obtaining a horizontal adjustment angle of the target rearview mirror based on the first auxiliary angle and the second auxiliary angle, and adjusting a horizontal angle of the target rearview mirror based on the horizontal adjustment angle.

Abstract: A control method and apparatus for displaying multimedia content, an electronic device, and a medium. The method includes: displaying a first-type multimedia content on a first content display layer of a first-type multimedia content display interface, the first-type multimedia content display interface including: a first user interaction layer and the first content display layer, the first user interaction layer being superimposed and displayed on the first content display layer; and receiving a first swiping operation inputted by a user on the user interaction layer, and exiting the first-type multimedia content display interface. That is, a swiping operation triggers to exit the first-type multimedia content display interface, as the swiping operation is significantly different from a click operation, the user is provided with a multimedia content display control method that is more consistent with the user's operation habit, improving the user experience.

Abstract: Embodiments of the disclosure provide for light control and systems. Some embodiments include obtaining driving information, where the driving information includes at least one of the following: navigation information, driving assistance information, and vehicle data. Some embodiments further include obtaining information about a to-be-driven path. Some embodiments further include obtaining a target light pattern based on the driving information and the information about the to-be-driven path. Some embodiments further include displaying, on the to-be-driven path and using the target light pattern, a beam emitted by a vehicle.

Abstract: Embodiments of the present disclosure provide a solution for multimodal data processing. A method comprises: obtaining image data and text data; and extracting a target visual feature of image data and a target textual feature of text data using a feature extraction model. The feature extraction model comprises alternatively deployed cross-modal encoding parts and visual encoding parts. The extracting comprises: performing, using a first cross-modal encoding part of the feature extraction model, cross-modal feature encoding on a first intermediate visual feature of the image data and a first intermediate textual feature of the text data, to obtain a second intermediate visual feature and a second intermediate textual feature; performing, using a first visual encoding part of the feature extraction model, visual modal feature encoding on the second intermediate visual feature, to obtain a third intermediate visual feature.

Abstract: A method, apparatus, device, and medium for image processing is provided. The method includes generating, using an image generation process, a first set of synthetic images based on a first set of codes associated with the first image class in a codebook and based on a first class feature associated with a first image class; generating, using a feature extraction process, a first set of reference features based on the first set of synthetic images and generating a first set of target features based on a plurality of sets of training images belonging to the first image class in a training image set; and updating the image generation process and the codebook according to at least a first training objective to reduce a difference between each reference feature in the first set of reference features and a corresponding target feature in the first set of target features.

Abstract: A control method and apparatus for displaying multimedia content, an electronic device, and a medium. The method includes: displaying a first-type multimedia content on a first content display layer of a first-type multimedia content display interface, the first-type multimedia content display interface including: a first user interaction layer and the first content display layer, the first user interaction layer being superimposed and displayed on the first content display layer; and receiving a first swiping operation inputted by a user on the user interaction layer, and exiting the first-type multimedia content display interface. That is, a swiping operation triggers to exit the first-type multimedia content display interface, as the swiping operation is significantly different from a click operation, the user is provided with a multimedia content display control method that is more consistent with the user's operation habit, improving the user experience.

Abstract: A high-performance positive electrode catalyst for a metal-air battery is disclosed, which is composed of transition metal nitride-transition metal oxide heterogeneous particles and a mesoporous carbon matrix. The nano heterogeneous particles, which are 10-50% based on the total mass of the catalyst, are dispersed in the mesoporous carbon matrix; and the oxide is 10-100% based on the heterogeneous particles. A preparation method of the catalyst includes: treating mesoporous carbon with a strong acid solution to obtain surface-functionalized mesoporous carbon; immersing the surface-functionalized mesoporous carbon in an aqueous solution of a transition metal salt, and stirring and washing; adding ammonia water and stirring to enable a confined complexation reaction; washing again, and vacuum drying; and calcining the product in an inert atmosphere or a vacuum condition.

Abstract: Disclosed are a method and device for timeout monitoring, comprising: determining key information of a request message, the key information including a sending time of the request message; storing the key information to a first level cache; scanning the first level cache according to a set frequency, if the first level cache includes a first request message, storing the key information of the first request message into a second level cache, the first request message being a request message for which a reply message is not received; scanning the second level cache, and determining by a message log whether a reply message is received for a second request message in the second level cache, if not, the second request message times out, the second request message is a request message in which the difference value between a sending time and a current time is greater than a time-out threshold.

Abstract: The present disclosure relates to methods and apparatuses for identifying an object of interest of a user. One example method includes obtaining information about a line-of-sight-gazed region of the user and an environment image corresponding to the user, obtaining information about a first gaze region of the user in the environment image based on the environment image, where the first gaze region is used to indicate a sensitive region determined by using a physical feature of a human body, and obtaining a target gaze region of the user based on the information about the line-of-sight-gazed region and the information about the first gaze region. The gaze region is used to indicate a region in which a target object gazed by the user in the environment image is located.

Abstract: This application discloses image processing methods and systems. One method includes: obtaining a first image, obtaining a template image having a life value that determines whether the template image is valid, comparing the first image with the template image, and storing the first image in a template library as a new template image in response to determining that the first image matches the template image.

Abstract: Example vehicle unlocking methods and apparatus are described. In one example method, wakeup keyword detection is performed on an obtained voice signal. If the voice signal includes a wakeup keyword, an acoustic source direction is determined, and a target camera is indicated to collect image data. The target camera is a target camera that is installed on the vehicle and whose angle of view can ensure that the acoustic source direction is captured. Then, the vehicle is unlocked based on the image data collected by the target camera.

Abstract: A high-performance positive electrode catalyst for a metal-air battery is disclosed, which is composed of transition metal nitride-transition metal oxide heterogeneous particles and a mesoporous carbon matrix. The nano heterogeneous particles, which are 10-50% based on the total mass of the catalyst, are dispersed in the mesoporous carbon matrix; and the oxide is 10-100% based on the heterogeneous particles. A preparation method of the catalyst includes: treating mesoporous carbon with a strong acid solution to obtain surface-functionalized mesoporous carbon; immersing the surface-functionalized mesoporous carbon in an aqueous solution of a transition metal salt, and stirring and washing; adding ammonia water and stirring to enable a confined complexation reaction; washing again, and vacuum drying; and calcining the product in an inert atmosphere or a vacuum condition.

Abstract: The present application discloses a method for presenting a face in a video call. The method includes: obtaining a key feature point of a facial expression of a user based on a face image of the user; driving a 3D head image of the user by using the key feature point of the facial expression of the user, to obtain a target 3D avatar of the user, where the target 3D avatar of the user has an expression of the user; rotating the target 3D avatar based on a preset presentation angle, to obtain a target 3D avatar at the preset presentation angle; and transmitting the target 3D avatar at the preset presentation angle to a peer video call device. An user can see a 3D avatar that is of a user and that is at a preset presentation angle in real time, to improve the video conference experience.

Abstract: This application provides an automatic vehicle window cleaning method and apparatus, used to detect a foreign object on a vehicle window based on a value of a pixel in a vehicle window image, and perform automatic cleaning. The method includes: obtaining a vehicle window image; determining a dark channel image corresponding to a single frame of vehicle window image; further determining, based on a quantity of pixels whose gray values exceed a preset gray threshold in the dark channel image and/or definition of the dark channel image, that a first-type foreign object exists on a vehicle window; and/or determining, based on RGB values of pixels in i consecutive frames of vehicle window images, that a second-type foreign object exists on the vehicle window; and controlling a cleaning tool to remove a foreign object. The foreign object includes the first-type foreign object and/or the second-type foreign object.

Abstract: An adaptive rearview mirror adjustment method includes obtaining a spatial position of a human eye of a driver of a current vehicle and a spatial position of a target rearview mirror, obtaining a horizontal field of view of the driver in the target rearview mirror based on the spatial position of the human eye of the driver and a spatial position of a target rearview mirror, obtaining a next-vehicle image, obtaining a first auxiliary angle of a target next vehicle based on the next-vehicle image, calculating a second auxiliary angle based on the horizontal field of view, the spatial position of the human eye, and the position of the target rearview mirror, obtaining a horizontal adjustment angle of the target rearview mirror based on the first auxiliary angle and the second auxiliary angle, and adjusting a horizontal angle of the target rearview mirror based on the horizontal adjustment angle.

Abstract: The present invention relates to a data processing system for providing business continuity protection, comprising: a business center comprising a first data processing unit, a first pre-communication unit, and at least a first post-communication unit; a disaster recovery center comprising a second data processing unit, a second pre-communication unit, and at least a second post-communication unit; wherein the first pre-communication unit and the second pre-communication unit are communicatively coupled to the first post-communication unit respectively when the business center operates normally; the first pre-communication unit and the second pre-communication unit are communicatively coupled to the second post-communication unit respectively when the disaster recovery center operates in place of the business center. The data processing system can reliably switch between the business center and the disaster recovery center, thereby providing a complete protection for the business continuity.