TENCENT TECHNOLOGY (SHENZHEN) COMPANY LTD

An image processing method and apparatus are provided. During image matting on an original image, a plurality of segmented images including different regions are first obtained through semantic segmentation. Further, according to the segmented images, lines of different widths are drawn on a contour line of a foreground region to obtain a target trimap. Finally, a target image is generated based on the target trimap. For the target trimap, because lines of different widths are drawn on the contour line of the foreground region, pertinent image matting may be implemented for different regions, so that image matting precision may be improved for a region requiring fine image matting, and image matting precision of another region can also be ensured. In this way, a final image obtained through image matting is fine and natural.

An electroencephalogram (EEG) signal classification method and apparatus, a device, a storage medium, and a program product are provided, and relate to the field of signal processing technologies. The method includes: obtaining a first EEG signal; obtaining time-frequency feature maps of at least two electrode signals in the first EEG signal; performing feature extraction based on the time-frequency feature maps of the at least two electrode signals to obtain a first extracted feature map; performing weighting processing based on an attention mechanism on the first extracted feature map to obtain an attention feature map; and obtaining a motor imagery type of the first EEG signal based on the attention feature map.

A video data processing method, performed by a computer device, includes: obtaining reconstructed video data; obtaining one or more candidate pixel threshold pairs, wherein at least one of the one or more candidate pixel threshold pairs include two candidate pixel thresholds having first relationship that is asymmetric; determining one or more distortion costs associated with correcting a target reconstructed video frame of the reconstructed video data in a target correction mode according to the one or more candidate pixel threshold pairs; screening, according to the one or more distortion costs, a target pixel threshold pair of the reconstructed video data, in the target correction mode, from the one or more candidate pixel threshold pairs; and obtaining a target correction result of the reconstructed video data by correcting, in the target correction mode according to the target pixel threshold pair, one or more reconstructed video frames of the reconstructed video data.

A graph data processing method includes acquiring a directed graph, where a directed edge in the directed graph is represented as pointing to a destination vertex from a start vertex; representing the directed edge in a unified form according to a specified order between a vertex identifier of the start vertex and a vertex identifier of the destination vertex; generating a tagged edge for recording an original edge direction of the directed edge, to obtain a tagged directed graph; and identifying a category of a triangle constituted by a vertex in the tagged directed graph, a neighbor vertex of the vertex, and a common vertex commonly adjacent to the vertex and the neighbor vertex, based on tagged edges between two of the vertex, the neighbor vertex, and the common vertex.

A qubit assembly preparation method includes preparing a waveguide film in two or more regions on a substrate spaced apart from each other. The method further includes preparing, by using a Dolan bridge photoresist structure, a qubit structure not connected to the waveguide film. The qubit structure includes a three-layer structure. The three-layer structure includes a first superconducting portion and a second superconducting portion intersecting with each other in a coverage region on the substrate. The method further includes removing the insulation layer on a first target region and a second target region of an upper surface of the superconducting portion. The method further includes evaporating a connection layer on the waveguide film, the qubit structure, and the substrate between the waveguide film and the qubit structure.

A display method and apparatus for a virtual prop, an electronic device and a non-transitory computer-readable storage medium. The method includes: displaying a virtual prop list in a virtual scene, the virtual prop list having an interactive option for at least one virtual prop, displaying based on a slide operation on the interactive option for a target virtual prop in the virtual prop list meeting a first target condition, a first target region, and marking the target virtual prop based on an operation end point of the slide operation being within the first target region.

A method for training a speech enhancement network, performed by an electronic device, includes: acquiring a first clean speech sample and a noise sample, and mixing them to generate a noisy speech sample; performing noise reduction on the noisy speech sample based on the speech enhancement network to obtain an enhanced speech sample; framing the enhanced speech sample into a plurality of enhanced speech frames, classifying speech effectiveness of the enhanced speech frames, and generating a first effectiveness distribution based on classification results of the enhanced speech frames; and determining a noise reduction accuracy based on the enhanced speech sample and the first clean speech sample, determining a speech classification accuracy based on the first effectiveness distribution, determining a speech enhancement accuracy based on the noise reduction accuracy and the speech classification accuracy, and training the speech enhancement network based on the speech enhancement accuracy.

An audio data processing method and apparatus, a device, and a storage medium that can obtain, in a game voice mode, a signal processing result associated with a first pre-signal processing policy in an application layer of a service application; control, in the application layer according to the signal processing result, a turned-on/turned-off state of a second optimization component in a second pre-signal processing policy in a terminal system layer, or a turned-on/turned-off state of a first optimization component in the first pre-signal processing policy, a first optimization component turned on in the first pre-signal processing policy being different from a second optimization component turned on in the second pre-signal processing policy; and obtain uplink voice data of a first user in the game voice mode, and perform voice optimization on the uplink voice data based on the turned-on first optimization component and the turned-on second optimization component.

A virtual object control method and apparatus, a terminal, and a storage medium. The method includes: displaying a virtual scene interface, the virtual scene interface including a first virtual object, second virtual objects, and ability controls; determining, in response to a first trigger operation on a first ability control, a target virtual object from the second virtual objects and according to the first trigger operation, and controlling the first virtual object to cast a first ability on the target virtual object; determining the target virtual object as an ability casting target corresponding to a second ability when a second trigger operation on a second ability control meets an ability continuous casting condition, the second trigger operation being triggered during ability casting of the first ability; and controlling the first virtual object to cast the second ability on the target virtual object.

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.