Abstract: A model training method includes obtaining first and second modal data sets including pieces of data, wherein the pieces of data include sub-modal data pieces, and wherein pieces of the first modal data correspond to pieces of the second modal data; obtaining masked data sets by masking a piece of the first and second sub-modal data, respectively; performing feature prediction on the first masked data set and the second modal data set and the second masked data set and the first modal data set, based on ta feature extraction model, to obtain first and second global features and first and second global recovery features; and generating a trained feature extraction model by optimizing the feature extraction model based on the first global recovery features, the first global features, the second global recovery features, and the second global features.

Abstract: An embodiment of this application discloses an artificial intelligence-based image generation method performed by a computer device. The method includes: acquiring a source image including a target object whose pose is to be transformed, and a target image including a reference object presenting a target pose; determining a pose transition matrix according to a model pose corresponding to the pose of the target object and a model pose corresponding to the target pose of the reference object; extracting a basic appearance feature of the target object from the source image; processing the basic appearance feature based on the pose transition matrix, to obtain a target appearance feature of the target object in the target pose; and generating a target synthetic image of the target object in the target pose based on the target appearance feature.

Abstract: In a facial expression recognition method, facial key points are identified as graph nodes. A facial graph structure for a first image is constructed with edges between pairs of the graph nodes based on relationships between the facial key points corresponding to the graph nodes. A first feature of a facial texture is extracted from color information of pixels in the first image. A second feature of the first image is extracted by processing the facial graph structure using a graph neural network (GNN). The first feature and the second feature are combined, to obtain a fused feature. A first expression type of a face in the first image that corresponds to the fused feature is determined. The first expression type is determined from a plurality of facial expression types.

Abstract: A data processing method and apparatus for point cloud media including obtaining a media file of point cloud media, the media file comprising a point cloud code stream of the point cloud media, an attribute parameter set, and attribute parameter set indication information, the attribute parameter set comprising parameter information required by decoding of attribute data in the point cloud code stream, and the attribute parameter set indication information indicating an encapsulation position of the attribute parameter set, and decoding the point cloud code stream according to the attribute parameter set indication information to present the point cloud media.

Abstract: This application discloses request processing methods and apparatuses. The request processing method, performed by a first computing device in a clustered database includes determining a lock synchronization mode of the clustered database in response to a data request; locking a data resource corresponding to the data request when the lock synchronization mode is a weak synchronization mode, and executing the data request; obtaining locking feedback information about the data resource by at least one second computing device in the clustered database when the lock synchronization mode is not the weak synchronization mode, the locking feedback information indicating whether the second computing device has locked the data resource; and locking the data resource corresponding to the data request when the obtained locking feedback information satisfies a target condition corresponding to the lock synchronization mode, and executing the data request.

Abstract: This application relates to a video encoding method, computer-readable storage medium, and computer device. The method includes: obtaining video data, the video data including a plurality of original video images; processing the video data according to a sequence generation rule to obtain a plurality of independent subsequence video image groups, each of the subsequence video image groups including subsequence images corresponding to the sequence generation rule; obtaining encoding manners corresponding to the subsequence video image groups; and independently encoding the corresponding subsequence video image groups according to the encoding manners to obtain subsequence video encoded data corresponding to each of the subsequence video image groups.

Abstract: An image processing method and apparatus, a computer device, a storage medium and a program product, which can be applied to scenes such as image processing, game, and game basic skill. The method includes: encoding a current frame image to obtain an initial encoded image, and obtaining an encoding parameter, the encoding parameter comprising: a parameter indicating whether a scene switch exists in the current frame image; and recoding the current frame image based on a determination, according to the encoding parameter, that the current frame image needs to be recoded to obtain a target encoded image, wherein a data volume of the target encoded image is less than the data volume of the initial encoded image.

Abstract: A method for displaying a virtual item includes: displaying a first virtual environment picture, the first virtual environment picture displaying a second virtual item in an effective state; controlling, in response to a use instruction, a target virtual character to use an equipped first virtual item; displaying, in response to the second virtual item being located within a first influence range of the first virtual item, that the second virtual item is switched from the effective state to an ineffective state; and displaying, in response to satisfying a restoration condition, that the second virtual item is switched from the ineffective state to the effective state.

Abstract: A key filling method includes transmitting first filling information to a server, the first filling information indicates to a security chip a filling permission of an electronic device; obtaining second filling information that includes a verification public key generated by the server. In accordance with a determination that the electronic device has the filling permission based on the first filling information, the second filling information comprises filling information transmitted by the server to the electronic device; and filling the verification public key into the security chip, the verification public key being used for encrypting information, and a verification private key corresponding to the verification public key being used for decrypting the encrypted information.

Abstract: A communication method includes: selecting an ith level user plane node corresponding to an ith level control plane node, i=1, . . . , N, N being a positive integer; transmitting a user plane MBS session establishment request to an ith level user plane node, and receiving a user plane MBS session establishment response fed back by the ith level user plane node; and transmitting an MBS session start request to a son control plane node of the ith level control plane node, so that the son control plane node allocates son user plane nodes to the ith level user plane node, the MBS session start request including identification information of the ith level user plane node and a first MBS Internet Protocol (IP) multicast transmission address and a first common-tunnel endpoint identifier (C-TEID) that are allocated by the ith level user plane node.

Abstract: This application provides a virtual object control method performed by a computer device. The method includes: displaying a virtual object in a virtual scene region of a first level; in response to a task execution instruction for the virtual object, controlling the virtual object to perform a task corresponding to the first level in the virtual scene region; when the virtual object completes the task, displaying level identifiers of at least two second levels adjacent to the first level; in response to an interaction instruction for a target level identifier in the level identifiers of the at least two second levels, controlling the virtual object to perform an interaction operation on the target level identifier; and when the interaction operation is completed, controlling the virtual object to move to a virtual scene region of a second level corresponding to the target level identifier.

Abstract: This application discloses a method for reconstructing a dendritic tissue in an image performed by a computer device. The method includes: acquiring original image data corresponding to a target image of a target dendritic tissue and corresponding reconstruction reference data determined based on a local reconstruction result of the target dendritic tissue in the target image; applying a target segmentation model to the original image data and the reconstruction reference data to acquire a target segmentation result for indicating a target category of each pixel in the target image, and the target category of any pixel being used for indicating whether the pixel belongs to the target dendritic tissue or not; and reconstructing the target dendritic tissue in the target image based on the target segmentation result to obtain a complete reconstruction result of the target dendritic tissue in the target image.

Abstract: An image classification method includes: performing image segmentation on an unlabeled sample image to obtain image blocks and performing feature extraction on each image block to obtain an initial image feature set including an initial image feature corresponding to each image block, rearranging and combining initial image features in the initial image feature set to obtain a first image feature set and a second image feature set, first image features in the first image feature set and second image features in the second image feature set corresponding to different rearrangement and combination manners, pre-training an image classification model based on the first image feature set and the second image feature set, the image classification model being configured to classify content in an image, and fine-tuning the pre-trained image classification model based on a labeled sample image.

Abstract: A virtual object control method includes: displaying a virtual environment screen and a target function control, the target function control being configured to control the first virtual object to perform a target operation which triggers change of a target attribute value of a second virtual object; determining, in response to a trigger operation on the target function control, one or more second virtual objects with target attribute values satisfying a preset condition as one or more candidate virtual objects, and displaying one or more candidate object identifiers of the one or more candidate virtual objects; and determining, in response to a selection operation on one of the one or more candidate object identifiers, that a selected candidate object identifier is a target object identifier, and controlling the first virtual object to perform the target operation on a target virtual object corresponding to the target object identifier.

Abstract: An image classification model training method and apparatus are provided. Classification results of each image outputted by an image classification model are obtained. When the classification results outputted by the image classification model do not meet a reference condition, a reference classification result is constructed based on the classification results outputted by the image classification model. Because the reference classification result can indicate a probability that images belong to each class, a parameter of the image classification model is updated to obtain a trained image classification model based on a total error value between the classification results of the each image and the reference classification result.

Abstract: The disclosure provides a virtual prop control method and apparatus, a storage medium, and an electronic device. The method includes displaying a target virtual throw prop controlled by a first virtual operation object in a target game application. The method also includes controlling the throwing of the target virtual throw prop based on a target throw operation. The method also includes setting a target region at which the virtual throw prop is thrown as a damage detection region, with the damage detection region configured to reduce hit points of at least one target virtual operation object located in the damage detection region. Based on a second virtual operation object being included in the at least one target virtual operation object and entering the target region, the method also includes reducing hit points of the second virtual operation object.

Abstract: A cloud application-based resource transfer method, apparatus, computer device and storage medium. The method includes: displaying, through a cloud application, a video interface generated after processing through a cloud; displaying, in response to a trigger operation on a target visual element in the video interface used for triggering resource transfer, a resource transfer channel interface generated by the cloud; displaying a target resource transfer channel in the resource transfer channel interface; displaying resource transfer-related information corresponding to the target resource transfer channel and generated by the cloud; and triggering, in response to a resource transfer confirmation operation for the resource transfer-related information, a resource transfer operation that is based on the resource transfer-related information through the target resource transfer channel.

Abstract: A transaction data processing method includes executing M transactions among N transactions in parallel during a process of packing the N transactions into a block and performing block consensus on the block, to obtain first transaction execution results. The method further includes identifying, after the block consensus succeeds, transactions identical to the M transactions as a first type of transactions in the block, and classifying the first transaction execution results associated with the M transactions as pre-execution transaction execution results. The method further includes identifying transactions in the block other than the first type of transactions as a second type of transactions, and executing, after the block consensus succeeds, the second type of transactions in parallel to obtain second transaction execution results. The method further includes combining the pre-execution transaction execution results with the second transaction execution results in sequence.

Abstract: A method for editing an electronic document includes: opening a second editing interface of an electronic document in a second client by using an insertion collaboration entry provided by a first client on a first computer device, the insertion collaboration entry being used for collaborating with the first client to perform a multimedia file insertion into the electronic document; obtaining a multimedia file; and inserting the multimedia file into a first insertion position in the second editing interface of the electronic document.

Abstract: A medical image processing method and apparatus, and an image processing method and apparatus, terminal and storage medium that obtains a to-be-processed medical image; generates a difference image according to the first image data, the second image data, and the third image data included in the to-be-processed medical image; and performs binarization processing on the difference image to obtain a binarized image, a foreground region of the binarized image corresponding to a pathological tissue region of the to-be-processed medical image. A difference image is generated based on color information of different channels before binarization processing is performed on an image, thereby effectively using the color information in the image. The pathological tissue region extracted based on the difference image is more accurate and facilitates subsequent image analysis.