Abstract: A video memory management method is provided. The method includes: determining priorities of a plurality of machine learning tasks executed by a graphics processing unit; if video memory resources are to be allocated for a higher-priority task, and an amount of allocatable video memory resources is smaller than an amount of video memory resources required by the higher-priority task, releasing at least a part of video memory resources occupied by a lower-priority task; and allocating video memory resources to the higher-priority task, wherein the higher-priority task is executed at least according to tensor data in a video memory space.

Abstract: Virtual machine hot migration method, device, electronic device, and computer storage medium are provided. The method includes: switching a first page table used for indicating a mapping relationship between a memory address of a virtual guest and a physical address of a host machine in a first physical machine to a second page table according to a switching trigger instruction; and performing a hot migration of the virtual guest from the first physical machine to a second physical machine according to the second page table, wherein a size of a physical memory page indicated by a last-level page table of the second page table satisfies a defined size, and a size of a physical memory page indicated by a last-level page table of the first page table is greater than the defined size. The embodiments of the present disclosure improve the performance of the virtual guest and its hot migration.

Abstract: Embodiments of the present disclosure provide frequency point changing methods and apparatuses.

Abstract: The present disclosure provides methods and apparatuses for image processing and image classification. In one embodiment, the method for image processing comprises: receiving an image; obtaining a first classification result for the image based on a classification model; processing the image for classification based on a preset process, and providing a processing result into a re-ranking model to obtain a second classification result for the image; and determining a target classification result for the image, based on the first classification result and the second classification result.

Abstract: A method including acquiring virtual location information of a virtual data buffer of a virtual network interface component; designating physical location information of a physical data buffer of a physical network interface component as the virtual location information; and storing data according to the virtual location information when the data is being stored into the physical data buffer. Since the physical location information of the physical data buffer of the physical network interface component is designated as the virtual location information instead of being determined via allocation by a physical device. Thus, the physical network interface component is capable of storing, upon receiving data, the data into a data buffer designated by the virtual network interface component, thereby eliminating the process of copying data from a data buffer designated by the physical network interface component to the data buffer designated by the virtual network interface component.

Abstract: Disclosed are systems and methods that determine whether instances of data (e.g., forward activations, backward derivatives of activations) that are used to train deep neural networks are to be stored on-chip or off-chip. The disclosed systems and methods are also used to prune the data (discard or delete selected instances of data). A system includes a hierarchical arrangement of on-chip and off-chip memories, and also includes a hierarchical arrangement of data selector devices that are used to decide whether to discard data and where in the system the data is to be discarded.

Abstract: The present disclosure provides methods and apparatuses for processing video content. The method can include: determining whether a picture-header (PH) network-abstraction-layer (NAL) unit is present in a picture unit (PU); in response to the PH NAL unit being present in the PU, determining a first video-coding-layer (VCL) NAL unit of the PU to be a first VCL NAL unit following the PH NAL unit, or in response to the PH NAL unit being not present in the PU, determining that the PU has only one VCL NAL unit and the first VCL NAL unit of the PU is the VCL NAL unit in the PU; and determining whether the first VCL NAL unit of the PU is the first VCL NAL unit of an access unit (AU) containing the PU.

Abstract: The present disclosure provides a computer-implemented method for decoding video. The method includes receiving a bitstream including a coding unit of a video frame, the coding unit being coded in a skip mode or a direct mode; determining whether the coding units has a width greater than a threshold width or a height greater than a threshold height; in response to the coding unit having the width greater than the threshold width or the height greater than the threshold height, enabling a motion vector angular prediction process to store one or more motion vector angular prediction candidates in a candidate list storing a plurality of motion candidates associated with the coding unit; and performing an inter prediction to the coding unit according to the candidate list and an index parsed from the bitstream.

Abstract: The present disclosure provides a computer-implemented method for encoding video. The method includes: determining whether a coded video sequence (CVS) contains equal number of profile, tier and level (PTL) syntax structures and output layer sets (OLSs); and in response to the CVS containing equal number of PTL syntax structures and OLSs, coding the bitstream without signaling a first PTL syntax element specifying an index, to a list of PTL syntax structures in the VPS, of a PTL syntax structure that applies to a corresponding OLS in the VPS.

Abstract: The present disclosure provides storage nodes, integrated circuits, and methods for storage node management. A method for storage node management includes performing one or more network tasks by one or more first processing cores in a network engine of an integrated circuit; and performing one or more storage tasks by one or more second processing cores in a storage engine of the integrated circuit.

Abstract: The present disclosure provides a video data processing method. The method includes: receiving a bitstream; decoding a first index from the bitstream; determining a maximum number of an adaptive loop filter (ALF) for a component of a picture based on the first index; and processing pixels in the picture with the ALF.

Abstract: Provided in embodiments of the disclosure are methods, apparatuses, and computer-readable media for data processing. Location information and personalized information of a user are obtained in real-time; then a target spatial region corresponding to the location information is determined, the target spatial region including at least one physical object matching the personalized information; and then shopping guide data is generated for the user based on the at least one physical object matching the personalized information. By obtaining the real-time location of the user in combination with the personalized information of the user such as a purchase preference of historical purchase behavior, a shopping place and a branded shop near the user are provided to guide the user to perform in-store shopping, thereby improving purchase efficiency for the user, effectively providing personalized shopping recommended content to the user, and greatly improving purchase experience of the user.

Abstract: Apparatus and methods are provided for improving thermal control, including collecting data of a plurality of systems, each of the plurality of systems including at least one first cooling element and at least one first heat-generating element; conducting a first simulation using a simulation model based on the collected data to generate a first set of simulation results; conducting a first training on a control system using the first set of simulation results to obtain a first trained control system; and using the first trained control system to monitor a field system with a space having at least one second cooling element and at least one second heat-generating element and to control the at least one second cooling element and the at least one second heat-generating element.

Abstract: The present disclosure provides methods, systems, and non-transitory computer readable media for migrating data in a data storage system. A method for migrating data comprises: transferring data from a storage cluster in a first data center to a migration cartridge of the first data center; converting, by the migration cartridge of the first data center, the data's format into an archival format; storing the data into a media card; and moving the media card from the first data center to a second data center, wherein the data is read from the media card into a migration cartridge of the second data center and stored into an archival cluster in the second data center according to the archival format of the data.

Abstract: The present application discloses a method, device, and system for processing a medical image. The method includes obtaining, by one or more processors, a first medical image, wherein the first medical image includes a representation of a target organ, and processing, by the one or more processors, the first medical image based at least in part on a first machine learning model. A classification result associated with the first medical image and a first image of a target region in the first medical image are obtained based at least in part on the processing of the first medical image using the first machine learning model.

Abstract: A method for video data processing includes: determining a category index of a target chroma sample, wherein the category index is determined based on a first reconstructed value associated with a co-located luma sample and a second reconstructed value associated with the target chroma sample; determining an offset based on the category index; and adding the offset to a third reconstructed value associated with the target chroma sample.

Abstract: A method including in response to voice information, acquiring a response object to which corresponding voice information is registered and which occupies an interactive medium; and informing the response object to respond to the voice information. The technical solution enables an operating system to automatically determine a response object responding to voice information triggered by a user from numerous response objects in view of voice information registered at each response object and the occupation of interactive media by each response object. The technical solution removes the restriction that a user must speak a wakening phrase to activate a required response object each time before speaking voice information. Thus, a user only needs to speak short voice information that shows the intention for use thereof during a process of using a certain response object, thereby improving the convenience of a user performing voice interaction.

Abstract: Systems, devices, and methods for determining prediction weight for sub-block are provided. One exemplary method includes: partitioning a coding unit into a plurality of sub-blocks; determining a weight of a sub-block of the plurality of sub-blocks based on weights of control points, wherein the control points are neighboring corners of the coding unit; and bi-predicting the sub-block based on the determined weight.

Abstract: Methods and apparatuses video processing include: in response to receiving a video sequence, encoding first flag data in a parameter set associated with the video sequence, wherein the first flag data represents whether gradual decoding refresh (GDR) is enabled or disabled for the video sequence; when the first flag data represents that the GDR is disabled for the video sequence, encoding a picture header associated with a picture in the video sequence to indicate that the picture is a non-GDR picture; and encoding the non-GDR picture.

Abstract: Embodiments of the present disclosure provides systems and methods for processing video content. The methods include: reconstructing a chroma block based on a plurality of luma samples associated with the chroma block, wherein reconstructing the chroma block includes: determining whether the chroma block has non-zero chroma residuals; and in response to a determination that the chroma block has no non-zero chroma residuals, bypassing a process of averaging the plurality of luma samples, wherein the process of averaging is used for reconstructing the chroma block.