Abstract

Provided in embodiments of the present application are a communication method, an apparatus, a computer-readable medium and an electronic device. The communication method comprises: generating a wireless communication message frame, the wireless communication message frame including information of each device which must access a channel, and an access sequence of said each device; if information of a usable transmission channel is acquired, sending the wireless communication message frame so as to start a data transmission process, and indicating that each device send communication data by means of the transmission channel in sequence in the data transmission process according to the access sequence thereof. Technical solutions of embodiments of the present application can improve channel utilization and communication efficiency between devices.

IPC Classes  ?

• H04W 74/00 - Wireless channel access, e.g. scheduled or random access

Abstract

A video stream decoding method. The video stream decoding method comprises: acquiring multiple video streams inputted by a monitoring device; creating a plurality of threads in a thread pool corresponding to a graphics processing unit; and transmitting, according to the plurality of threads, the multiple video streams to the graphics processing unit for video decoding processing, so as to obtain multi-channel decoded video data.

IPC Classes  ?

• H04N 19/44 - Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder
• H04N 19/436 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals - characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation using parallelised computational arrangements

Abstract

A data binding method, apparatus, and device of a mini program, and a storage medium. The mini program is a program that depends on a host program to run. The method comprises: displaying a mini program fabrication interface (10) of a visual fabrication program, wherein a panel area (101) and an editing area (102) are displayed in the mini program fabrication interface (10), the panel area (101) is provided with n basic UI controls (310), and n is a positive integer; when a user operation on a selected basic UI control is received, editing in the editing area (102) to obtain a program interface of the mini program according to the selected basic UI control (320); obtaining a data source (330); and when a data binding operation corresponding to a target basic UI control on the program interface of the mini program is received, binding the target basic UI control to the data source according to the data binding operation, the data source being used by the target basic UI control for displaying (340).

IPC Classes  ?

• G06F 9/44 - Arrangements for executing specific programs

Abstract

The present application relates to a sign language recognition method and apparatus, a computer-readable storage medium, and a computer device. The method comprises: acquiring a sign language video on which recognition is to be performed; extracting a gesture feature from each frame of image in the sign language video; extracting a gesture change feature of each frame of image in the sign language video; extracting sign language word information from a fused feature obtained by fusing the gesture feature and the gesture change feature; and combining the sign language word information into a sign language sentence according to contextual information corresponding to the sign language word information.

IPC Classes  ?

• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

Abstract

Embodiments of the present application provide a data synchronization method and apparatus for a distributed system, a computer readable medium, and an electronic device. The method comprises: obtaining historical data synchronization information of an adjacent node connected to a target node; determining a selection weight value of the adjacent node according to the historical data synchronization information, wherein the selection weight value is used for indicating the efficiency of synchronizing data from the adjacent node to the target node; and selecting, according to the selection weight value of the adjacent node, a target adjacent node for synchronizing data to the target node, so as to synchronize target data in the target adjacent node to the target node.

IPC Classes  ?

• H04L 29/08 - Transmission control procedure, e.g. data link level control procedure
• G06F 16/23 - Updating

Abstract

A character recommending method and apparatus, and a computer device and a storage medium. The method comprises: converting an entered code in a code entry interface into a syntax tree (302), wherein the syntax tree comprises multiple nodes, a hierarchical relationship among the multiple nodes, and a position range of the multiple nodes; determining, according to a position of a cursor and the syntax tree in the code entry interface, at least one reference node corresponding to the position of the cursor in the syntax tree; and parsing the at least one reference node, and determining target characters to be recommended according to a parsed result (305). In the code entry interface, the target characters are recommended, and the target characters to be entered later are recommended according to the currently entered code, without needing a developer to enter the characters by himself, especially without needing to switch a keyboard interface when the developer needs to switch a character type, so that the developer is facilitated to write a code, operation is simplified, entering time is saved, and entering efficiency is improved.

IPC Classes  ?

• G06F 3/023 - Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes

Abstract

Disclosed in embodiments of the present invention are an image processing method and device, a terminal, and a medium. The image processing method is executable by a terminal, and comprises: acquiring a sample image, a category label of the sample image and a label value of the category label; calling a preset image processing model, performing segmentation on the sample image according to feature similarity between respective pixels in the sample image, and obtaining at least two sub-regions; calling the preset image processing model, performing a category prediction operation with respect to the sample image according to a relevance between the respective sub-regions and the category label, and obtaining a predicted value for the category label; updating network parameters of the preset image processing model according to central coordinates of the respective sub-regions, the label value of the category label and the predicted value of the category label; and performing iterative training on the preset image processing model according to the updated network parameters, and obtaining a target image processing model.

IPC Classes  ?

• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

Abstract

Disclosed are a mini-program production method and apparatus, and a terminal and a storage medium, belonging to the field of programming. The method comprises: displaying a mini-program production interface, wherein a panel region and an editing region are displayed on the mini-program production interface, and n basic UI elements are provided in the panel region, n being a positive integer; in response to the reception of an arrangement operation on a target basic UI element, obtaining, in the editing region, a program interface of a mini-program by means of editing according to the target basic UI element, wherein the target basic UI element is a UI element selected from the n basic UI elements; and generating a program packet of the mini-program according to the program interface. Compared with a development process of a relevant mini-program, by using the method of the present application, a program related to the current operation can be synchronously generated, by means of receiving a visible operation of a user, in the background during the whole process, such that an ordinary user can complete production of a mini-program, without manually editing codes.

IPC Classes  ?

• G06F 8/38 - Creation or generation of source code for implementing user interfaces

Abstract

A machine translation method. The method comprises: receiving a sentence to be translated, wherein the sentence to be translated comprises a word to be translated; invoking a machine translation model obtained by means of training, wherein the machine translation model comprises a segmentation model and a translation model; based on the segmentation model, segmenting the sentence to be translated according to the word vector of the word to be translated so as to obtain a segment to be translated, wherein the segment to be translated comprises at least one word to be translated; and based on the translation model and the segment to be translated, translating the sentence to be translated so as to obtain a translation result.

IPC Classes  ?

• G06F 40/40 - Processing or translation of natural language

Abstract

The present application discloses a message pushing method, a storage medium, and a server, pertaining to the field of Internet of Vehicles. The method comprises: acquiring basic data associated with a target vehicle, wherein the basic data at least comprises vehicular user data and driving environment data; performing recognition on a current scenario of the target vehicle on the basis of the acquired basic data; determining, according to an acquired scenario recognition result, at least one service currently needed by the target vehicle, and generating a service message of the at least one service; and pushing the generated service message to the target vehicle. The present application achieves active scenario recognition, actively determines, on the basis of the scenario recognition result, the service that may be needed by a user in the current time and space, and actively pushes the corresponding service message to the vehicle side. Since the user can receive a desired service without performing any additional operations, the method greatly improves driving safety in an in-vehicle environment, meets service requirements of the user in the in-vehicle environment, and is effective.

IPC Classes  ?

• H04L 29/08 - Transmission control procedure, e.g. data link level control procedure
• G06F 16/9537 - Spatial or temporal dependent retrieval, e.g. spatiotemporal queries

Abstract

Embodiments of the present application disclose a resource display method, a device, an apparatus, and a storage medium, pertaining to the technical field of computers. The method comprises: acquiring one or more target sub-videos of a target video; acquiring, on the basis of image frames of any one of the target sub-videos, one or more key frames of said target sub-video; dividing, according to color clustering, any one of the key frames of said target sub-video into multiple regions, setting a region meeting an area requirement in the multiple regions as a selectable region of said key frame; setting the selectable region of each of the key frames as selectable regions of said target sub-video; and selecting a target region from selectable regions of the target sub-videos, and displaying a resource in the target region. The invention automatically divides a key frame into multiple regions by means of color clustering, selects a target region from selectable regions to display a resource, and accordingly determines a suitable position for displaying the resource by means of an automatic search, thereby achieving high efficiency in displaying resources, and saving time and on manual labor cost.

IPC Classes  ?

• H04N 21/4782 - Web browsing

Abstract

The present application provides a data processing method and apparatus, a computer-readable storage medium, and an electronic device. The method comprises: a server obtains a flow control signaling transmitted by an interface machine, wherein the interface machine is used for obtaining, from the server, audio and video data to be transmitted to a client; in response to the flow control signaling, the server obtain a first rate at which the interface machine transmits the audio and video data; when the first rate is less than a first predetermined threshold, the server transmits audio data to the interface machine, and discards video data that is located at the same picture group as the audio data.

IPC Classes  ?

• H04N 21/643 - Communication protocols
• H04N 21/238 - Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams

Abstract

Disclosed are a method and apparatus for providing a multi-player room, and a device and a medium, relating to the field of computer networks. The method comprises: receiving a creation request, sent by a client, for a cloud room, wherein the creation request carries configuration information of the cloud room, and the cloud room is used by at least two clients to play a multi-player game; creating and configuring the cloud room at a cloud server according to the configuration information, wherein the cloud room is used for running the multi-player game; receiving an access request, sent by the client, for the cloud room; and enabling, according to the access request, the cloud room to allow the client access to the cloud room. The creation request that carries the configuration information of the cloud room is sent, by the client, to the cloud server to create the cloud room, and the client sends the configuration information of the cloud room to the cloud server, thereby simplifying a creation means of the cloud room.

IPC Classes  ?

• H04L 29/06 - Communication control; Communication processing characterised by a protocol

Abstract

A method and device for generating a neural network model in games, and a computer-readable storage medium. The method comprises: determining hyperparameter values corresponding to hyperparameters in a first neural network model from a hyperparameter value set (S102); obtaining first model accuracy of a second neural network model obtained by training the first neural network model, wherein the first model accuracy is used for indicating the accuracy of performing detection on a verification sample set by using the second neural network (S104); and if the first model accuracy is lower than target accuracy, updating the hyperparameter values of some hyperparameters in the first neural network model to corresponding target hyperparameter values in the hyperparameter value set to obtain a third neural network model (S106).

IPC Classes  ?

• G06N 3/04 - Architecture, e.g. interconnection topology

Abstract

Provided in embodiments of the present application are a method and apparatus for detecting an information embedding region, an electronic device, and a computer storage medium. The method comprises: obtaining a video to be embedded, and segmenting the video to be embedded so as to obtain video segments; obtaining a target frame in the video segments, and identifying and segmenting an object in the target frame so as to obtain annotation information corresponding to the object; determining a target object according to the annotation information, and clustering the target object so as to obtain a plurality of candidate regions to be embedded; and determining a target candidate region to be embedded from among the candidate regions to be embedded, and performing a maximum rectangular search on the target candidate region to be embedded so as to obtain a target region to be embedded.

IPC Classes  ?

• H04N 21/845 - Structuring of content, e.g. decomposing content into time segments

Abstract

Embodiments of the present invention relate to a machine learning technology in the field of artificial intelligence and provide a neural network model training method and apparatus, an image processing method, an electronic device, a medical image processing system, and a computer readable storage medium. The neural network model training method comprises: performing initial training by using a first training sample set to obtain an initial neural network model; predicting a second training sample set by using the initial neural network model to obtain a prediction result of each training sample of the second training sample set; determining multiple preferred samples from the second training sample set on the basis of the prediction results; receiving a labeling result for the multiple preferred samples, and adding the multiple labeled preferred samples into the first training sample set to obtain an extended first training sample set; performing update training by using the extended first training sample set to obtain an updated neural network model; when a training ending condition is satisfied, ending the training; and when the training ending condition is not satisfied, repeating the steps of prediction, determination of preferred samples, sample extension and training updating.

IPC Classes  ?

• G06N 20/00 - Machine learning
• G06K 9/62 - Methods or arrangements for recognition using electronic means

Abstract

An image classification method, apparatus and device, a storage medium, and a medical electronic device. The image classification method comprises: receiving a target image and a reference image related to the target image, wherein the target image is a medical image; determining a first image feature of the target image and a second image feature of the reference image in the same manner; fusing the first image feature and the second image feature to determine a feature of an image to be classified; and determining, by means of the feature of the image to be classified, the probability of the target image belonging to a preset category.

IPC Classes  ?

• G06K 9/62 - Methods or arrangements for recognition using electronic means

Abstract

A clock drift processing method. An SMF network element obtains a drift range value of the clock domain of a data network relative to the clock domain of a wireless network, and then sends the drift range value to a UPF network element. The UPF network element sends a notification message to the SMF network element only upon detecting that the maximum drift value is greater than the drift range value, and triggers the SMF network element to initiate a PDU session modification process and to send an updated TSCAI to a RAN device, thus saving signaling overheads while achieving accurate time control over data flows in the clock domain of the data network.

IPC Classes  ?

• H04J 3/06 - Synchronising arrangements

Abstract

Provided are a multimedia data processing method and apparatus, a storage medium and an electronic device, which relate to the technical field of communications. The multimedia data processing method comprises: establishing a session between a terminal device and a media processing module and controlling the session by means of a signaling processing module deployed at a remote node; and after the session is established with the terminal device, receiving multimedia data sent by the terminal device and processing the multimedia data by means of the media processing module deployed at an edge computing node. The present disclosure can reduce the system deployment complexity and deployment costs, improve the operating and maintenance efficiency and reduce operating and maintenance costs.

IPC Classes  ?

• H04L 29/06 - Communication control; Communication processing characterised by a protocol
• H04L 29/08 - Transmission control procedure, e.g. data link level control procedure

Abstract

Provided in the present application are a 3D image classification method and apparatus, a device and a storage medium. The method comprises: obtaining a 3D image, the 3D image comprising first-dimensional image information, second-dimensional image information, and third-dimensional image information; extracting from the 3D image a first image feature corresponding to plane image information; extracting from the 3D image a second image feature corresponding to the third-dimensional image information; fusing the first image feature and the second image feature to obtain a fused image feature corresponding to the 3D image; and determining a classification result corresponding to the 3D image according to the fused image feature corresponding to the 3D image. According to the technical solution provided in embodiments of the present application, the efficiency and accuracy with which a 3D image is classified may be improved.

IPC Classes  ?

• G06K 9/62 - Methods or arrangements for recognition using electronic means

Abstract

Embodiments of the present application provide an information embedding method and device, a server, and a computer storage medium. The method comprises: acquiring background tilt information of a preconfigured entity subject to embedding in a preconfigured video frame awaiting embedding; acquiring at least one piece of foreground tilt information corresponding to at least one piece of preconfigured multimedia information, wherein each of the at least one piece of foreground tilt information is tilt information of a surface to be contacted of the corresponding preconfigured multimedia information; acquiring a tilt difference between the background tilt information and the at least one piece of foreground tilt information, and obtaining at least one piece of tilt difference information; determining, according to the at least one piece of tilt difference information, target multimedia information that meets a preconfigured tilt difference condition from the at least one piece of preconfigured multimedia information; and embedding the target multimedia information into a carrying surface of the preconfigured entity of the preconfigured video frame, and acquiring a target video frame.

IPC Classes  ?

• H04N 21/2668 - Creating a channel for a dedicated end-user group, e.g. by inserting targeted commercials into a video stream based on end-user profiles

Abstract

Disclosed in the present application is a communication method, comprising: when a first network function entity communicates indirectly with a second network function entity by means of a first service communication proxy (SCP) network element, if the first SCP network element fails, indirect communication can be performed by means of a second SCP network element, or direct communication can be performed. In this way, when the SCP network element used for communication between the first network function entity and the second network function entity fails, it can be ensured that the first network function entity and the second network function entity can continue to perform business communication, so as not only to ensure business continuity, but also to ensure the reliability and availability of a system.

IPC Classes  ?

• H04W 36/00 - Handoff or reselecting arrangements
• H04W 28/08 - Load balancing or load distribution

Abstract

The present invention relates to the field of communications. Disclosed are a broadcast method and device for a virtual network group, an apparatus, and a system. The method comprises: receiving a broadcast data packet sent by one member in a virtual network group, a destination address of the broadcast data packet being a broadcast address; sending the broadcast data packet to a target internal broadcast interface of at least two internal broadcast interfaces; and sending, by means of the target internal broadcast interface, the broadcast data packet to all members in the virtual network group or all local members. The present invention realizes broadcast communication in a 5G VN.

IPC Classes  ?

• H04W 4/06 - Selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS]; Services to user groups; One-way selective calling services
• H04W 4/08 - User group management

Abstract

Disclosed by the present application are a multicast method, apparatus, device and system for a multicast group of a virtual network group, which are connected to the field of communications. The method comprises: receiving a multicast data packet sent by a member of the virtual network group, the destination address of the multicast data packet being a multicast address; sending the multicast data packet to a target multicast internal interface among at least two multicast internal interfaces; and sending the multicast data packet to all members of the virtual network group or all local multicast group members by means of the target multicast internal interface. The present application achieves multicast communication in a 5G VN.

IPC Classes  ?

• H04L 12/18 - Arrangements for providing special services to substations for broadcast or conference
• H04L 12/46 - Interconnection of networks
• H04W 4/06 - Selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS]; Services to user groups; One-way selective calling services

Abstract

Provided are a communication method and apparatus, and a computer-readable medium and an electronic device. The communication method comprises: a core network entity of a first network carrying out communication negotiation with a core network entity of a second network, and acquiring a communication resource pool allocated, by the core network entity of the second network, to a terminal device located in the first network, wherein the first network and the second network share an access network; and configuring the communication resource pool for the terminal device, such that the terminal device uses, after moving to the second network, communication resources in the communication resource pool for communication. The technical solution in the embodiments of the present application can ensure the communication continuity of a terminal device during a network switching process, thereby facilitating an improvement in the communication efficiency of the terminal device.

IPC Classes  ?

• H04W 72/04 - Wireless resource allocation

Abstract

An image region recognition method, comprising: obtaining an image to be segmented, wherein said image comprises a plurality of extreme points (101); generating first image feature information according to said image, wherein the first image feature information comprises N image matrices and a first heat map, the first heat map being generated according to the plurality of extreme points (102); obtaining, by means of a first image segmentation model, a first image segmentation region corresponding to the first image feature information (103); obtaining a second heat map according to annotation points corresponding to the first image segmentation region, and the first image segmentation region (104); obtaining, by means of a second image segmentation model, a second image segmentation region corresponding to said image (105); and generating an image recognition result of said image according to the second image segmentation region (106). According to the method, by means of auxiliary segmentation in the second stage, the region having a poor effect in the image segmentation in the first stage is further segmented, so as to obtain a more accurate image segmentation result, thereby improving the performance of image segmentation.

IPC Classes  ?

• G06T 7/11 - Region-based segmentation
• G06T 7/187 - Segmentation; Edge detection involving connected component labelling

Abstract

Provided are a translation method, apparatus and device based on a machine translation model, and a storage medium. The method comprises: encoding an ith source word embedding of a source sentence into an intermediate vector; decoding the intermediate vector in conjunction with the ith source word embedding to obtain a decoded intermediate vector; fusing the decoded intermediate vector with an (i-1)th target word embedding of a target sentence to obtain a fused intermediate vector; decoding the fused intermediate vector to obtain a decoded word vector; and performing probability prediction on an ith target word according to the decoded word vector, and outputting a prediction result. According to the method, a previous target word embedding is added in a decoding process to serve as input data so as to capture a dependency relationship between words, such that the accuracy of generation of a target sentence is improved, mistakes such as over-translation and under-translation are reduced, and the generated target sentence is more fluent.

IPC Classes  ?

• G06F 17/00 - Digital computing or data processing equipment or methods, specially adapted for specific functions

Abstract

An image identification method and device, an identification model training method and device, and a storage medium. The image identification method comprises: obtaining a target three-dimensional image to be identified (S202); inputting the target three-dimensional image into a first identification model (S204), wherein the first identification model is used for identifying the target three-dimensional image to obtain an image type of the target three-dimensional image, a convolution block of the first identification model is the same as a convolution block of a second identification model for identifying the target three-dimensional image, and the second identification model is a model obtained by training an original identification model by using a target training sample, wherein the target training sample comprises cubes obtained by rotating and sorting N target cubes obtained from a three-dimensional sample image, and N is a natural number greater than 1; and obtaining the image type of the target three-dimensional image output by the first identification model (S206).

IPC Classes  ?

• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

Abstract

Disclosed is a data transmission method, applied to a first device, such as a user equipment (UE) or a user panel function (UPF) network element. The method comprises: acquiring data to be transmitted; determining, according to network type information comprised in an extended Ethernet packet filter set, that the data is time sensitive communication (TSC) data; determining a transmission characteristic of the TSC data according to the extended Ethernet packet filter set; and mapping, according to the transmission characteristic, the TSC data to a quality of service (QoS) flow corresponding to the transmission characteristic for transmission. The present application can transmit different types of TSC data by means of different QoS flows, thereby improving the accuracy of TSC data transmission.

IPC Classes  ?

• H04W 28/24 - Negotiating SLA [Service Level Agreement]; Negotiating QoS [Quality of Service]

Abstract

Embodiments of the present application provide a sidelink monitoring method for vehicle communication and a related device. The sidelink monitoring method for vehicle communication comprises: a first vehicle communication terminal establishing a unicast link with a second vehicle communication terminal; monitoring the communication quality of the unicast link, and determining, according to the communication quality of the unicast link, whether or not the unicast link has experienced a failure; and if it is determined that the unicast link has experienced a failure and has not recovered within a first predetermined period, releasing the unicast link. The technical solution in the embodiments of the application ensures the transmission effectiveness and reliability of unicast links, and helps to improve the utilization rate of wireless communication resources.

IPC Classes  ?

• H04W 24/08 - Testing using real traffic
• H04W 4/40 - Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
• H04W 24/10 - Scheduling measurement reports
• H04W 76/10 - Connection setup
• H04W 76/30 - Connection release
• H04B 17/309 - Measuring or estimating channel quality parameters

Abstract

A multi-layer bounding box determination method, and collision detection and motion control methods and devices. The method for determining a multi-layer bounding box of an object comprises: determining a polyhedron capable of accommodating the object therein, and taking same as a first-layer bounding box (S101); selecting one of a plurality of vertexes of the first-layer bounding box, and taking same as a target vertex (S102); determining the object's support plane which takes a specific direction corresponding to the target vertex as a normal vector, and is closest to the target vertex (S103), the support plane being a plane which passes through a point on the surface of the object and enables the object to be completely located at one side of the support plane; and at least using the support plane to cut the first-layer bounding box, so as to form a smaller bounding box, and taking same as a second-layer bounding box (S104).

IPC Classes  ?

• G06T 17/10 - Volume description, e.g. cylinders, cubes or using CSG [Constructive Solid Geometry]

Abstract

Provided are an audio collection device positioning method, an audio collection device positioning apparatus, an electronic device, a speaker recognition method and system, and a computer-readable storage medium. The audio collection device positioning method comprises: acquiring an image to be tested; recognizing an audio collection device in said image, so as to obtain first coordinate data of the audio collection device; and determining displacement data according to the first coordinate data and historical coordinate data of the audio collection device, and determining, according to the displacement data, the coordinates of the audio collection device in said image. In the present application, the correctness of the first coordinate data can be determined in combination with the historical coordinate data, and the coordinate data is optimized, such that the precision of the obtained coordinates of the audio collection device is further improved.

IPC Classes  ?

• G06T 7/73 - Determining position or orientation of objects or cameras using feature-based methods

Abstract

The present application relates to a detection model training method and apparatus, a computer device, and a storage medium, relating to machine learning in artificial intelligence. The method comprises: obtaining a candidate image area set obtained by segmenting a first sample image; obtaining a first relation degree corresponding to each candidate image area, the first relation degree being a relation degree, about the circumstance that the candidate image area comprises a target object, output after the candidate image area is input into a first detection model; obtaining a second relation degree corresponding to each candidate image area, the second relation degree being a relation degree, about the circumstance that the candidate image area comprises the target object, output after the candidate image area is inputted into a second detection model; obtaining a relation degree change value according to the first relation degree and the second relation degree, and screening out a target image area comprising the target object from the candidate image area set according to the relation degree change value; and performing model training according to the target image area to obtain a target detection model.

IPC Classes  ?

• G06K 9/32 - Aligning or centering of the image pick-up or image-field

Abstract

Disclosed are a translation method, a method and apparatus for training a machine translation model, and a storage medium, relating to the field of artificial intelligence. The method comprises: acquiring a sample sentence labeled with a reference translation result; inputting the sample sentence into a machine translation model to be trained to obtain a sample translation result of the sample sentence (302), wherein the machine translation model is a non-autoregressive model with model parameters to be adjusted; calculating a gradient estimation value of a word translation result according to the reference translation result, wherein the gradient estimation value is obtained through calculation according to a reward value; and adjusting the model parameters of the machine translation model according to a gradient estimation result. The gradient estimation value is calculated according to the reward value, and the model parameters of the machine translation model are adjusted according to the gradient estimation value. A gradient estimation value is obtained by taking a reward value as adjustment reference data so as to adjust the model parameters, thus realizing the sequence-level training of a machine translation model, and thereby improving the translation accuracy of the trained machine translation model.

IPC Classes  ?

• G06F 17/00 - Digital computing or data processing equipment or methods, specially adapted for specific functions

Abstract

The present application discloses a method for controlling multiple virtual characters, a device, an apparatus, and a storage medium, pertaining to the technical field of computers. The method comprises: displaying a first user interface of an application program provided with an augmented reality function, the first user interface comprising selection options for multiple virtual characters; receiving a first selection operation on the first user interface with respect to at least two virtual characters; displaying a second user interface of the application program, the second user interface comprising the at least two virtual characters; receiving a second selection operation on the second user interface; and determining a target virtual character from the at least two virtual characters according to the second selection operation and a rendering order. The method determines the rendering order by means of depth information, and draws the at least two virtual characters according to the rendering order, thereby preventing the situation in which virtual characters clip through one another when superimposed, enabling the target virtual character to be accurately determined when a user performs a selection operation on the virtual characters.

IPC Classes  ?

• G06T 19/00 - Manipulating 3D models or images for computer graphics

Abstract

The present application relates to the technical field of video processing. Disclosed are a video clip positioning method and apparatus, a computer device, and a storage medium. According to the present application, a plurality of video frame characteristics of a target video, and the text characteristics of a target text are obtained by a video recognition model, one candidate clip matching the target text is determined, the candidate clip is finely divided on the basis of the matching degree between a video frame in the candidate clip and the target text to obtain a plurality of subclips, and a subclip having the highest matching degree with the target text is used as a target video clip. The approach of performing characteristics matching for multiple times, dividing the obtained candidate clip again so as to determine the target video clip makes it unnecessary for the video recognition model to learn the boundary characteristics of the target video clip and accurately annotate a sample video during model training, shortens the training period of the video recognition model, and avoids the problem of low accuracy of the model output result caused by the inaccurate sample video annotation.

IPC Classes  ?

• H04N 21/845 - Structuring of content, e.g. decomposing content into time segments
• H04N 21/44 - Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to MPEG-4 scene graphs

Abstract

A perspective switching control method and apparatus, an electronic device and a computer-readable storage medium. The method comprises: after entering a game, displaying a virtual scene image of the game from a target perspective in a client (S302); when a key process in the game is triggered, determining a key perspective matching the key process (S304); and automatically performing switching from the target perspective to the key perspective so as to present a key scene image from the key perspective in the client (S306).

IPC Classes  ?

• A63F 13/5258 - Changing parameters of virtual cameras by dynamically adapting the position of the virtual camera to keep a game object or game character in its viewing frustum, e.g. for tracking a character or a ball
• A63F 13/5252 - Changing parameters of virtual cameras using two or more virtual cameras concurrently or sequentially, e.g. automatically switching between fixed virtual cameras when a character changes room or displaying a rear-mirror view in a car-driving game

Abstract

Provided are a medical image classification method, a model training method and a server. The medical image classification method includes: acquiring a data set of medical images (501); performing quality analysis on the data set of medical images to extract feature information (502); based on the feature information, using a pre-trained neural network to classify the medical images (503) to obtain the classification result.

IPC Classes  ?

• G06T 7/90 - Determination of colour characteristics

Abstract

A method and apparatus for controlling a virtual object to mark a virtual item, and a medium. The method comprises: displaying a virtual environment picture, the virtual environment picture comprising a virtual object and a virtual item located in the virtual environment (401); controlling the virtual object to move to a range around the virtual item in the virtual environment (402); displaying a list of picked up items on the virtual environment picture, with a field of the virtual item available for picking up being displayed in the list of picked up items (403); and receiving a mark operation on the field of the virtual item, and marking the virtual item in the virtual environment according to the mark operation, with the mark operation being an operation different from a pick-up operation (404).

IPC Classes  ?

• A63F 13/822 - Strategy games; Role-playing games 

Abstract

An animation file processing method and apparatus, a computer readable storage medium, and a computer device. The method comprises: obtaining a bitmap image sequence corresponding to an original animation file (S302); when bitmap images in the bitmap image sequence are non-critical bitmap images, encoding difference pixel regions between the bitmap images and corresponding critical bitmap images according to a picture encoding mode to obtain encoding pictures corresponding to the bitmap images (S304); and generating an animation export file corresponding to the original animation file according to the encoding pictures corresponding to the bitmap images in the bitmap image sequence (S306).

IPC Classes  ?

• G06T 13/80 - 2D animation, e.g. using sprites

Abstract

A method and device for image search, a computer apparatus, and a storage medium. The method comprises: displaying an image selection page of an instant messaging client (101); determining, on the basis of an operation performed by a user with respect to the image selection page, an input image for a search operation (102); presetting multiple image similarity dimensions according to the level of similarity between two chat session interactive images (103); searching on the basis of the multiple image similarity dimensions to obtain one or more output images similar to the input image (104), wherein each output image is similar to the input image in at least one image similarity dimension; and generating a search result list according to the one or more output images, and displaying same (105).

IPC Classes  ?

• G06F 16/53 - Querying

Abstract

The present application relates to animation data encoding and decoding methods and apparatuses, a storage medium and a computer device. The animation data encoding method comprises: acquiring, from an animation engineering file, animation data corresponding to each animation tag code; when an attribute type exists in an attribute structure table corresponding to the animation tag code, determining attribute identifier information corresponding to each attribute; encoding, according to the attribute identifier information, attribute values corresponding to attributes in the animation data, to obtain attribute content corresponding to the attributes; and sequentially storing, according to the attribute ranking corresponding to the attributes in the attribute structure table, the attribute identifier information and the attribute content corresponding to the attributes, so as to obtain a dynamic attribute data block corresponding to the animation tag code.

IPC Classes  ?

• G06T 9/00 - Image coding
• G06T 13/00 - Animation

Abstract

Disclosed in the present application are an image processing method, apparatus and device, and a storage medium. The method is executed by a computing device, and comprises: receiving an input image; determining a first image feature of a first size of the input image, wherein the first image feature has at least two channels; determining a first weight adjustment parameter for the first image feature, and performing weight adjustment on each channel in the first image feature by using the first weight adjustment parameter to obtain an adjusted first image feature, wherein the first weight adjustment parameter comprises at least two parameter components, and each parameter component is used for adjusting a pixel point of the corresponding channel; downsampling the adjusted first image feature to obtain a second image feature having a second size; combining the first image feature and the second image feature to obtain a combined image feature; and determining an image processing result according to the combined image feature.

IPC Classes  ?

• G06T 11/60 - Editing figures and text; Combining figures or text

Abstract

The present application discloses a video decoding method and apparatus, and a storage medium. In the case where a playback instruction for a target video is received, target type information concerning the target video is acquired; specified type information concerning a plurality of decoders configured by a terminal is acquired; by comparing the specified type information concerning the plurality of decoders with the target type information, a target decoder among the plurality of decoders supporting decoding of the target video is determined; and the target decoder is used to decode the target video, and the decoded video is played back. A default decoder is not used directly to decode the target video, instead, by comparing the specified type information of the plurality of decoders with the target type information, the target decoder supporting decoding of the target video is selected. Using the target decoder to decode the target video may avoid the decoding failure problem, increasing the video decoding success rate, thereby increasing the video playback success rate.

IPC Classes  ?

• H04N 19/42 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals - characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation

Abstract

The present invention relates to a method and device for rendering an animation, a computer readable storage medium, and a computer apparatus. The method comprises: acquiring an animation file in a target format; when the animation file is decoded, determining, from animation drawing data obtained from decoding, an animation drawing data interval meeting a non-motion condition; buffering initial animation drawing data in the animation drawing data interval; in the process of playing the animation file, if animation drawing data corresponding to a frame to be played meets the non-motion condition, reading the buffered initial animation drawing data corresponding to said frame; and performing animation rendering according to the read initial animation drawing data.

IPC Classes  ?

• G06T 9/00 - Image coding
• G06T 13/00 - Animation

Abstract

Disclosed are a method and apparatus for controlling a virtual object to discard a virtual item, and a medium, which relate to the field of computers. The method comprises: displaying at least one virtual item possessed by a virtual object, and part of a discarding region (401); receiving a first drag operation, wherein the first drag operation is used for dragging target virtual items of the at least one virtual item to part of the discarding region; and the possessing number corresponding to the target virtual items is n, and n is an integer greater than 1 (402); and according to the first drag operation, controlling the virtual object to discard some of the n target virtual items (403), wherein a user can set the number of discarded target virtual items.

IPC Classes  ?

• A63F 13/822 - Strategy games; Role-playing games 
• G06F 3/0488 - Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures

Abstract

The present application provides a method and apparatus for controlling movement of a virtual object, and a terminal and a storage medium. The method comprises: displaying a target view angle picture of a target application program, and displaying a virtual rocker and a movable area of the virtual rocker on the target view angle picture in an overlapping manner; when a trigger operation corresponding to the virtual rocker is obtained, starting the virtual rocker; when the virtual rocker is in the starting state, adjusting the position of the virtual rocker in the movable area according to the position change amount of a touch point, wherein when the touch point moves in the effective touch range, the position of the virtual rocker and the position of the touch point synchronously change in real time, and the effective touch range contains and is larger than the movable area; and controlling the virtual object to move according to the position of the virtual rocker. Compared with the prior art, when the touch point is located outside the movable area of the virtual rocker, under the condition that the virtual rocker moves by the same distance, the moving distance of the touch point is smaller, and the operation efficiency of a user is improved.

IPC Classes  ?

• G06F 3/0488 - Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures

Abstract

A financial product recommendation method and apparatus, and a device and a computer storage medium. The method comprises: receiving a financial product recommendation request of a client, and constructing M types of product recommendation features of each financial product respectively according to historical data of set parameters of each of N financial products, N and M being positive integers (301); for each type of product recommendation features in the M types of product recommendation features, obtaining comprehensive product recommendation features corresponding to the type (302); determining a user recommendation proportion of each financial product according to the deviation degree of each type of product recommendation features of each financial product relative to the comprehensive product recommendation features corresponding to the type (303), the user recommendation proportion being the proportion of the recommended user corresponding to each financial product among all users; and determining a financial recommendation product according to the user recommendation proportion of each financial product to respond to the financial product recommendation request (304).

IPC Classes  ?

• G06Q 40/06 - Investment, e.g. financial instruments, portfolio management or fund management

Abstract

A method and an apparatus for language sequence labeling, a storage medium and a computing device. Said method comprises: reading a first embedded representation of a language sequence, the first embedded representation comprising a character-level word embedded representation, a pre-trained word embedded representation and a global word embedded representation of the language sequence, the global word embedded representation referring to a global context representation of the language sequence (501); on the basis of a first deep-conversion recurrent neural network, performing first deep-conversion encoding on the first embedded representation, and outputting a first hidden layer state representation corresponding to each word in the language sequence (502); and decoding the first hidden layer state representation to obtain the labeling result of the language sequence (503).

IPC Classes  ?

• G06N 3/04 - Architecture, e.g. interconnection topology

Abstract

An information display method and apparatus, and an electronic device and a computer storage medium. The method comprises: obtaining a motion state of a target vehicle in a virtual scene at the current moment, and a first operation for controlling the target vehicle to move in the virtual scene (S202); predicting a first motion trajectory of the target vehicle within a period of time after the current moment by combining the motion state of the target vehicle at the current moment and the first operation (S204), wherein the first motion trajectory comprises a motion line and the motion direction of the target vehicle within a period of time after the current moment; and displaying the first motion trajectory and a second motion trajectory, wherein the second motion trajectory is a motion trajectory of the target vehicle prior to the current moment, and the position where the first motion trajectory and the second motion trajectory meet is the position of the target vehicle at the target vehicle (S206).

IPC Classes  ?

• A63F 13/52 - Controlling the output signals based on the game progress involving aspects of the displayed game scene
• A63F 13/5375 - Controlling the output signals based on the game progress involving additional visual information provided to the game scene, e.g. by overlay to simulate a head-up display [HUD] or displaying a laser sight in a shooting game using indicators, e.g. showing the condition of a game character on screen for graphically or textually suggesting an action, e.g. by displaying an arrow indicating a turn in a driving game
• A63F 13/803 - Driving vehicles or craft, e.g. cars, airplanes, ships, robots or tanks

Abstract

An image recognition device and image recognition method, comprising: acquiring an image to be segmented (101); generating image feature information according to said image (102); acquiring, by means of an image segmentation model, an image segmentation region corresponding to the image feature information (103); and generating an image recognition result of said image according to the image segmentation region (104). The method uses a heat map, generated by extreme points, as a portion of image feature information to enrich features of an image, so as to generate a more accurate image segmentation region, thereby improving the universality and applicability of image segmentation.

IPC Classes  ?

• G06K 9/32 - Aligning or centering of the image pick-up or image-field

Abstract

Disclosed are a method and apparatus for recognizing a facial expression, and an electronic device and a storage medium. The method comprises: extracting a first feature from color information of pixel points in a first image; extracting, from the first image, a second feature of a key point of a face; and performing fusing processing on the first feature and the second feature to obtain a fused feature, and then determining, by means of the fused feature, a first expression type of the face of an object in the first image.

IPC Classes  ?

• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

Abstract

An image processing method and apparatus, and a computer device. The method comprises: obtaining, according to a convolutional neural network, corresponding feature points of various training images in a feature embedding space, and a feature center point of respective images of the types, to which the training images belong, of a second number; performing network training by means of central collision to obtain a target feature center point corresponding to an image of each type; and then, mapping the obtained various feature points and various target feature center points to a Hamming space to obtain a first number of Hash codes and a second number of Hash center points, in which case, network training can still be performed by means of the central collision to obtain an image Hash model. By means of learning the similarities between the Hash codes of the various training images and the similarities between the feature points of the training images and the corresponding central points, global distribution of images is not only learned, but the network training time is also greatly shortened, and the learning efficiency and the precision of the image Hash codes are improved.

IPC Classes  ?

• G06F 16/55 - Clustering; Classification

Abstract

A video target tracking method and apparatus, a computer device and a storage medium. The method comprises: acquiring a local detection map corresponding to a target image frame in a video to be detected (201); acquiring a relative motion saliency map corresponding to the target image frame (202); determining, according to the local detection map and the relative motion saliency map, constraint information corresponding to the target image frame (203); adjusting parameters of an image segmentation model by means of the constraint information, to obtain an adjusted image segmentation model (204); and extracting a target object from the target image frame by means of the adjusted image segmentation model (205).

IPC Classes  ?

• G06T 7/246 - Analysis of motion using feature-based methods, e.g. the tracking of corners or segments

Abstract

An image labeling method, a pathological image-based labeling display method, an apparatus, a device, and a computer storage medium. The image labeling method comprises: acquiring an image to be labeled, said image corresponding to a first magnifying power (101); acquiring a labeled image from a labeled image set, the labeled image corresponding to a second magnifying power, the second magnifying power being inconsistent with the first magnifying power, and the labeled image set comprising at least one labeled image (102); performing matching between the image to be labeled and the labeled image, to obtain an affine transformation matrix (103); and according to the affine transformation matrix and the labeled image, generating labeling information concerning the image to be labeled (104).

IPC Classes  ?

• G06F 16/58 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• G06F 16/51 - Indexing; Data structures therefor; Storage structures

Abstract

An image state determination method and device for pathology images, an apparatus, and a computer storage medium. The method comprises: obtaining a pathology image set, wherein the pathology image set comprises at least an image to be evaluated and an associated image, and the associated image and the image to be evaluated are consecutive image frames (101); determining, according to the pathology image set, a first state corresponding to the image to be evaluated, wherein the first state is used to indicate a motion change of the image to be evaluated (102); and if the first state indicates a stationary state, determining, according to the pathology image set, a second state corresponding to the image to be evaluated, wherein the second state is used to indicate a sharpness change of the image to be evaluated (103).

IPC Classes  ?

• G06T 7/00 - Image analysis

Abstract

A virtual object control method and apparatus, a storage medium, and an electronic device. The method comprises: during an acceleration operation performed by a virtual object controlled by a client, acquiring a first operation instruction, wherein the first operation instruction is used to indicate execution of a first target action; during a first target time period after completion of the first target action, acquiring a second operation instruction, wherein the second operation instruction is used to indicate execution of a second target action; and during a second target time period after the virtual object starts to execute the second target action, performing controlling to switch a movement status of the virtual object from a first state to a second state, wherein per unit of time, the value of first energy acquired by the virtual object in the first state is less than the value of second energy acquired by the virtual object in the second state. The present invention resolves the issue in which operation in the process of controlling a virtual object provided by related techniques is complex.

IPC Classes  ?

• A63F 13/55 - Controlling game characters or game objects based on the game progress

Abstract

An image processing method and apparatus, and an electronic device and a storage medium. The method comprises: acquiring a remaining mileage parameter of a vehicle (S101); acquiring an initial position of the vehicle in a map display interface (S102); determining a target endurance area in the map display interface according to the initial position and the remaining mileage parameter (S103); and displaying the target endurance area in the map display interface (S104).

IPC Classes  ?

• G01C 21/36 - Input/output arrangements for on-board computers

Abstract

An audio clip matching method and apparatus (1200), a computer-readable medium and an electronic device. The matching method comprises: acquiring a first feature sequence corresponding to a first audio clip and a second feature sequence corresponding to a second audio clip (S210); constructing a distance matrix, wherein an element in the distance matrix represents the distance between a first position point in the first feature sequence and a second position point in the second feature sequence (S220); calculating a first cumulative distance from a starting position to a target position in the distance matrix, and calculating a second cumulative distance from an ending position to the target position in the distance matrix (S230); calculating the minimum distance between the first feature sequence and the second feature sequence on the basis of the first cumulative distance and the second cumulative distance (S240); and determining, according to the minimum distance, the degree of matching between the first audio clip and the second audio clip (S250). The matching accuracy of audio clips can be improved.

IPC Classes  ?

• G10L 25/27 - Speech or voice analysis techniques not restricted to a single one of groups characterised by the analysis technique
• G10L 25/51 - Speech or voice analysis techniques not restricted to a single one of groups specially adapted for particular use for comparison or discrimination
• G10L 25/60 - Speech or voice analysis techniques not restricted to a single one of groups specially adapted for particular use for comparison or discrimination for measuring the quality of voice signals

Abstract

An information processing method and device, and a storage medium. The method comprises: obtaining source data (S101); on the basis of a target word feature vector, encoding subdata in the source data to obtain a hidden feature vector corresponding to the subdata, wherein the target word feature vector characterizes an emotion feature standard (S102); obtaining a word feature vector corresponding to the source data on the basis of the hidden feature vector corresponding to the subdata (S103); and inputting the word feature vector into a preset emotion classification network for prediction to obtain an emotion polarity prediction result of the source data (S104). The method can improve the accuracy of emotion polarity prediction.

IPC Classes  ?

• G06K 9/62 - Methods or arrangements for recognition using electronic means
• G06N 3/04 - Architecture, e.g. interconnection topology
• G06N 3/08 - Learning methods

Abstract

Provided is a method for running a standalone program, comprising: running a standalone program in a first main program, and establishing a network connection between a first terminal and a second terminal by means of the first main program running on the first terminal and a second main program running on the second terminal; the first terminal transmitting image information to the second terminal by means of the described network connection, and the second terminal transmitting a second operation command to the first terminal; the standalone program is a program which supports separate control of a first operation object and a second operation object.

IPC Classes  ?

• G06F 3/0487 - Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser
• G06F 3/14 - Digital output to display device
• A63F 13/42 - Processing input control signals of video game devices, e.g. signals generated by the player or derived from the environment by mapping the input signals into game commands, e.g. mapping the displacement of a stylus on a touch screen to the steering angle of a virtual vehicle

Abstract

A network content processing method and device, an apparatus, and a computer storage medium. The method comprises: determining, according to generation time of elements in historical browsing data, a long-term feature group and a recent feature group from the historical browsing data (S201); determining, according to similarities between elements in the long-term feature group, a long-term encoding vector corresponding to the long-term feature group (S202); determining, according to similarities between elements in the recent feature group and the long-term encoding vector, a user feature vector corresponding to the recent feature group (S203); and determining, according to the user feature vector, a network content to serve as a candidate item to be recommended (S204).

IPC Classes  ?

• G06F 16/9535 - Search customisation based on user profiles and personalisation

Abstract

A text-based event detection method, comprising: acquiring a first data set and a second data set, each of the first data set and the second data set including an event instance, wherein the first data set includes a standard event instance, and the second data set includes a non-standard event instance; and training an adversarial network through the first data set and the second data set, and acquiring a standard event instance in the second data set through the trained adversarial network.

IPC Classes  ?

• G06F 16/33 - Querying
• G06N 3/04 - Architecture, e.g. interconnection topology

Abstract

A data processing method, comprising: acquiring target sequence data, the target sequence data comprising N groups of data that are sorted in chronological order; according to an ith group of data among the N groups of data, a processing result of a target neural network model on the ith group of data, and a processing result of the target neural network model on jth data in an i+1th group of data, using the target neural network model to process j+1th data in the i+1th group of data so as to obtain a processing result of the target neural network model on the j+1th data in the i+1th group of data.

IPC Classes  ?

• G06F 16/483 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content

Abstract

Disclosed are a method and apparatus for detecting an object in an image, and a vehicle and a robot. The method is executed by a computing device, and the method comprises: determining image features of an image; determining a correlation relating to pixels in the image by using the image features; updating the image features of the image by using the correlation so as to obtain updated image features; and determining an object detection result of the image according to the updated image features.

IPC Classes  ?

• G06K 9/46 - Extraction of features or characteristics of the image

Abstract

A method and apparatus for generating mark information in a virtual environment, an electronic device, and a storage medium, relating to the technical field of computers. The method comprises: displaying a target viewing angle picture (201); receiving a first viewing angle switching operation (202); changing an observation direction of the target viewing angle picture and an aiming position of a sight bead according to the first viewing angle switching operation (203); receiving a mark operation triggered on a mark control (204); and generating, on a virtual object to which the aiming position points, mark information according to the mark operation (205).

IPC Classes  ?

• A63F 13/5372 - Controlling the output signals based on the game progress involving additional visual information provided to the game scene, e.g. by overlay to simulate a head-up display [HUD] or displaying a laser sight in a shooting game using indicators, e.g. showing the condition of a game character on screen for tagging characters, objects or locations in the game scene, e.g. displaying a circle under the character controlled by the player

Abstract

Provided are a payment processing method, an electronic device and a computer readable storage medium, which relates to the field of communication technology. The first device obtains the order information and sends it to the second device for display (300); the first device receives the payment code sent by the second device, the payment code is generated by the second device according to the user payment identity after receiving the payment confirmation instruction triggered by the user (310); based on the payment code, the first device performs payment processing on the order information from the payment account corresponding to the payment code to obtain the payment result (320); the first device sends the payment result to the second device for display (330). The first device and the second device are capable of two-way communication, and can send and receive information to each other, which solves the problem that the payment device and the cash register in the prior art cannot realize two-way communication, and is helpful to improve the functional diversity of payment device and enhance user consumption experience.

IPC Classes  ?

• G06Q 20/10 - Payment architectures specially adapted for home banking systems

Abstract

Disclosed are a key management method and a related device. The method comprises: a client selects a random number and a key according to an operation instruction inputted by a user; generate a first encrypted ciphertext of the key according to the random number, the key, a first public key, and a second public key, wherein the first public key is determined according to a point on an elliptic curve and a private key of a hardware security module, and the second public key is determined according to the point on the elliptic curve and a private key of the client; generate a symmetric key sequence according to the key and a preselected Hash function, and according to the symmetric key sequence, encrypt cloud storage data to obtain a data ciphertext; and transmit the first encrypted ciphertext and the data ciphertext to a cloud server.

IPC Classes  ?

• H04L 9/00 - Arrangements for secret or secure communication

Abstract

The present application provides an image segmentation model training method and apparatus, a device, and a storage medium. The method comprises: training an initial image segmentation model by using a source domain sample to obtain a pre-trained image segmentation model; extracting a predicted segmentation result of a source domain image and a predicted segmentation result of a target domain image by means of the pre-trained image segmentation model; training a first discriminator according to the predicted segmentation result of the source domain image and the predicted segmentation result of the target domain image; training a second discriminator according to the predicted segmentation result of the source domain image and a standard segmentation result thereof; and retraining, according to a loss function of the pre-trained image segmentation model, an adversarial loss function of the first discriminator, and an adversarial loss function of the second discriminator, the pre-trained image segmentation model, and performing iterative cyclic training until converging to obtain an image segmentation model completed training.

IPC Classes  ?

• G06T 7/11 - Region-based segmentation
• G06K 9/62 - Methods or arrangements for recognition using electronic means

Abstract

The embodiment of the present application discloses an image acquisition method, device, a server and a storage medium, wherein the method includes: acquiring the target application process corresponding to the user identity from the application process set; when detecting that the image rendering function in the target application process is called, calling the data interception module to intercept a plurality of window image data currently generated; performing image synthesis processing on the plurality of window image data to obtain a user interface image to be displayed; sending the notification message carrying the user interface image to the user terminal corresponding to the user identity, so that the user terminal displays the user interface image on the user interface.

IPC Classes  ?

• H04L 29/06 - Communication control; Communication processing characterised by a protocol

Abstract

The present application relates to an image segmentation method, a model training method, apparatuses, a device and a storage medium. The image segmentation method comprises: acquiring the current frame in a video frame sequence and historical affine transformation information transferred by a preceding video frame; performing affine transformation on the current frame according to the historical affine transformation information to obtain a candidate region image corresponding to the current frame; performing feature extraction on the candidate region image to obtain a feature map corresponding to the candidate region image; performing semantic segmentation on the basis of the feature map to obtain a segmentation result corresponding to a target in the current frame; correcting the historical affine transformation information according to the feature map, to obtain updated affine transformation information, and taking the updated affine transformation information as historical affine transformation information corresponding to the subsequent video frame in the video frame sequence. The solutions provided in the present application improve the accuracy of image segmentation.

IPC Classes  ?

• G06K 9/32 - Aligning or centering of the image pick-up or image-field
• G06K 9/34 - Segmentation of touching or overlapping patterns in the image field
• G06K 9/62 - Methods or arrangements for recognition using electronic means
• G06T 7/00 - Image analysis
• G06T 7/11 - Region-based segmentation

Abstract

A dialogue generation model training method, comprising: acquiring a dialogue data set (201); performing, by means of an encoder in a dialogue generation model, encoding processing on an interrogative sentence in the dialogue data set to obtain an encoded representation of the interrogative sentence (202); fusing, by means of a decoder in the dialogue generation model, the encoded representation of the interrogative sentence and knowledge information corresponding to the interrogative sentence and obtained from a knowledge base question answering model by means of transfer learning, and obtaining a predicted reply corresponding to the interrogative sentence (203); determining, on the basis of the predicted reply corresponding to the interrogative sentence and a marked reply, the value of a loss function of the dialogue generation model (204); and updating, on the basis of the value of the loss function, model parameters of the dialogue generation model (205).

IPC Classes  ?

• G06F 16/332 - Query formulation
• G06N 3/04 - Architecture, e.g. interconnection topology

Abstract

A method for controlling shooting performed with respect to a virtual object comprises: acquiring a locked region of a first virtual object; if a timed duration in which an aiming point of a second virtual object remains in the locked region reaches a target duration, displaying locking complete indication information at a location at which the aiming point is located in the locked region, the locking complete indication information indicating that locking of the first virtual object is complete; and if the aiming point is located within a hit region of the first virtual object, controlling the second virtual object to perform automatic shooting, wherein the size of the hit region is smaller than the size of the locked region.

IPC Classes  ?

• A63F 13/837 - Shooting of targets
• A63F 13/53 - Controlling the output signals based on the game progress involving additional visual information provided to the game scene, e.g. by overlay to simulate a head-up display [HUD] or displaying a laser sight in a shooting game

Abstract

A method can include receiving member information from first member devices at a first gateway UE in a device-to-device (D2D) LAN. The D2D LAN can include a first sub-LAN and at least one second sub-LAN. The first sub-LAN can include the first gateway UE and the first member devices. Each of the second sub-LANs can include a second gateway UE and a set of second member devices. The first and second gateway UEs can be connected with sidelinks. The member information can indicate a first DEVICE ID associated with the respective first member device. A first member device list including the first DEVICE IDs of the first member devices can be updated based on the received member information. A first mapping update message including the updated first member device list and a first gateway ID of the first gateway UE can be broadcast to the second gateway UEs.

IPC Classes  ?

• H04W 8/00 - Network data management
• H04W 92/18 - Interfaces between hierarchically similar devices between terminal devices
• H04L 29/02 - Communication control; Communication processing

Abstract

The present application relates to the technical field of artificial intelligence. Disclosed are a model training method and apparatus, a storage medium, and a device. The method comprises: obtaining a plurality of training samples, each training sample comprising an interactive screen and an action tag which gives an interactive action taken by a character object in the interactive screen; performing feature extraction on the interactive screen comprised in each training sample, and performing clustering on the basis of the extracted features; determining at least one key sample in the plurality of training samples according to the obtained clustering result; and setting a weight for each training sample, and updating network parameters of a deep network on the basis of the training samples having the weights, the weight of each key sample being greater than the weights of other samples in the plurality of training samples.

IPC Classes  ?

• G06N 5/02 - Knowledge representation

Abstract

The present application relates to a video coding method and apparatus, a computer-readable storage medium and a computer device. The method comprises: acquiring video data, wherein the video data comprises multiple original video images; processing, according to a sequence generation rule, the video data to obtain multiple independent subsequence video image groups, wherein each subsequence video image group comprises a subsequence image corresponding to the sequence generation rule; acquiring a coding scheme corresponding to each subsequence video image group; and performing, according to the coding scheme, independent coding on the corresponding subsequence video image group to obtain subsequence video coding data corresponding to each subsequence video image group. According to the solution provided by the present application, the video quality can be adaptively adjusted according to a network condition, so as to improve the video playing fluency. In addition, further provided are a video decoding method and apparatus, a computer-readable storage medium and a computer device.

IPC Classes  ?

• H04N 19/42 - Methods or arrangements for coding, decoding, compressing or decompressing digital video signals - characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation
• H04N 21/234 - Processing of video elementary streams, e.g. splicing of video streams or manipulating MPEG-4 scene graphs

Abstract

A medical image processing method and apparatus, an electronic medical device, and a storage medium, the method comprising: acquiring a biological tissue image comprising biological tissue, identifying a first area in the biological tissue image of a lesion object in the biological tissue, and identifying lesion attributes matching the lesion object (S101); dividing the image areas of the biological tissue in the biological tissue image into a plurality of quadrant position areas (S102); acquiring target quadrant position information of the quadrant position area in which the first area is located and, on the basis of the target quadrant position information and the lesion attributes, generating medical task data (S103).

IPC Classes  ?

• G16H 30/40 - ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing

Abstract

Embodiments of the present invention disclose a data transmission method and apparatus, a device, and a computer storage medium. The method comprises: obtaining, from a target application, an application data packet to be transmitted; transmitting the application data packet to a first routing device over a network data channel, and then transmitting the application data packet to a proxy server by means of the first routing device, while simultaneously transmitting the application data packet to a second routing device over a short-range wireless data channel different from the network data channel, and then transmitting the application data packet to the proxy server by means of the second routing device; and transmitting, by means of the proxy server, the application data packet to an application server on the basis of a deduplication mechanism.

IPC Classes  ?

• H04L 12/28 - Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]

Abstract

An object control method and apparatus, and a storage media and an electronic device. The method comprises: during execution of a game task by a first target object controlled by a client, detecting a target angle generated by the first target object during an action combination (S302), the action combination comprising at least two target actions executed in the same direction; in the case that the target angle reaches a trigger threshold, triggering a state adjustment instruction (S304); and in response to the state adjustment instruction, adjusting the movement state of the first target object from a first state to a second state (S306), wherein the time duration taken by the first target object to complete the game task in the first state is longer than the time duration taken by the first target object to complete the game task in the second state. The method solves the technical problem of increasing the control duration due to high operation complexity.

IPC Classes  ?

• A63F 13/2145 - Input arrangements for video game devices characterised by their sensors, purposes or types for locating contacts on a surface, e.g. floor mats or touch pads the surface being also a display device, e.g. touch screens

Abstract

Disclosed by the present application are an illumination rendering method and apparatus, a storage medium, and an electronic apparatus. The method comprises: acquiring a first picture at a target angle of view in a virtual three-dimensional scene, the first picture comprising a virtual object to undergo illumination rendering in the virtual three-dimensional scene at the target angle of view; determining a target virtual light source point set for performing illumination rendering on the virtual object in the first picture; and according to the target virtual light source point set, performing illumination rendering on the virtual object in the first picture. The present application solves the technical problem in related technology in which overall illumination rendering efficiency is low for a virtual object in a virtual three-dimensional scene.

IPC Classes  ?

• G06T 15/50 - Lighting effects

Abstract

A method and apparatus for detecting a target in a video, and a vehicle, a robot, a monitoring device, a computing device and a storage medium. The method is executed by a computing device, and comprises: applying a target detection operation to a first frame to determine a first target detection result, in the first frame, of a target (S202); applying a target tracking operation to a second frame located after the first frame to determine a change, between the second frame and the first frame, in the target (S204); and determining a second target detection result, in the second frame, of the target according to the first target detection result and the change, between the second frame and the first frame, in the target (S206).

IPC Classes  ?

• G06T 7/238 - Analysis of motion using block-matching using non-full search, e.g. three-step search

Abstract

A personalized face display method for a three-dimensional character, which is implemented by a computer device. The method comprises: acquiring a texture identifier of a personalized face model of a three-dimensional character, wherein the texture identifier is used for identifying a target style texture of a target part on the personalized face model, the target part is provided with a corresponding UV texture image, and the UV texture image comprises at least two candidate style textures of the target part; drawing a basic UV texture image on the personalized face model, wherein the basic UV texture image comprises textures of regions other than the target part; determining, according to the texture identifier, the target style texture from the at least two candidate style textures corresponding to the target part; and drawing the target style texture in a region, corresponding to the target part, on the personalized face model.

IPC Classes  ?

• G06T 15/04 - Texture mapping

Abstract

An image processing method and apparatus, a computer device and a storage medium. Said method comprises: acquiring an image to be processed, the image to be processed being fused with a content image to be promoted (101); performing context feature extraction on the image to be processed to obtain context feature information of the image to be processed (102); extracting, on the basis of the context feature information, multi-layer semantic information of an object in the image to be processed (103); and performing image reconstruction on the basis of the context feature information and the multi-layer semantic information, so as to obtain a reconstructed image (104). Said method can improve the image fusion effect.

IPC Classes  ?

• G06T 5/50 - Image enhancement or restoration by the use of more than one image, e.g. averaging, subtraction

Abstract

The present application discloses an application starting method and apparatus, a computer device and a storage medium, belonging to the field of computer technology. Said method comprises: displaying an application starting management interface, the application starting management interface including a starting control of at least one application; acquiring a first trigger signal, the first trigger signal being a signal generated when a trigger operation on a starting control of a target application is received; when a terminal has not finished downloading a client of the target application, acquiring an application video stream, the application video stream being a video stream composed of application pictures generated when a first client runs, and the first client being the client of the target application running in a server; and displaying a first application interface of the target application according to the application video stream. The server above may be a cloud server.

IPC Classes  ?

• G06F 8/60 - Software deployment

Abstract

An object jumping control method and apparatus, and a computer device and a storage medium, relating to the technical field of networks. The method comprises: in a process that a first virtual object moves to a destination in a virtual scene, when the current position of the first virtual object satisfies a jumping condition, determining a target landing point of the first virtual object according to the position of the destination; determining, according to the current position and the target landing point, a jumping trajectory that the first virtual object jumps to the target landing point from the current position; and controlling, according to the jumping trajectory, the first virtual object to jump to the target landing point from the current position.

IPC Classes  ?

• A63F 13/55 - Controlling game characters or game objects based on the game progress
• A63F 13/426 - Processing input control signals of video game devices, e.g. signals generated by the player or derived from the environment by mapping the input signals into game commands, e.g. mapping the displacement of a stylus on a touch screen to the steering angle of a virtual vehicle involving on-screen location information, e.g. screen coordinates of an area at which the player is aiming with a light gun

Abstract

A quantum noise process analysis method and apparatus, a device, and a storage medium, relating to the technical field of quanta. The method comprises: performing quantum process tomography on a quantum noise process of a target quantum system to obtain dynamical mapping of the quantum noise process (201); extracting tensor transfer mapping of the quantum noise process from the dynamical mapping (202); and analyzing the quantum noise process according to the tensor transfer mapping (203). The tensor transfer mapping is used for representing the dynamical evolution of the quantum noise process, i.e., embodying the evolution law of the dynamical mapping of the quantum noise process over time. Therefore, a quantum noise process is analyzed on the basis of tensor transfer mapping of the quantum noise process. Compared with pure quantum process tomography, richer and more comprehensive information about the quantum noise process can be obtained, so that the quantum noise process is analyzed more accurately and comprehensively.

IPC Classes  ?

• G06F 17/10 - Complex mathematical operations

Abstract

The present application relates to the field of mobile communications, and in particular, to a method, device, and system for selecting a mobile edge computing node. The method comprises: an edge cloud gateway receives a domain name request forwarded by a user plane function, wherein in the domain name request, the domain name is a service acceleration domain name or a destination address is an IP address of a global server load balancer, or protocol port information in the domain name request satisfies a protocol port requirement; obtain a corresponding edge application VIP from the global server load balancer and send same to a terminal by means of the user plane function; receive a service request sent by the terminal and forwarded by the user plane function; shunt the service request to a corresponding MED processing server. The present application newly adds an edge cloud gateway, is compatible with both DNS and HTTP DNS addressing scenes, and achieves the purpose of selecting a nearest edge computing node to implement edge acceleration.

IPC Classes  ?

• H04L 29/12 - Arrangements, apparatus, circuits or systems, not covered by a single one of groups characterised by the data terminal
• H04L 29/08 - Transmission control procedure, e.g. data link level control procedure

Abstract

The present application relates to the field of artificial intelligence, and disclosed are a content recommendation method and apparatus, and a device and a storage medium. The method comprises: obtaining a target user vector of a target user; determining n groups of seed user vectors according to the target user vector, the n groups of seed user vectors corresponding to n candidate recommendation contents; calling a similar crowd diffusion model to calculate a similarity between the target user vector and each group of seed user vectors, the similar crowd diffusion model being used for calculating a similarity between user vectors on the basis of an attention mechanism; and determining a target content recommended to the target user among the n candidate recommendation contents according to the respective similarities of the n groups of seed user vectors. According to the present application, the problem of low accuracy of a recommendation method in the related art can be solved.

IPC Classes  ?

• G06F 16/9535 - Search customisation based on user profiles and personalisation

Abstract

Provided are a place recognition method and apparatus, a model training method and apparatus for place recognition, a computer readable storage medium and an electronic device. The model training method comprises: extracting local features of a sample image on the basis of a first portion of a CNN model (310); aggregating the local features into feature vectors having a first number of dimensions on the basis of a second portion of the CNN model (330); obtaining, on the basis of a third portion of the CNN model, compressed representation vectors of the feature vectors, wherein the compressed representation vectors have a second number of dimensions that is less than the first number of dimensions (350); and adjusting model parameters of the first to third portions with the aim of minimizing the distance between the compressed representation vectors corresponding to a plurality of images photographed at the same place (370). According to the method, a compression process with trainable parameters is introduced into a CNN model, so that an end-to-end training place recognition model can be fully realized, and the obtained CNN model can directly obtain low-dimensional image features, thereby improving the place recognition performance on the basis of artificial intelligence.

IPC Classes  ?

• G06F 16/58 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• G06N 3/04 - Architecture, e.g. interconnection topology

Abstract

Disclosed are a background audio signal filtering method and apparatus, and a storage medium, belonging to the technical field of audio processing. The method comprises: acquiring a first audio signal collected during the process of playing a background audio signal, wherein the background audio signal is an audio signal obtained after watermark information is added to an original audio signal; performing a separation operation on the first audio signal to obtain the watermark information and a second audio signal excluding the watermark information; querying a preset correlation according to the watermark information to obtain the original audio signal corresponding to the watermark information; and filtering the original audio signal out from the second audio signal to obtain a target audio signal. The embodiments of the present application provide a solution of filtering a background audio signal. There is no need to acquire an additional individual background audio signal, and the background audio signal can be filtered out from the collected audio signal, such that the influence of the background audio signal is avoided, the universality is strong, and the application range is expanded.

IPC Classes  ?

• G10L 19/018 - Audio watermarking, i.e. embedding inaudible data in the audio signal
• G10L 21/0208 - Noise filtering
• G10L 21/0272 - Voice signal separating

Abstract

Disclosed is a mammogram image processing method, which is executed by a computer device. The method comprises: acquiring mammogram images of a unilateral breast, the mammogram images comprising a CC position mammogram image and an MLO position mammogram image; calling a target detection model to process the CC position mammogram image and the MLO position mammogram image so as to obtain an image detection result of the unilateral breast; and generating a detection report according to the image detection result and outputting same.

IPC Classes  ?

• G16H 50/20 - ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Provided are a method, apparatus, electronic device, and computer storage medium for optimizing a quantization model, relating to the technical field of deep learning. The method comprises: determining the jump ratios of parameters of the embedding layer of a trained quantization model within a predetermined time range, the quantization model being a neural network model in which the parameters of the embedding layer have been quantized (S110); then, according to each jump ratio, determining a jump curve within a predetermined time range, and fitting the jump curve to obtain a corresponding temporal scaling parameter (S120); then, optimizing an original optimization algorithm of the quantization model on the basis of the temporal scaling parameter to obtain an optimized target optimization algorithm, and training the quantization model on the basis of the target optimization algorithm (S130). When optimizing the original optimization algorithm, the time parameters are fully considered, enabling rapid learning of new data features, improving online prediction ability and recommendation accuracy.

IPC Classes  ?

• G06F 16/9535 - Search customisation based on user profiles and personalisation
• G06N 3/04 - Architecture, e.g. interconnection topology

Abstract

Disclosed in embodiments of the present application are a model training method and apparatus, and a terminal and a storage medium. The method comprises: obtaining a template image and a test image; calling a first object recognition model to process the features of a tracking object in the template image to obtain a first reference response, and calling a second object recognition model to process the features of the tracking object in the template image to obtain a second reference response; calling the first object recognition model to process the features of the tracking object in the test image to obtain a first test response, and calling the second object recognition model to process the features of the tracking object in the test image to obtain a second test response; tracking the first test response to obtain a tracking response of the tracked object; and updating the first object recognition model based on difference information between the first reference response and the second reference response, difference information between the first test response and the second test response, and difference information between a tracking tag and a tracking response.

IPC Classes  ?

• G06K 9/62 - Methods or arrangements for recognition using electronic means

Abstract

The present application relates to the field of mobile communication, and in particular relates to a method, apparatus, and system for selecting a mobile edge computing node, the method comprising: an edge cloud gateway receiving a first HTTP service request forwarded by a user plane function, the target address of the first HTTP service request being an edge application VIP; on the basis of the first HTTP service request and a preset shunting policy, determining a corresponding MEC processing server, and shunting the first HTTP service request to the corresponding MEC processing server. As such, by means of a newly added edge cloud gateway, the problem in which there is still no MEC node selection scheme for 5G mobile communication networks is solved, and an edge computing node may be selected nearby, thus attaining the goal of edge acceleration.

IPC Classes  ?

• H04L 29/06 - Communication control; Communication processing characterised by a protocol

Abstract

A vision-based tactile measurement method; the method is executed by a chip, and the chip is connected to a tactile sensor; the tactile sensor comprises a sensing surface and an image sensing assembly, and the sensing surface is provided with a marking pattern; the method comprises: acquiring an image sequence collected by the image sensing assembly for the sensing surface, images in the image sequence comprising marking patterns; according to the marking patterns within adjacent images in the image sequence, calculating a difference feature of the marking patterns; and calling a feedforward neural network to process the difference feature of the marking patterns to obtain a tactile measurement result.

IPC Classes  ?

• G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer

Abstract

Provided are an image occlusion processing method, a device, an apparatus and acomputer storage medium, wherein the method includes: determining a current viewpoint parameter for drawing a current frame image; acquiring a predicted depth map matching the current viewpoint parameter as a target depth map of the current frame image; and determining an occlusion culling result of an object in the current frame image according to the target depth map.

IPC Classes  ?

• G06T 5/00 - Image enhancement or restoration

Abstract

The present application relates to the technical field of image processing, and disclosed are a character recognition method and apparatus, a computer device and a storage medium. The technical solutions of the embodiments of the present application may be used to extract a target number of characters from an image to be recognized. By means of extracting an image feature of an image, the image feature comprising a plurality of image feature vectors, an attention mechanism is employed; and according to the plurality of image feature vectors, attention weight values corresponding to the target number of characters are calculated and outputted by means of parallel calculation. One attention weight value may represent the degree of importance of the plurality of image feature vectors for a character corresponding to an attention weight value. As a result, during the process of character recognition, by means of the foregoing attention mechanism-based recognition, characters of any shape may be effectively recognized by means of a simple process, thereby avoiding the process of cyclic calculation and greatly improving the calculation efficiency.

IPC Classes  ?

• G06K 9/00 - Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints

Abstract

An embodiment of the present invention discloses an image transformation method and device, a storage medium and a computer equipment; the method comprises the steps of acquiring identification information of an original image, converting the identification information into an image to be added, performing frequency domain transformation on the original image to obtain a pixel matrix of the original image in the frequency domain space, performing matrix decomposition on the pixel matrix to obtain an image brightness matrix, converting the pixel values of the corresponding pixels in the image brightness matrix based on the pixel values of the pixels in the image to be added to obtain a converted brightness matrix, and performing inverse frequency domain transformation on the converted brightness matrix to obtain a transformed image. The solution can implement the addition of invisible identification information in the original image.

IPC Classes  ?

• G06T 1/00 - General purpose image data processing
• G06T 5/10 - Image enhancement or restoration by non-spatial domain filtering

Abstract

The present application provides a target tracking method, comprising: obtaining a plurality of consecutive picture frames of a target video; setting a tracking target area of the nth picture frame in the plurality of consecutive picture frames; performing a feature extraction operation on the tracking target area of the nth picture frame to obtain an input feature map of the tracking target area; calculating a weighting filter corresponding to the input feature map of the tracking target area according to a correlation filtering algorithm and an average pooling constraint condition; calculating, by using the weighting filter and the input feature map of the (n+1)th picture frame in the plurality of consecutive picture frames, output response of the (n+1)th picture frame, and confirming the tracking target area of the (n+1)th picture frame according to the output response of the (n+1)th picture frame; and performing operations until the tracking target areas of all the consecutive picture frames are obtained. The present application further provides a target tracking apparatus. The present application improves the precision and effectiveness of target tracking.

IPC Classes  ?

• G06T 7/246 - Analysis of motion using feature-based methods, e.g. the tracking of corners or segments

Abstract

A fence address-based coordinate data processing method and apparatus, a computer readable storage medium, and a computer device. The method comprises: acquiring an address text and a corresponding fence identifier (S202); performing semantic extension on the address text to obtain a plurality of extended addresses (S204); converting each of the extended addresses into corresponding geographic coordinates (S206); recognizing key address elements in the plurality of extended addresses (S208); screening the geographic coordinates according to whether the extended address corresponding to any geographic coordinates comprises the recognized key address elements and whether the geographic coordinates are located in a target fence corresponding to the fence identifier (S210); and determining, according to clustering features of the geographic coordinates obtained by screening, a geographic coordinate obtained by screening as a processed coordinate (S212). The present method can improve the efficiency and accuracy of coordinate processing.

IPC Classes  ?

• G06F 16/29 - Geographical information databases