WO2021082460A1 - 一种ai处理方法及装置 - Google Patents

一种ai处理方法及装置 Download PDF

Info

Publication number
WO2021082460A1
WO2021082460A1 PCT/CN2020/095570 CN2020095570W WO2021082460A1 WO 2021082460 A1 WO2021082460 A1 WO 2021082460A1 CN 2020095570 W CN2020095570 W CN 2020095570W WO 2021082460 A1 WO2021082460 A1 WO 2021082460A1
Authority
WO
WIPO (PCT)
Prior art keywords
processing
computing device
characteristic information
cloud service
data stream
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2020/095570
Other languages
English (en)
French (fr)
Inventor
刘洪宽
惠卫锋
蔡世顺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP20882057.1A priority Critical patent/EP4016396A4/en
Publication of WO2021082460A1 publication Critical patent/WO2021082460A1/zh
Priority to US17/677,733 priority patent/US20220179713A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5055—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering software capabilities, i.e. software resources associated or available to the machine
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/06—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons
    • G06N3/063—Physical realisation, i.e. hardware implementation of neural networks, neurons or parts of neurons using electronic means
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061—Partitioning or combining of resources
    • G06F9/5072—Grid computing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/2866—Architectures; Arrangements
    • H04L67/30—Profiles
    • H04L67/306—User profiles
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/535—Tracking the activity of the user
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G06F2009/45562—Creating, deleting, cloning virtual machine instances
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44—Arrangements for executing specific programs
    • G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533—Hypervisors; Virtual machine monitors
    • G06F9/45558—Hypervisor-specific management and integration aspects
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/10—Interfaces, programming languages or software development kits, e.g. for simulating neural networks
    • G06N3/105—Shells for specifying net layout

Definitions

  • This application relates to the field of artificial intelligence (AI), and in particular to an AI processing method and device.
  • AI artificial intelligence
  • AI is a new technological science that studies and develops theories, methods, technologies and application systems used to simulate, extend and expand human intelligence.
  • AI technology can be used in smart electronic devices (such as smart phones, smart TVs, etc.) Perform AI processing to simulate human intelligence in different scenarios. For example, using AI technology can realize mobile phone recognition software, image search software, smart TV recognition software, etc., and then provide intelligent services based on the recognition results to improve intelligent user experience.
  • the AI processing process includes two parts: training and inference.
  • Training uses massive data as input to train a deep neural network model; inference refers to using a trained deep neural network model to input the data to be judged into the model to "infer"
  • This conclusion as a result of AI processing.
  • AI processing can identify people, objects, or scenes in the data.
  • the content of AI processing results is different in different scenarios.
  • the AI processing result can be the label information (such as physical name, attributes, etc.) of the recognized object.
  • advertising push software the AI processing result can be advertising push content.
  • AI processing For example, the process of identifying objects with pictures in a smart phone.
  • a smart phone generates pictures (which can be from a mobile phone camera or pictures downloaded from other channels).
  • the smart phone sends the pictures to the server through the network, and the server performs AI processing and the processed AI
  • the result is sent to the smartphone for display.
  • the AI processing process is a typical request-response type.
  • the smartphone is in a state of waiting for results and cannot perform real-time AI processing on the data stream.
  • Another type of AI processing is that the electronic device itself performs AI processing, and the electronic device has AI processing capability and can perform AI processing in real time. Although this solution can process data streams in real time, it has high requirements on electronic equipment.
  • Another AI processing process is that in a scenario where an intelligent edge station is deployed between a server and an electronic device, the intelligent edge station performs AI processing.
  • the smart camera synchronizes the image data stream to the smart edge station in real time, and the smart edge station performs AI processing, and the AI processing result is directly displayed to the user or returned to the camera for display.
  • this solution handles real-time AI processing of data streams, it is only suitable for real-time monitoring scenarios.
  • Edge smart nodes need to be deployed, which is costly and difficult to deploy on a large scale.
  • This application provides an AI processing method and device to realize real-time data stream AI processing that is low-cost, easy to deploy, and does not require equipment capabilities.
  • an AI processing method which is applied to a computing device, and the method may include: the computing device obtains characteristic information of a user of the computing device, the characteristic information is used to indicate the user's behavior habits and/or interests Phantomies; the computing device sends the acquired user's characteristic information to the cloud service; the computing device receives the AI processing result from the cloud service, and the AI processing result is that the cloud service performs AI on the data stream of the computing device according to the characteristic information of the user of the computing device Handle it.
  • the cloud service by configuring the cloud service, when there is a data stream sent to the computing device, the cloud service performs AI processing on the data stream in real time, and the computing device receives the AI processing result from the cloud service in real time.
  • This process Using the existing architecture, the real-time AI processing of the data stream is performed on the server without increasing the cost, simple deployment, and no requirement for computing device capabilities.
  • the cloud service that interacts with the computing device can be mapped to the computing device.
  • the programs in the computing device for example, application (APP)
  • APP application
  • the distribution server requests a data stream.
  • the AI processing method provided by this application may further include: before sending the user's characteristic information to the cloud service, the computing device sends a registration request to the cloud server, and the registration request is used for Cloud server configures cloud services.
  • the registration request may be used by the cloud server to configure the cloud service corresponding to the computing device that sent the registration request.
  • the cloud server and the content distribution server that sends the data stream may be the same server, or the cloud server and the content distribution server that send the data stream may be the same server.
  • the content distribution server of is a different server.
  • the foregoing cloud service may be a virtual device configured according to virtual machine technology or container technology to implement cloud services corresponding to different terminal devices Isolated from each other.
  • the AI processing method provided in this application may further include: before receiving the AI processing result from the cloud service, the computing device obtains the user's Configuration parameters, sending configuration parameters to the cloud service.
  • the configuration parameters can include one or more of the following: AI function on or off, person recognition function on or off, item recognition function on or off, character mental activity recognition function on or off, character emotion recognition function on or off, or Turn off, or turn on or turn off the scene recognition function.
  • the computing device receives the AI processing result from the cloud service, which can be specifically implemented as: the computing device receives the AI processing result from the cloud service, where the AI processing result is the cloud service's data on the computing device according to the characteristic information and configuration parameters The stream is processed by AI.
  • the characteristic information of the user may include: user behavior information, and/or user interest information.
  • the user behavior element information is used to indicate the characteristics of the user's use of the computing device.
  • the user behavior information may include one or more of the following information: loyal product brand information, commonly used APP information, online duration information; user interest information To indicate one or more types of interests of the user.
  • the AI processing result may include one or more of the following: advertisement placement, item labeling, and commentary.
  • the data stream may include one or more of the following content: video stream, picture stream, and text stream.
  • the cloud service described in this application may be a cloud server corresponding to the computing device.
  • the communication between computing device and cloud service described in this article (for example, computing device sends certain content to cloud service, or computing device receives certain content from cloud service, or cloud service sends certain content to computing device, or cloud service receives certain content from computing device Certain content) can be replaced by: the computing device communicates with its corresponding cloud service.
  • the cloud server may configure a cloud service corresponding to each computing device.
  • cloud services corresponding to different computing devices can be isolated from each other.
  • another AI processing method is provided, which can be applied to a cloud server in which cloud services are configured.
  • the method may include: the cloud server receives the characteristic information of the user of the computing device from the computing device; wherein the characteristic information is used to indicate the behavior and/or hobbies of the user; the cloud server performs AI processing on the data stream according to the characteristic information of the user Obtain the AI processing result; the cloud server sends the AI processing result to the computing device.
  • the cloud service by configuring the cloud service, when there is a data stream sent to the computing device, the cloud service performs AI processing on the data stream in real time, and the computing device receives the AI processing result from the cloud service in real time.
  • This process Using the existing architecture, the real-time AI processing of the data stream is performed on the server without increasing the cost, simple deployment, and no requirement for computing device capabilities.
  • the above-mentioned data stream may refer to the data stream of the computing device, and can be understood as the data stream sent to the computing device.
  • the aforementioned cloud server may be a content distribution server.
  • the cloud server and the content distribution server that provides the data stream are different servers.
  • the AI processing method provided in this application may further include: before performing AI processing on the data stream according to the characteristic information, the cloud server obtains the data stream of the computing device from the content distribution server.
  • the cloud server obtains the data stream of the computing device from the content distribution server, which can be specifically implemented as follows: request the computing device from the content distribution server The data stream of the computing device is received from the content distribution server.
  • the AI processing method provided in this application may further include: a cloud server from The content distribution server receives the data stream.
  • the cloud service may be a virtual device configured according to virtual machine technology or container technology.
  • the AI processing method provided by this application may further include: performing AI processing on the data stream according to the characteristic information to obtain the AI processing result
  • the cloud server received configuration parameters from the computing device.
  • the configuration parameters may include one or more of the following: AI function on or off, person recognition function on or off, item recognition function on or off, character mental activity recognition The function is turned on or off, the character recognition function is turned on or off, or the scene recognition function is turned on or off.
  • the cloud server performs AI processing on the data stream according to the characteristic information to obtain the AI processing result, which can be specifically implemented as follows: performing AI processing on the data stream according to the characteristic information and configuration parameters to obtain the AI processing result.
  • the characteristic information of the user may include: user behavior information, and/or user interest information.
  • the user behavior element information is used to indicate the characteristics of the user's use of the computing device.
  • the user behavior information may include one or more of the following information: loyal product brand information, commonly used APP information, online duration information; user interest information To indicate one or more types of interests of the user.
  • the AI processing result may include one or more of the following: advertising, item labeling, and commentary.
  • the data stream may include one or more of the following content: video stream, picture stream, and text stream.
  • AI processing method provided in the second aspect of the present application and the AI processing method provided in the first aspect are descriptions of the same technical solution from different perspectives, and their specific implementations can be referred to each other.
  • the present application provides an AI processing device.
  • the AI processing device may be a computing device, a device or a chip system in a computing device, or a device that can be matched and used with a computing device.
  • the AI processing apparatus can realize the functions performed by the computing devices in the above-mentioned aspects or various possible designs, and the functions can be realized by hardware, or by hardware executing corresponding software.
  • the hardware or software includes one or more modules corresponding to the above-mentioned functions.
  • the AI processing device may include: an acquiring unit, a sending unit, and a receiving unit.
  • the acquiring unit is used to acquire characteristic information of the user of the AI processing device, and the characteristic information is used to indicate the user's behavior habits and/or hobbies.
  • the sending unit is used to send the characteristic information acquired by the acquiring unit to the cloud service.
  • the receiving unit is used to receive the AI processing result from the cloud service.
  • the AI processing result is obtained by the cloud service performing AI processing on the data stream of the AI processing device according to the characteristic information.
  • AI processing device provided in the third aspect is used to execute the AI processing method provided in the first aspect, and the specific implementation can refer to the specific implementation in the first aspect.
  • the present application provides an AI processing device.
  • the AI processing device may be a cloud server, a device or a chip system in a cloud server, or a device that can be matched and used with a cloud server.
  • the cloud server is configured with cloud services.
  • the AI processing device can implement the functions performed by the cloud server in the above aspects or various possible designs, and the functions can be implemented by hardware, or by hardware executing corresponding software.
  • the hardware or software includes one or more modules corresponding to the above-mentioned functions.
  • the AI processing device may include: a receiving unit, a processing unit, and a sending unit.
  • the receiving unit is configured to receive the characteristic information of the user of the computing device from the computing device.
  • the characteristic information is used to indicate the user's behavior habits and/or hobbies.
  • the processing unit is configured to perform AI processing on the data stream according to the characteristic information received by the receiving unit to obtain an AI processing result.
  • the sending unit is used to send the AI processing result obtained by the processing unit to the computing device.
  • AI processing device provided in the fourth aspect is used to execute the AI processing method provided in the second aspect, and the specific implementation may refer to the specific implementation in the second aspect.
  • the present application provides a computing device.
  • the computing device may include: a processor, a memory; the processor, and the memory are coupled, the memory may be used to store computer instructions, and the processor is used to call the computer instructions to perform operations such as the first aspect or The AI processing method described in any possible implementation manner.
  • the present application provides a cloud server.
  • the cloud server may include: a processor, a memory; the processor, the memory is coupled, the memory may be used to store computer instructions, and the processor is used to call the computer instructions to execute the second aspect or The AI processing method described in any possible implementation manner.
  • the present application provides a computer-readable storage medium.
  • the computer-readable storage medium may include: computer software instructions; when the computer software instructions run in a computing device, the computing device executes operations such as the first aspect or the first aspect.
  • the present application provides a computer-readable storage medium.
  • the computer-readable storage medium may include: computer software instructions; when the computer software instructions run in a cloud server, the cloud server executes operations such as the second aspect or the first aspect.
  • this application provides a computer program product, which when the computer program product runs on a computer, causes the computer to execute the AI according to the first aspect of the claim or any one of the possible implementation manners Approach.
  • the present application provides a chip system that is applied to a computing device; the computing device includes an interface circuit and a processor; the interface circuit and the processor are interconnected by wires; the interface circuit is used to receive signals from the memory of the computing device, And send a signal to the processor, the signal includes a computer instruction stored in the memory; when the processor executes the computer instruction, the chip system executes the AI processing method as described in the first aspect or any one of the possible implementation manners .
  • the present application provides a chip system that is applied to a cloud server;
  • the cloud server includes an interface circuit and a processor;
  • the interface circuit and the processor are interconnected through a wire;
  • the interface circuit is used to receive signals from the memory of the cloud server , And send a signal to the processor, the signal includes a computer instruction stored in the memory; when the processor executes the computer instruction, the chip system executes the AI processing as described in the second aspect or any one of the possible implementation manners method.
  • this application provides an AI processing system, including a computing device and a cloud server.
  • the computing device executes the AI processing method as described in the first aspect or any one of the possible implementations of the first aspect
  • the cloud server executes the AI processing method as described in any one of the second aspect or the possible implementation of the second aspect AI processing method.
  • this application provides another AI processing system, including computing devices and cloud servers, in which cloud services are configured.
  • the computing device is used to: obtain the characteristic information of the user of the computing device, the characteristic information is used to indicate the behavior and/or hobbies of the user; and to send the acquired characteristic information of the user to the cloud service.
  • the cloud server is used to: receive the characteristic information of the user of the computing device from the computing device; wherein the characteristic information is used to indicate the user's behavior habits and/or hobbies; according to the characteristic information of the user, perform AI processing on the data stream of the computing device Obtain the AI processing result; send the AI processing result to the computing device.
  • the computing device is further configured to: receive an AI processing result from the cloud service, where the AI processing result is obtained by the cloud service performing AI processing on the data stream of the computing device according to the characteristic information of the user of the computing device.
  • Figure 1 is a schematic diagram of an AI processing scenario in the prior art
  • Figure 1a is a schematic diagram of another AI processing scenario in the prior art
  • Figure 2 is a schematic diagram of the architecture of a data service system provided by the prior art
  • FIG. 3 is a schematic diagram of the architecture of a data service system provided by the prior art
  • FIG. 4 is a schematic structural diagram of an AI processing device provided by an embodiment of the application.
  • FIG. 5 is a schematic flowchart of an AI processing method provided by an embodiment of this application.
  • FIG. 6 is a schematic flowchart of another AI processing method provided by an embodiment of the application.
  • FIG. 7 is a schematic structural diagram of another AI processing device provided by an embodiment of the application.
  • FIG. 8 is a schematic structural diagram of yet another AI processing device provided by an embodiment of this application.
  • FIG. 9 is a schematic structural diagram of yet another AI processing device provided by an embodiment of this application.
  • FIG. 10 is a schematic structural diagram of yet another AI processing device provided by an embodiment of this application.
  • words such as “exemplary” or “for example” are used as examples, illustrations, or illustrations. Any embodiment or design solution described as “exemplary” or “for example” in the embodiments of the present application should not be construed as being more preferable or advantageous than other embodiments or design solutions. To be precise, words such as “exemplary” or “for example” are used to present related concepts in a specific manner.
  • a data stream can be a collection of data sent to a computing device.
  • the data stream can include but is not limited to one or more of the following: video stream, picture stream, text stream, etc.
  • the content distribution server may be a server that provides data streams to computing devices.
  • the data stream may be a business data stream or others, which is not specifically limited in the embodiment of the present application.
  • the content distribution server may provide a data stream to the computing device according to the rule or according to the request of the computing device.
  • the cloud server may be a server deployed in the cloud to implement centralized services.
  • the computing device can be the device on demand for the data stream.
  • the computing device can display or perform corresponding processing after acquiring the data stream, or the computing device displays the data stream through other devices after acquiring the data stream, or the computing device displays the data stream through other devices after corresponding processing after acquiring the data stream.
  • This application is not limited.
  • the computing device may be a server or a terminal device, and this application does not limit the actual product form of the computing device.
  • AI processing can include training and inference.
  • Training uses massive data input to train a network model (such as a complex deep neural network model).
  • Inference refers to using the trained network model to "infer” various conclusions using the data to be judged, and use the conclusion as the result of AI processing.
  • the conclusion drawn by the inference is that the content is different in different scenarios, such as object recognition software.
  • the processed result is the label information of the recognized object, such as the physical name, attributes, and so on.
  • the computing device can request a data stream from the content distribution server.
  • AI processing is performed according to the aforementioned request-response style. The process takes a long time and cannot be implemented on the server side. Real-time AI processing.
  • the embodiment of the application provides an AI processing method.
  • the program and operation of the computing device are mapped in the cloud service, and the AI processing requirements of the computing device are mapped to the cloud service, and the cloud service uses the server's AI processing capability, real-time AI processing of the data stream of the computing device, realizing the real-time AI processing of the data stream on the server under the premise that the computing device does not have the AI processing capability.
  • the AI processing method provided by the embodiment of the present application can be applied to the data service system shown in FIG. 3.
  • the data service system may include a content distribution server 301, a cloud server 302, and a computing device 303.
  • a cloud service 3021 is configured in the cloud server 302.
  • the computing device 303 requests a data stream from the content distribution server 301 through the cloud service 3021.
  • the content distribution server 301 sends a data stream to the computing device 303 through the cloud service 3021.
  • the cloud service 3021 when the cloud service 3021 receives a data stream (also referred to as a data stream of the computing device 303) sent to the computing device 303 from the content distribution server 301, it can perform AI processing on the data stream.
  • the data stream and the AI processing result of the data stream are sent to the computing device 303.
  • the cloud service 3021 may perform AI processing on the data stream of the computing device 303 according to the configuration of the computing device 303, and send the data stream and the AI processing result of the data stream to the computing device 303.
  • the computing device 303 may display the AI processing result, or the computing device 303 may send the AI processing result to other devices for display, which is not limited in this embodiment of the application.
  • the content distribution server 301 may be a server of a content provider.
  • the cloud server 302 may be a server that provides cloud services.
  • the cloud server 302 is the content distribution server 301.
  • the cloud server 302 and the content distribution server 301 are deployed independently, and the two communicate through a hypertext transport protocol (HTTP) protocol.
  • HTTP hypertext transport protocol
  • the computing device 303 may be a server or a terminal device or others, and the embodiment of the present application does not limit the specific product form of the computing device 303.
  • the terminal device involved in the embodiments of this application may be a device with wireless transceiver function, which can be deployed on land, including indoor or outdoor, handheld or vehicle-mounted; or on the water (such as ships, etc.); Can be deployed in the air (for example, airplanes, balloons, satellites, etc.).
  • the terminal device may be a user equipment (UE), and the UE includes a handheld device with a wireless communication function, a vehicle-mounted device, a wearable device, or others.
  • the UE may be a mobile phone, a tablet computer, or a computer with wireless transceiver function.
  • Terminal equipment can also be virtual reality (VR) terminal equipment, augmented reality (AR) terminal equipment, wireless terminals in industrial control, wireless terminals in unmanned driving, wireless terminals in telemedicine, and smart Wireless terminals in power grids, wireless terminals in smart cities, wireless terminals in smart homes, and so on.
  • VR virtual reality
  • AR augmented reality
  • Wireless terminals in power grids wireless terminals in smart cities, wireless terminals in smart homes, and so on.
  • the architecture of the data service system shown in FIG. 3 is only an exemplary architecture diagram, and the embodiment of the present application does not limit the number of network elements included in the architecture shown in FIG. 3.
  • the system shown in FIG. 3 may also include other functional entities.
  • the naming of the network elements in the above-mentioned FIG. 3 architecture is only an example, and is not limited.
  • an embodiment of the present application provides an AI processing device for executing the AI processing method provided in this application.
  • the AI processing device may be deployed in a computing device or a cloud server in the data service system shown in FIG. 3.
  • FIG. 4 shows an AI processing device 40 related to various embodiments of the present application.
  • the AI processing device 40 may include a processor 401, a memory 402, and a transceiver 403.
  • the memory 402 may be a volatile memory (volatile memory), such as a random-access memory (random-access memory, RAM); or a non-volatile memory (non-volatile memory), such as a read-only memory (read-only memory).
  • volatile memory such as a random-access memory (random-access memory, RAM); or a non-volatile memory (non-volatile memory), such as a read-only memory (read-only memory).
  • memory ROM), flash memory (flash memory), hard disk (HDD) or solid-state drive (solid-state drive, SSD); or a combination of the above-mentioned types of memory, used to store materials that can implement the method of this application Program code, configuration files or other content.
  • the processor 401 is the control center of the AI processing device 40.
  • the processor 401 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
  • Circuits for example: one or more microprocessors (digital signal processors, DSP), or one or more field programmable gate arrays (FPGA).
  • the transceiver 403 is used to communicate with other devices.
  • the processor 401 executes by running or executing software programs and/or modules stored in the memory 402, and calling data stored in the memory 402.
  • Acquire characteristic information of the user of the computing device which is used to indicate the user’s behavior habits and/or hobbies; send the acquired characteristic information of the user to the cloud service through the transceiver 403; receive it from the cloud service through the transceiver 403
  • the AI processing result, the AI processing result is obtained by the cloud service performing AI processing on the data stream of the computing device where the AI processing device 40 is located according to the characteristic information of the user of the computing device where the AI processing device 40 is located.
  • the processor 401 runs or executes software programs and/or modules stored in the memory 402, and calls data stored in the memory 402, Perform the following functions:
  • an embodiment of the present application provides an AI processing method. This method can be applied to the interaction process between the cloud server 302 and the computing device 303 in the data service system illustrated in FIG. 3. Among them, cloud services are configured in the cloud server.
  • the cloud server may configure the corresponding cloud service for each computing device.
  • the cloud service described in this application may be a cloud server corresponding to a computing device.
  • the communication between computing device and cloud service described in this article (for example, computing device sends certain content to cloud service, or computing device receives certain content from cloud service, or cloud service sends certain content to computing device, or cloud service receives certain content from computing device Certain content) can be replaced by: the computing device communicates with its corresponding cloud service.
  • the interaction process between the cloud server and different computing devices is the interaction process between the cloud services corresponding to the different computing devices in the cloud server and the corresponding computing devices.
  • the embodiment of the present application only takes the interaction process between the cloud server and a computing device as an example to describe the AI processing method provided in the embodiment of the present application.
  • the interaction process between the cloud server and each computing device is similar and will not be repeated one by one.
  • the AI processing method provided by the embodiment of the present application may include:
  • the computing device acquires characteristic information of the user of the computing device.
  • the user of the computing device may be a user who uses the computing device.
  • the characteristic information may be used to indicate the user's behavior habits and/or interests.
  • the characteristic information of the user may include: user behavior information, and/or user interest information.
  • the user behavior element information is used to indicate the characteristics of the user's use of the computing device.
  • the user behavior information may include one or more of the following information: loyal product brand information, commonly used APP information, online duration information; user interest information To indicate one or more types of interests of the user.
  • the type of interest may include, but is not limited to: fashion, sports, reading, travel, food, or communication.
  • the computing device may learn to record the characteristic information of the user during the user's use of the computing device.
  • the embodiment of the present application does not limit the process by which the computing device obtains the characteristic information.
  • the computing device sends the characteristic information of the user to the cloud service.
  • the cloud service may correspond to the computing device.
  • Cloud services can be virtual devices in configuration and cloud servers.
  • cloud services corresponding to different computing devices can be isolated from each other.
  • the cloud service may be a virtual device configured according to virtual machine technology or container technology.
  • the computing device sends the user's characteristic information to the cloud service corresponding to it, which can also be replaced by: the computing device sends the user's characteristic information to the cloud server, and the cloud server is configured with the cloud service.
  • the cloud server may be a content distribution server.
  • the cloud server and the content distribution server may be different servers.
  • the deployment mode of the cloud server can be configured according to actual requirements, which is not limited in the embodiment of the present application.
  • the cloud server receives the characteristic information of the user of the computing device from the computing device.
  • the characteristic information of the user received by the cloud server in S503 is the characteristic information sent by the computing device in S502, which will not be repeated here.
  • S504 The cloud server performs AI processing on the data stream according to the characteristic information of the user to obtain an AI processing result.
  • the data stream can be a data stream of a computing device, and can be understood as a data stream sent to the computing device.
  • the data stream may be received by the cloud server from the content distribution server.
  • the data stream may also be obtained by the cloud server in other ways, which is not limited in this embodiment of the application.
  • the AI processing method provided by the embodiment of the present application may further include: the cloud server receives the data stream from the content distribution server.
  • the AI processing method provided in the embodiment of the present application may further include: the cloud server receives a request message from the computing device, and according to the request message, requests the content distribution server for the computing device data flow.
  • the cloud server is a content distribution server, and the cloud server reads the data stream requested by the computing device within the content distribution server according to the request of the computing device as the data stream of the computing device.
  • the cloud server and the content distribution server are different servers.
  • the cloud server obtains the data stream of the computing device from the content distribution server, which can be specifically implemented as follows: the cloud server requests the content distribution server for the data stream; The cloud server receives the requested data stream from the content distribution server.
  • the type of data stream is related to the type of service performed by the computing device, which is not limited in this application.
  • the data stream may include one or more of the following content: video stream, picture stream, or text stream.
  • the cloud server inputs the user's characteristic information and data stream into the trained network model, infers and obtains a conclusion, and uses the inferred conclusion as the AI processing result.
  • the embodiments of the present application do not specifically limit the AI processing process and the form of the result.
  • the computing device is a smart phone, and the smart phone performs video playback services.
  • the data stream of the computing device can be a video stream.
  • AI processing can be to identify people, objects, and scenes in images in the video stream, and combine user characteristic information to obtain AI
  • the processing result may include targeted push of advertisements, item labeling, and other information.
  • the computing device is an e-book
  • the e-book performs text reading services
  • the data stream of the computing device is a text stream.
  • AI processing can be to recognize the text in the e-book text stream, combined with user characteristic information, and the AI processing result can include text commentary. Words and other information.
  • S505 The cloud server sends the AI processing result to the computing device.
  • the computing device receives the AI processing result from the cloud service.
  • the computing device receives the AI processing result from the cloud service, which can be understood as: the computing device receives the AI processing result from its corresponding cloud service.
  • S507 can be replaced with: the computing device receives the AI processing result from the cloud server where the cloud service is located.
  • the AI processing result is sent by the cloud server in S505, and the AI processing result is obtained by the cloud service performing AI processing on the data stream of the computing device according to the characteristic information.
  • the cloud service by configuring the cloud service, when there is a data stream sent to the computing device, the cloud service performs AI processing on the data stream in real time, and the computing device receives the AI processing result from the cloud service in real time.
  • This process Using the existing architecture, the real-time AI processing of the data stream is performed on the server without increasing the cost, simple deployment, and no requirement for computing device capabilities.
  • the AI processing method provided in the embodiment of the present application may further include S507.
  • the computing device displays the AI processing result.
  • the AI processing method provided in the embodiment of the present application may further include S502a to S502c.
  • the computing device sends a registration request to the cloud server.
  • the registration request is used to configure the cloud service.
  • the registration request is used to configure the cloud service corresponding to the computing device.
  • the computing device can execute S502a when it is powered on, or the computing device can execute S502a when an APP installed in the computing device is started.
  • the cloud server receives a registration request from the computing device.
  • the cloud server configures cloud services.
  • the cloud server configures cloud services inside. This application does not specifically limit the configuration process.
  • the cloud server configures the cloud service corresponding to the computing device in the cloud server.
  • the cloud server may also send a registration response message to the computing device to notify the computing device that the cloud service is configured.
  • the cloud server in S502c may use virtual machine technology or container technology to configure the cloud service corresponding to the computing device.
  • the AI processing method provided in the embodiment of the present application may further include S506a.
  • the computing device obtains configuration parameters of the user of the computing device.
  • the configuration parameters can be input by the user of the computing device in the human-computer interaction interface of the computing device, or the configuration parameters can also be default parameters of the computing device.
  • Configuration parameters can include one or more of the following: AI function on or off, person recognition function on or off, object recognition function on or off, character mental activity recognition function on or off, character emotion recognition function on or off , Or the scene recognition function is turned on or off.
  • whether the AI function is turned on or off indicates whether to perform AI processing. If the AI function is turned on, AI processing is performed on the data stream of the computing device; if the AI function is turned off, there is no need to perform AI processing on the data stream of the computing device.
  • the opening or closing of the person recognition function indicates whether to perform person recognition during AI processing. If the person recognition function is turned on, person recognition is performed when AI processing is performed on the data stream of the computing device; if the person recognition function is turned off, no person recognition is performed when performing AI processing on the data stream of the computing device.
  • the item recognition function is turned on or off to indicate whether to perform item recognition during AI processing. If the item identification function is enabled, item identification is performed when AI processing is performed on the data stream of the computing device; if the item identification function is disabled, item recognition is not performed when AI processing is performed on the data stream of the computing device.
  • the character mental activity recognition function is turned on or off to indicate whether to perform character mental activity recognition during AI processing. If the character mental activity recognition function is turned on, the character mental activity recognition will be performed when the data stream of the computing device is processed by AI; if the character mental activity recognition function is turned off, the character mental activity will not be performed when the data stream of the computing device is processed by AI Activity recognition.
  • the character emotion recognition function is turned on or off to indicate whether to perform character emotion recognition during AI processing. If the character emotion recognition function is turned on, the character emotion recognition will be performed when AI processing is performed on the data stream of the computing device; if the character emotion recognition function is turned off, the character emotion recognition will not be performed when AI processing is performed on the data stream of the computing device.
  • the opening or closing of the scene recognition function indicates whether to perform scene recognition during AI processing. If the scene recognition function is turned on, scene recognition is performed when AI processing is performed on the data stream of the computing device; if the scene recognition function is turned off, scene recognition is not performed when AI processing is performed on the data stream of the computing device.
  • the computing device sends the user's configuration parameters to the cloud service.
  • the AI processing method used in the embodiment of the present application may further include S504a.
  • the cloud server receives configuration parameters of the user from the computing device.
  • the cloud server performs AI processing on the data stream of the computing device according to the user's characteristic information to obtain the AI processing result, which can be specifically implemented as follows: the cloud server performs AI on the data stream of the computing device according to the characteristic information and configuration parameters Processing to get the AI processing result.
  • the computing device receives the AI processing result from the cloud service, which can be specifically implemented as: the computing device receives the AI processing result from the cloud service.
  • the AI processing result is that the cloud service performs the data flow of the computing device according to the characteristic information and configuration parameters. AI processed.
  • a computing device and a cloud server include hardware structures and/or software modules corresponding to each function.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is executed by hardware or computer software-driven hardware depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.
  • the embodiment of the present application may divide the computing device and the cloud server into functional modules according to the foregoing method examples.
  • each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module.
  • the above-mentioned integrated modules can be implemented in the form of hardware or software function modules. It should be noted that the division of modules in the embodiments of the present application is illustrative, and is only a logical function division, and there may be other division methods in actual implementation.
  • an AI processing device 70 provided in an embodiment of this application is used to implement the function of the computing device in the foregoing method.
  • the AI processing device 70 may be a computing device, a device in a computing device, or a device that can be matched and used with a computing device.
  • the AI processing device 70 may be a chip system.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • the AI processing device 70 may include: an acquiring unit 701, a sending unit 702, and a receiving unit 703.
  • the acquiring unit 701 is used to execute S501 and S506a in FIG. 5 or FIG.
  • the sending unit 702 is used to execute S502, S502a, and S506b in FIG. 5 or FIG. 6, and the receiving unit 703 is used to execute S506 in FIG. 5 or FIG. .
  • the sending unit 702 is used to execute S502, S502a, and S506b in FIG. 5 or FIG. 6, and the receiving unit 703 is used to execute S506 in FIG. 5 or FIG. .
  • all relevant content of the steps involved in the above method embodiments can be cited in the functional description of the corresponding functional module, which will not be repeated here.
  • an AI processing device 80 provided in this embodiment of the present application is used to implement the functions of the computing device in the foregoing method.
  • the AI processing device 80 may be a computing device, a device in a computing device, or a device that can be matched and used with a computing device. Wherein, the AI processing device 80 may be a chip system.
  • the AI processing device 80 includes at least one processing module 801 and a communication module 802. Exemplarily, the processing module 801 may be used to execute S501 and S506a in FIG. 5 or FIG. 6; the processing module 801 may be used to execute S502, S502a, S506b, and S506 in FIG. 5 or FIG. 6 through the communication module 802. For details, please refer to the detailed description in the method example, which will not be repeated here.
  • the AI processing device 80 may also include at least one storage module 803 for storing program instructions and/or data.
  • the storage module 803 and the processing module 801 are coupled.
  • the coupling in the embodiments of the present application is an indirect coupling or communication connection between devices, units or modules, and may be in electrical, mechanical or other forms, and is used for information exchange between devices, units or modules.
  • the processing module 801 may cooperate with the storage module 802 to operate.
  • the processing module 801 may execute program instructions stored in the storage module 802. At least one of the at least one storage module may be included in the processing module.
  • the AI processing device 80 involved in FIG. 8 in the embodiment of the present application may be the AI processing device 40 shown in FIG. 4.
  • the AI processing device 70 or the AI processing device 80 provided by the embodiments of the present application can be used to implement the functions of the computing devices in the methods implemented by the various embodiments of the present application.
  • the various embodiments of this application please refer to the various embodiments of this application.
  • an AI processing device 90 provided in this embodiment of the present application is used to implement the function of the cloud server in the above method.
  • the AI processing device 90 may be a cloud server, a device in a cloud server, or a device that can be matched and used with a cloud server.
  • the AI processing device 90 may be a chip system.
  • the chip system may be composed of chips, or may include chips and other discrete devices.
  • the AI processing device 90 may include: a receiving unit 901, a processing unit 902, and a sending unit 903.
  • the receiving unit 901 is configured to execute S503, S502b, and S504a in FIG. 5 or FIG.
  • the processing unit 902 is configured to execute S504 in FIG. 5 or FIG. 6, and the sending unit 903 is configured to execute S505 in FIG. 5 or FIG.
  • the processing unit 902 is configured to execute S504 in FIG. 5 or FIG. 6
  • the sending unit 903 is configured to execute S505 in FIG. 5 or FIG.
  • all relevant content of the steps involved in the above method embodiments can be cited in the functional description of the corresponding functional module, which will not be repeated here.
  • an AI processing apparatus 100 provided in an embodiment of the present application is used to implement the function of the cloud server in the above method.
  • the AI processing device 100 may be a cloud server, a device in a cloud server, or a device that can be matched and used with a cloud server. Wherein, the AI processing device 100 may be a chip system.
  • the AI processing device 100 includes at least one processing module 1001 and a communication module 1002. Exemplarily, the processing module 1001 may be used to execute S504 in FIG. 5 or FIG. 6; the processing module 1001 may be used to execute S503, S502b, S504a, and S505 in FIG. 5 or FIG. 6 through the communication module 1002. For details, please refer to the detailed description in the method example, which will not be repeated here.
  • the AI processing device 100 may also include at least one storage module 1003 for storing program instructions and/or data.
  • the storage module 1003 and the processing module 1001 are coupled.
  • the coupling in the embodiments of the present application is an indirect coupling or communication connection between devices, units or modules, and may be in electrical, mechanical or other forms, and is used for information exchange between devices, units or modules.
  • the processing module 1001 may operate in cooperation with the storage module 1002.
  • the processing module 1001 may execute program instructions stored in the storage module 1002. At least one of the at least one storage module may be included in the processing module.
  • the AI processing device 100 involved in FIG. 10 in the embodiment of the present application may be the AI processing device 40 shown in FIG. 4.
  • the AI processing device 90 or the AI processing device 100 provided in the embodiments of the present application can be used to implement the functions of the cloud server in the methods implemented by the various embodiments of the present application.
  • the same as those in the embodiments of the present application are shown.
  • specific technical details that are not disclosed please refer to the various embodiments of this application.
  • FIG. 5 or FIG. 6 shows the various steps performed by the computing device in the embodiment.
  • FIG. 5 or FIG. 6 shows the various steps performed by the cloud server in the embodiment.
  • FIG. 5 or FIG. 6 Other embodiments of the present application also provide a computer program product.
  • the computer program product runs on a computer, the computer executes the steps performed by the computing device or cloud server in the embodiment shown in FIG. 5 or FIG. 6.
  • the chip system includes an interface circuit and a processor; the interface circuit and the processor are interconnected by wires; the interface circuit is used to receive signals from the memory of the computing device and send signals to the processor.
  • the signals include computer instructions stored in the memory; when the processor When the computer instruction is executed, the chip system executes the various steps executed by the computing device in the embodiment shown in FIG. 5 or FIG. 6 above.
  • the chip system includes an interface circuit and a processor; the interface circuit and the processor are interconnected by wires; the interface circuit is used to receive signals from the memory of the cloud server and send signals to the processor.
  • the signals include computer instructions stored in the memory; when the processor When the computer instruction is executed, the chip system executes each step performed by the cloud server in the embodiment shown in FIG. 5 or FIG. 6 above.
  • the AI processing system includes the computing device described in any of the foregoing embodiments, and the cloud server described in any of the foregoing embodiments.
  • the disclosed device and method can be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods, for example, multiple units or components may be divided. It can be combined or integrated into another device, or some features can be omitted or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate parts may or may not be physically separate.
  • the parts displayed as units may be one physical unit or multiple physical units, that is, they may be located in one place, or they may be distributed to multiple different places. . Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium.
  • the technical solutions of the embodiments of the present application are essentially or the part that contributes to the prior art, or all or part of the technical solutions can be embodied in the form of a software product, and the software product is stored in a storage medium. It includes several instructions to make a device (may be a single-chip microcomputer, a chip, etc.) or a processor (processor) execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code .

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Neurology (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Computer Hardware Design (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

一种AI处理方法及装置，涉及人工智能领域，实现低成本易部署且不要求设备能力的实时数据流AI处理。具体方案为：计算设备获取该计算设备的用户的特征信息（S501），该特征信息用于指示该用户的行为习惯和/或兴趣爱好；计算设备向云服务发送获取的用户的特征信息（S502）；计算设备从云服务接收AI处理结果（S506），该AI处理结果是云服务根据该计算设备的用户的特征信息对该计算设备的数据流进行AI处理得到的（S504）。

Description

一种AI处理方法及装置 技术领域
本申请涉及人工智能(artificial intelligence，AI)领域，尤其涉及一种AI处理方法及装置。
背景技术
AI是研究、开发用于模拟、延伸和扩展人的智能的理论、方法、技术及应用系统的一门新的技术科学，可以在智能电子设备(例如智能手机、智能电视等)中采用AI技术进行AI处理，在不同场景中模拟实现人的智能。例如，采用AI技术可以实现手机识物软件、以图搜图软件、智能电视识物软件等，进而根据识别结果提供智能服务，以提高智能化的用户体验。
AI处理过程包括训练和推理两部分，训练是用海量数据作为输入训练深度神经网络模型；推断指利用训练好的深度神经网络模型，将待判断的数据输入该模型，去“推断”得出各种结论，作为AI处理的结果。例如，AI处理可以识别数据中的人、物、或者场景。不同场景下AI处理结果的内容不同。例如识物软件，AI处理结果可以是被识别物体的标注信息(比如物理名称、属性等)。例如广告推送软件，AI处理结果可以是广告推送内容。
当前，一种AI处理过程为请求-响应式。例如智能手机中图片识物的过程。如图1所示的AI处理场景，智能手机生成图片(可以来源于手机摄像头或者从其他渠道下载的图片)，该智能手机通过网络将图片发送到服务器，服务器进行AI处理并将AI处理后的结果发送给智能手机显示。该AI处理过程是典型的请求-响应式，在服务器AI处理阶段，智能手机处于等待结果状态，无法对数据流做实时AI处理。
另一种AI处理过程为电子设备自身进行AI处理，电子设备具备AI处理能力，可实时进行AI处理。该方案虽然能实时处理数据流，但对电子设备要求高。
再一种AI处理过程为服务器与电子设备之间部署了智能边缘小站的场景中，由智能边缘小站进行AI处理。如图1a所示的智能视频监控场景中，智能摄像机把图像数据流实时同步给智能边缘小站，由智能边缘小站进行AI处理，并把AI处理结果直接显示给用户或返回给摄像机显示。该方案虽然对数据流实时AI处理，但仅适用于实时监控场景，需要部署边缘智能节点，成本高，大规模部署困难。
因此，如何实现低成本易部署且不要求设备能力的实时数据流AI处理，成为亟待解决的问题。
发明内容
本申请提供一种AI处理方法及装置，实现低成本易部署且不要求设备能力的实时数据流AI处理。
为了达到上述目的，本申请采用如下技术方案：
第一方面，提供一种AI处理方法，该方法应用于计算设备，该方法可以包括：计算 设备获取该计算设备的用户的特征信息，该特征信息用于指示该用户的行为习惯和/或兴趣爱好；计算设备向云服务发送获取的用户的特征信息；计算设备从云服务接收AI处理结果，该AI处理结果是云服务根据该计算设备的用户的特征信息对该计算设备的数据流进行AI处理得到的。
通过本申请提供的AI处理方法，通过配置云服务，当有发往计算设备的数据流时，由云服务实时对该数据流进行AI处理，计算设备从云服务实时接收AI处理结果，该过程利用现有架构，在不增加成本、部署简单，且对计算设备能力无要求的前提下，在服务端对数据流实时AI处理。
其中，与计算设备交互的云服务，可以与该计算设备映射，计算设备中的程序(例如应用程序(application，APP))都在其对应的云服务中映射，计算设备可以通过云服务从内容分发服务器请求数据流。
结合第一方面，在一种可能的实现方式中，本申请提供的AI处理方法还可以包括：在向云服务发送用户的特征信息之前，计算设备向云服务器发送注册请求，该注册请求用于云服务器配置云服务。
例如，注册请求可以用于云服务器配置与发送注册请求的计算设备对应的云服务。
结合第一方面或上述任一种可能的实现方式，在另一种可能的实现方式中，上述云服务器与发送数据流的内容分发服务器可以为同一个服务器，或者，上述云服务器与发送数据流的内容分发服务器为不同的服务器。
结合第一方面或上述任一种可能的实现方式，在另一种可能的实现方式中，上述云服务可以为按照虚拟机技术或容器技术配置的虚拟设备，以实现不同终端设备对应的云服务相互隔离。
结合第一方面或上述任一种可能的实现方式，在另一种可能的实现方式中，本申请提供的AI处理方法还可以包括：在从云服务接收AI处理结果之前，计算设备获取用户的配置参数，向云服务发送配置参数。该配置参数可以包括下述内容中一项或多项：AI功能开启或关闭、人物识别功能开启或关闭、物品识别功能开启或关闭、人物心理活动识别功能开启或关闭、人物情感识别功能开启或关闭、或者场景识别功能开启或关闭。相应的，计算设备从云服务接收AI处理结果，具体可以实现为：计算设备从云服务接收AI处理结果，其中，该AI处理结果是该云服务根据特征信息及配置参数对该计算设备的数据流进行AI处理得到的。
结合第一方面或上述任一种可能的实现方式，在另一种可能的实现方式中，用户的特征信息可以包括：用户行为信息，和/或，用户兴趣信息。其中，用户行为要素信息用于指示用户使用计算设备行为的特征，用户行为信息可以包括下述信息中一项或多项：忠实商品品牌信息、常用APP信息、在网时长信息；用户兴趣信息用于指示用户的一种或多种兴趣类型。
结合第一方面或上述任一种可能的实现方式，在另一种可能的实现方式中，AI处理结果可以包括下述内容中一项或多项：广告投放、物品标注、及解说词。
结合第一方面或上述任一种可能的实现方式，在另一种可能的实现方式中，数据流可以包括下述内容中一项或多项：视频流、图片流、文字流。
需要说明的是，本申请中描述的云服务可以为与计算设备对应的云服务器。本文所描 述的计算设备与云服务的通信(例如计算设备向云服务发送某内容，或者，计算设备从云服务接收某内容，或者云服务向计算设备发送某内容，或者云服务从计算设备接收某内容)，可以替代为：计算设备与其对应的云服务通信。
一种可能的实现方式中，当存在多个计算设备(例如进行某一数据业务的多个智能手机)时，云服务器可以为每个计算设备配置一个与其对应的云服务。
可选的，不同计算设备对应的云服务之间可以互相隔离。
第二方面，提供另一种AI处理方法，该方法可以应用于云服务器，该云服务器中配置云服务。该方法可以包括：云服务器从计算设备接收计算设备的用户的特征信息；其中，特征信息用于指示用户的行为习惯和/或兴趣爱好；云服务器根据用户的特征信息，对数据流进行AI处理得到AI处理结果；云服务器向该计算设备发送AI处理结果。
通过本申请提供的AI处理方法，通过配置云服务，当有发往计算设备的数据流时，由云服务实时对该数据流进行AI处理，计算设备从云服务实时接收AI处理结果，该过程利用现有架构，在不增加成本、部署简单，且对计算设备能力无要求的前提下，在服务端对数据流实时AI处理。
其中，上述数据流可是指该计算设备的数据流，可以理解为发往该计算设备的数据流。
结合第二方面，在一种可能的实现方式中，上述云服务器可以为内容分发服务器。
结合第二方面，在另一种可能的实现方式中，上述云服务器与提供数据流的内容分发服务器为不同的服务器。本申请提供的AI处理方法还可以包括：在根据特征信息对数据流进行AI处理之前，云服务器从内容分发服务器获取计算设备的数据流。
结合第二方面或上述任一种可能的实现方式，在另一种可能的实现方式中，云服务器从内容分发服务器获取计算设备的数据流，具体可以实现为：向内容分发服务器请求该计算设备的数据流；从内容分发服务器接收该计算设备的数据流。
结合第二方面或上述任一种可能的实现方式，在另一种可能的实现方式中，在根据特征信息对数据流进行AI处理之前，本申请提供的AI处理方法还可以包括：云服务器从内容分发服务器接收数据流。
结合第二方面或上述任一种可能的实现方式，在另一种可能的实现方式中，云服务可以为按照虚拟机技术或容器技术配置的虚拟设备。
结合第二方面或上述任一种可能的实现方式，在另一种可能的实现方式中，本申请提供的AI处理方法还可以包括：在根据特征信息，对数据流进行AI处理得到AI处理结果之前，云服务器从计算设备接收配置参数，该配置参数可以包括下述内容中一项或多项：AI功能开启或关闭、人物识别功能开启或关闭、物品识别功能开启或关闭、人物心理活动识别功能开启或关闭、人物情感识别功能开启或关闭、或者场景识别功能开启或关闭。相应的，云服务器根据特征信息，对数据流进行AI处理得到AI处理结果，具体可以实现为：根据特征信息及配置参数，对数据流进行AI处理得到AI处理结果。
结合第二方面或上述任一种可能的实现方式，在另一种可能的实现方式中，用户的特征信息可以包括：用户行为信息，和/或，用户兴趣信息。其中，用户行为要素信息用于指示用户使用计算设备行为的特征，用户行为信息可以包括下述信息中一项或多项：忠实商品品牌信息、常用APP信息、在网时长信息；用户兴趣信息用于指示用户的一种 或多种兴趣类型。
结合第二方面或上述任一种可能的实现方式，在另一种可能的实现方式中，AI处理结果可以包括下述内容中一项或多项：广告投放、物品标注、及解说词。
结合第二方面或上述任一种可能的实现方式，在另一种可能的实现方式中，数据流可以包括下述内容中一项或多项：视频流、图片流、文字流。
需要说明的是，本申请第二方面提供的AI处理方法，与第一方面提供的AI处理方法，是同一技术方案不同角度的描述，其具体实现可以相互参考。
第三方面，本申请提供一种AI处理装置，该AI处理装置可以是计算设备，也可以是计算设备中的装置或者芯片系统，或者是能够和计算设备匹配使用的装置。该AI处理装置可以实现上述各方面或者各可能的设计中计算设备所执行的功能，所述功能可以通过硬件实现，也可以通过硬件执行相应的软件实现。所述硬件或软件包括一个或多个上述功能相应的模块。如：该AI处理装置可以包括：获取单元、发送单元及接收单元。
其中，获取单元，用于获取该AI处理装置的用户的特征信息，特征信息用于指示用户的行为习惯和/或兴趣爱好。
发送单元，用于向云服务发送获取单元获取的特征信息。
接收单元，用于从云服务接收AI处理结果。其中，AI处理结果是云服务根据特征信息对该AI处理装置的数据流进行AI处理得到的。
需要说明的是，第三方面提供的AI处理装置，用于执行上述第一方面提供的AI处理方法，具体实现可以参考上述第一方面的具体实现。
第四方面，本申请提供一种AI处理装置，该AI处理装置可以是云服务器，也可以是云服务器中的装置或者芯片系统，或者是能够和云服务器匹配使用的装置。该云服务器中配置有云服务。该AI处理装置可以实现上述各方面或者各可能的设计中云服务器所执行的功能，所述功能可以通过硬件实现，也可以通过硬件执行相应的软件实现。所述硬件或软件包括一个或多个上述功能相应的模块。如：该AI处理装置可以包括：接收单元、处理单元及发送单元。
其中，接收单元，用于从计算设备接收计算设备的用户的特征信息。其中，特征信息用于指示用户的行为习惯和/或兴趣爱好。
处理单元，用于根据接收单元接收的特征信息，对数据流进行AI处理得到AI处理结果。
发送单元，用于向计算设备发送处理单元得到的AI处理结果。
需要说明的是，第四方面提供的AI处理装置，用于执行上述第二方面提供的AI处理方法，具体实现可以参考上述第二方面的具体实现。
第五方面，本申请提供一种计算设备，该计算设备可以包括：处理器，存储器；处理器，存储器耦合，存储器可用于存储计算机指令，处理器用于调用计算机指令，以执行如第一方面或任一种可能的实现方式所述的AI处理方法。
第六方面，本申请提供一种云服务器，该云服务器可以包括：处理器，存储器；处理器，存储器耦合，存储器可用于存储计算机指令，处理器用于调用计算机指令，以执行如第二方面或任一种可能的实现方式所述的AI处理方法。
第七方面，本申请提供一种计算机可读存储介质，该计算机可读存储介质可以 包括：计算机软件指令；当计算机软件指令在计算设备中运行时，使得该计算设备执行如第一方面或第一方面的可能实现方式中任一项所述的AI处理方法。
第八方面，本申请提供一种计算机可读存储介质，该计算机可读存储介质可以包括：计算机软件指令；当计算机软件指令在云服务器中运行时，使得该云服务器执行如第二方面或第二方面的可能实现方式中任一项所述的AI处理方法。
第九方面，本申请提供一种计算机程序产品，当该计算机程序产品在计算机上运行时，使得该计算机执行如权利要求第一方面或任一种可能的实现方式中任一项所述的AI处理方法。
第十方面，本申请提供一种芯片系统，该芯片系统应用于计算设备；计算设备包括接口电路和处理器；接口电路和处理器通过线路互联；接口电路用于从计算设备的存储器接收信号，并向处理器发送信号，信号包括存储器中存储的计算机指令；当处理器执行该计算机指令时，芯片系统执行如第一方面或任一种可能的实现方式中任一项所述的AI处理方法。
第十一方面，本申请提供一种芯片系统，该芯片系统应用于云服务器；云服务器包括接口电路和处理器；接口电路和处理器通过线路互联；接口电路用于从云服务器的存储器接收信号，并向处理器发送信号，信号包括存储器中存储的计算机指令；当处理器执行该计算机指令时，芯片系统执行如第二方面或任一种可能的实现方式中任一项所述的AI处理方法。
第十二方面，本申请提供一种AI处理系统，包括计算设备及云服务器。其中，计算设备执行如第一方面或第一方面的可能实现方式中任一项所述的AI处理方法，云服务器执行如第二方面或第二方面的可能实现方式中任一项所述的AI处理方法。
第十三方面，本申请提供另一种AI处理系统，包括计算设备及云服务器，云服务中配置了云服务。其中：计算设备用于：获取该计算设备的用户的特征信息，该特征信息用于指示该用户的行为习惯和/或兴趣爱好；向云服务发送获取的用户的特征信息。云服务器用于：从计算设备接收计算设备的用户的特征信息；其中，特征信息用于指示用户的行为习惯和/或兴趣爱好；根据用户的特征信息，对该计算设备的数据流进行AI处理得到AI处理结果；向该计算设备发送AI处理结果。计算设备还用于：从云服务接收AI处理结果，该AI处理结果是云服务根据该计算设备的用户的特征信息对该计算设备的数据流进行AI处理得到的。
应当理解的是，本申请中对技术特征、技术方案、有益效果或类似语言的描述并不是暗示在任意的单个实施例中可以实现所有的特点和优点。相反，可以理解的是对于特征或有益效果的描述意味着在至少一个实施例中包括特定的技术特征、技术方案或有益效果。因此，本说明书中对于技术特征、技术方案或有益效果的描述并不一定是指相同的实施例。进而，还可以任何适当的方式组合本实施例中所描述的技术特征、技术方案和有益效果。本领域技术人员将会理解，无需特定实施例的一个或多个特定的技术特征、技术方案或有益效果即可实现实施例。在其他实施例中，还可在没有体现所有实施例的特定实施例中识别出额外的技术特征和有益效果。
附图说明
图1为现有技术中一种AI处理场景示意图；
图1a为现有技术中另一种AI处理场景示意图；
图2为现有技术提供的一种数据服务系统的架构示意图；
图3为现有技术提供的一种数据服务系统的架构示意图；
图4为本申请实施例提供的一种AI处理装置的结构示意图；
图5为本申请实施例提供的一种AI处理方法的流程示意图；
图6为本申请实施例提供的另一种AI处理方法的流程示意图；
图7为本申请实施例提供的另一种AI处理装置的结构示意图；
图8为本申请实施例提供的再一种AI处理装置的结构示意图；
图9为本申请实施例提供的再一种AI处理装置的结构示意图；
图10为本申请实施例提供的又一种AI处理装置的结构示意图。
具体实施方式
本申请说明书和权利要求书及上述附图中的术语“第一”、“第二”和“第三”等是用于区别不同对象，而不是用于限定特定顺序。
在本申请实施例中，“示例性的”或者“例如”等词用于表示作例子、例证或说明。本申请实施例中被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。确切而言，使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念。
为了便于理解，先对本申请涉及的名词进行解释。
数据流，可以是发往计算设备的数据集合，数据流可以包括但不限于下述内容中一项或多项：视频流、图片流、文字流等。
内容分发服务器，可以是向计算设备提供数据流的服务器。该数据流可以为业务数据流或者其他，本申请实施例对此不进行具体限定。内容分发服务器可以根据规则或者根据计算设备的请求，向计算设备提供数据流。
云服务器，可以是云端部署的实现集中式服务的服务器。
计算设备，可以是数据流的需求设备。计算设备在获取数据流后可以显示或进行相应处理，或者计算设备在获取数据流后通过其他设备显示，或者计算设备在获取数据流后经过相应处理后通过其他设备显示，本申请不予限定。计算设备可以为服务器或者终端设备，本申请对于计算设备的实际产品形态不予限定。
AI处理，可以包括训练和推理两部分，训练是用海量数据输入，训练出一个网络模型(例如复杂的深度神经网络模型)。推断指利用训练好的网络模型，使用待判断的数据去“推断”得出各种结论，将该结论作为AI处理结果。推断得出的结论，不同场景下内容不同，例如识物软件，处理后的结果是被识别物体的标注信息，比如物理名称、属性等。
目前，在图2示意的数据服务系统中，计算设备可以从内容分发服务器请求获取数据流。对于不具备AI处理能力的计算设备，在获取数据流或生成数据流后如需进行AI处理，则按照前述的请求-响应式进行AI处理，过程耗时很长，无法在服务端实现数据流的实时AI处理。
基于此，本申请实施例提供一种AI处理方法，通过配置云服务，云服务中映射了计算设备的程序及操作，将计算设备的AI处理需求映射到云服务中，由云服务利用服务端的AI处理能力，实时对计算设备的数据流进行AI处理，实现在计算设备不具备AI处理能力的前提下，服务端实现数据流的实时AI处理。
下面将结合附图对本申请实施例的实施方式进行详细描述。
本申请实施例提供的AI处理方法可以应用于图3所示的数据服务系统中。如图3所示，该数据服务系统可以包括内容分发服务器301、云服务器302以及计算设备303。
其中，云服务器302中配置了云服务3021。计算设备303通过云服务3021，从内容分发服务器301请求数据流。内容分发服务器301通过云服务3021，向计算设备303发送数据流。
一种可能的实现方式中，云服务3021从内容分发服务器301接收到发往计算设备303的数据流(也可以称之为计算设备303的数据流)时，可以对该数据流进行AI处理，将数据流及数据流的AI处理结果发送给计算设备303。
一种可能的实现方式中，云服务3021可以按照计算设备303的配置，对该计算设备303的数据流进行AI处理，将数据流及数据流的AI处理结果发送给计算设备303。
进一步的，计算设备303可以对AI处理结果进行显示，或者，计算设备303可以将AI处理结果发送给其他设备显示，本申请实施例对此不予限定。
内容分发服务器301可以为内容提供商的服务器。云服务器302可以为提供云服务的服务器。
一种可能的实现方式中，云服务器302即为内容分发服务器301。
另一种可能的实现方式中，云服务器302与内容分发服务器301分别独立部署，两者之间通过超文本传输协议(hyper text transport protocol，HTTP)协议通信。
计算设备303可以为服务器或者终端设备或者其他，本申请实施例对于计算设备303的具体产品形态不进行限定。
本申请实施例涉及到的终端设备，可以是一种具有无线收发功能的设备，其可以部署在陆地上，包括室内或室外、手持或车载；也可以部署在水面上(如轮船等)；还可以部署在空中(例如飞机、气球和卫星上等)。终端设备可以是用户设备(user equipment，UE)，UE包括具有无线通信功能的手持式设备、车载设备、可穿戴设备或其他。示例性地，UE可以是手机(mobile phone)、平板电脑或带无线收发功能的电脑。终端设备还可以是虚拟现实(virtual reality，VR)终端设备、增强现实(augmented reality，AR)终端设备、工业控制中的无线终端、无人驾驶中的无线终端、远程医疗中的无线终端、智能电网中的无线终端、智慧城市(smart city)中的无线终端、智慧家庭(smart home)中的无线终端等等。
需要指出的是，图3所示数据服务系统的架构仅为示例性架构图，本申请实施例不限定图3所示架构中包括的网元的数量。虽然未示出，但除图3所示网元外，图3所示系统还可以包括其他功能实体。此外，上述图3架构中的网元的命名只是一个示例，不予限定。
下面结合附图，对本申请的实施例进行具体阐述。
一方面，本申请实施例提供一种AI处理装置，用于执行本申请提供的AI处理方法，该AI处理装置可以部署于图3所示的数据服务系统中的计算设备或云服务器中。图4示出的是与本申请各实施例相关的一种AI处理装置40。如图4所示，AI处理装置40可以包括处理器401、存储器402以及收发器403。
下面结合图4对AI处理装置40的各个构成部件进行具体的介绍：
其中，存储器402可以是易失性存储器(volatile memory)，例如随机存取存储器(random-access memory，RAM)；或者非易失性存储器(non-volatile memory)，例如只读存储器(read-only memory，ROM)，快闪存储器(flash memory)，硬盘(hard disk drive，HDD)或固态硬盘(solid-state drive，SSD)；或者上述种类的存储器的组合，用于存储可实现本申请方法的程序代码、配置文件或其他内容。
处理器401是AI处理装置40的控制中心。例如，处理器401可以是一个中央处理器(central processing unit，CPU)，也可以是特定集成电路(application specific integrated circuit，ASIC)，或者是被配置成实施本申请实施例的一个或多个集成电路，例如：一个或多个微处理器(digital signal processor，DSP)，或，一个或者多个现场可编程门阵列(field programmable gate array，FPGA)。
收发器403用于与其他设备进行通信。
一种可能的实现方式中，当AI处理装置40部署于计算设备时，处理器401通过运行或执行存储在存储器402内的软件程序和/或模块，以及调用存储在存储器402内的数据，执行如下功能：
获取该计算设备的用户的特征信息，该特征信息用于指示该用户的行为习惯和/或兴趣爱好；通过收发器403向云服务发送获取的用户的特征信息；通过收发器403从云服务接收AI处理结果，该AI处理结果是云服务根据该AI处理装置40所在的计算设备的用户的特征信息，对该AI处理装置40所在的计算设备的数据流进行AI处理得到的。
另一种可能的实现方式中，当AI处理装置40部署于云服务器时，处理器401通过运行或执行存储在存储器402内的软件程序和/或模块，以及调用存储在存储器402内的数据，执行如下功能：
从计算设备接收计算设备的用户的特征信息；其中，特征信息用于指示用户的行为习惯和/或兴趣爱好；根据用户的特征信息，对数据流进行AI处理得到AI处理结果；向该计算设备发送AI处理结果。
另一方面，本申请实施例提供一种AI处理方法。该方法可以应用于图3示意的数据服务系统中，云服务器302与计算设备303的交互过程。其中，云服务器中配置了云服务。
当云服务器与多个计算设备通信时，云服务器可以为每个计算设备配置与其对应的云服务。本申请中描述的云服务可以为与计算设备对应的云服务器。本文所描述的计算设备与云服务的通信(例如计算设备向云服务发送某内容，或者，计算设备从云服务接收某内容，或者云服务向计算设备发送某内容，或者云服务从计算设备接收某内容)，可以替代为：计算设备与其对应的云服务通信。
云服务器与不同计算设备的交互过程，是云服务器中与不同计算设备分别对应 的云服务与各自对应的计算设备的交互过程。本申请实施例仅以云服务器与一个计算设备的交互过程为例，描述本申请实施例提供的AI处理方法，云服务器与每个计算设备的交互过程相似，不再一一赘述。
如图5所示，本申请实施例提供的AI处理方法可以包括：
S501、计算设备获取该计算设备的用户的特征信息。
其中，计算设备的用户，可以为使用该计算设备的用户。
具体的，特征信息可以用于指示用户的行为习惯和/或兴趣爱好。
例如，用户的特征信息可以包括：用户行为信息，和/或，用户兴趣信息。其中，用户行为要素信息用于指示用户使用计算设备行为的特征，用户行为信息可以包括下述信息中一项或多项：忠实商品品牌信息、常用APP信息、在网时长信息；用户兴趣信息用于指示用户的一种或多种兴趣类型。
其中，兴趣类型可以包括但不限于：时尚、运动、阅读、旅游、美食、或交际等。
一种可能的实现方式中，计算设备可以在用户使用该计算设备的过程中，学习记录用户的特征信息。当然，本申请实施例对于计算设备获取特征信息的过程不予限定。
S502、计算设备向云服务发送用户的特征信息。
其中，该云服务可以为与该计算设备对应。云服务可以是配置与云服务器中的虚拟设备。
一种可能的实现中，不同计算设备对应的云服务可以相互隔离。例如，云服务可以为按照虚拟机技术或容器技术配置的虚拟设备。
具体的，在S502中，计算设备向与其对应的云服务发送用户的特征信息，也可以替换为：计算设备向云服务器发送用户的特征信息，该云服务器中配置了与云服务。
一种可能的实现方式中，云服务器可以为内容分发服务器。
另一种可能的实现中，云服务器与内容分发服务器可以为不同的服务器。
需要说明的是，在实际应用中，云服务器的部署方式可以根据实际需求配置，本申请实施例不予限定。
S503、云服务器从计算设备接收计算设备的用户的特征信息。
其中，S503中云服务器接收的用户的特征信息，即S502中计算设备发送的特征信息，此处不再赘述。
S504、云服务器根据用户的特征信息，对数据流进行AI处理得到AI处理结果。
其中，该数据流可以为计算设备的数据流，可以理解为发往计算设备的数据流。
可选的，数据流可以为云服务器从内容分发服务器接收到的，当然，数据流也可以为云服务器通过其他方式获取的，本申请实施例不予限定。
相应的，本申请实施例提供的AI处理方法还可以包括：云服务器从内容分发服务器接收数据流。
进一步的，在云服务器从内容分发服务器接收数据流之前，本申请实施例提供的AI处理方法还可以包括：云服务器接收计算设备的请求消息，根据该请求消息向内容分发服务器请求该计算设备的数据流。
一种可能的实现方式中，云服务器为内容分发服务器，云服务器则根据计算设备的请求在内容分发服务器内部读取计算设备请求的数据流，作为计算设备的数据流。
另一种可能的实现方式中，云服务器与内容分发服务器为不同的服务器，S504中云服务器从内容分发服务器获取计算设备的数据流，具体可以实现为：云服务器向内容分发服务器请求数据流；云服务器从内容分发服务器接收请求的数据流。
具体的，数据流的类型与计算设备进行的业务类型相关，本申请对此不予限定。示例性的，数据流可以包括下述内容中一项或多项：视频流、图片流、或文字流。
具体的，S504中云服务器将用户的特征信息以及数据流，输入训练好的网络模型，推断得到结论，将推断结论作为AI处理结果。本申请实施例对于AI处理的过程以及结果的形式均不进行具体限定。
例如，计算设备为智能手机，智能手机进行视频播放业务，计算设备的数据流可以为视频流，AI处理可以是识别视频流中的图像中的人物、物品和场景，结合用户特征信息，得到AI处理结果可以包括针对性地推送广告、物品标注等信息。
例如，计算设备为电子书，电子书进行文字阅读业务，计算设备的数据流为文字流，AI处理可以是识别电子书文字流中的文字，结合用户特征信息，得到AI处理结果可以包括文字解说词等信息。
S505、云服务器向计算设备发送AI处理结果。
S506、计算设备从云服务接收AI处理结果。
具体的，S506中计算设备从云服务接收AI处理结果，可以理解为：计算设备从其对应的云服务接收AI处理结果。在实际应用中，从设备的角度来描述，S507可以替换为：计算设备从云服务所在的云服务器接收AI处理结果。
其中，该AI处理结果即S505中云服务器发送的，该AI处理结果是云服务根据特征信息对计算设备的数据流进行AI处理得到的。
通过本申请提供的AI处理方法，通过配置云服务，当有发往计算设备的数据流时，由云服务实时对该数据流进行AI处理，计算设备从云服务实时接收AI处理结果，该过程利用现有架构，在不增加成本、部署简单，且对计算设备能力无要求的前提下，在服务端对数据流实时AI处理。
进一步的，如图6所示，在S506之后，本申请实施例提供的AI处理方法还可以包括S507。
S507、计算设备显示AI处理结果。
进一步的，如图6所示，在S502之前，本申请实施例提供的AI处理方法还可以包括S502a至S502c。
S502a、计算设备向云服务器发送注册请求。
其中，注册请求用于配置云服务。
一种可能的实现方式中，当云服务与计算设备对应存在时，注册请求用于配置计算设备对应的云服务。
例如，计算设备可以在上电时，或者，计算设备可以在其内部安装的APP启动时，执行S502a。
S502b、云服务器从计算设备接收注册请求。
S502c、云服务器配置云服务。
具体的，云服务器在其内部配置云服务。本申请对于该配置过程不进行具体限定。
一种可能的实现方式中，当云服务与计算设备对应存在时，云服务器在其内部配置与计算设备对应的云服务。
需要说明的是，在S502c之后，云服务器还可以向计算设备发送注册响应消息，以通知计算设备配置了云服务。
一种可能的实现方式中，S502c中云服务器可以采用虚拟机技术或者容器技术配置计算设备对应的云服务。
进一步的，如图6所示，在S506之前，本申请实施例提供的AI处理方法还可以包括S506a。
S506a、计算设备获取该计算设备的用户的配置参数。
其中，配置参数可以由计算设备的用户在计算设备的人机交互界面中输入，或者，配置参数也可以为计算设备的默认参数。配置参数可以包括下述内容中一项或多项：AI功能开启或关闭、人物识别功能开启或关闭、物品识别功能开启或关闭、人物心理活动识别功能开启或关闭、人物情感识别功能开启或关闭、或者场景识别功能开启或关闭。
具体的，AI功能开启或关闭指示了是否进行AI处理。若AI功能开启，则对该计算设备的数据流进行AI处理；若AI功能关闭，则无需对该计算设备的数据流进行AI处理。
人物识别功能开启或关闭指示了在进行AI处理时，是否进行人物识别。若人物识别功能开启，则对该计算设备的数据流进行AI处理时进行人物识别；若人物识别功能关闭，则对该计算设备的数据流进行AI处理时不进行人物识别。
物品识别功能开启或关闭指示了在进行AI处理时，是否进行物品识别。若物品识别功能开启，则对该计算设备的数据流进行AI处理时进行物品识别；若物品识别功能关闭，则对该计算设备的数据流进行AI处理时不进行物品识别。
人物心理活动识别功能开启或关闭指示了在进行AI处理时，是否进行人物心理活动识别。若人物心理活动识别功能开启，则对该计算设备的数据流进行AI处理时进行人物心理活动识别；若人物心理活动识别功能关闭，则对该计算设备的数据流进行AI处理时不进行人物心理活动识别。
人物情感识别功能开启或关闭指示了在进行AI处理时，是否进行人物情感识别。若人物情感识别功能开启，则对该计算设备的数据流进行AI处理时进行人物情感识别；若人物情感识别功能关闭，则对该计算设备的数据流进行AI处理时不进行人物情感识别。
场景识别功能开启或关闭指示了在进行AI处理时，是否进行场景识别。若场景识别功能开启，则对该计算设备的数据流进行AI处理时进行场景识别；若场景识别功能关闭，则对该计算设备的数据流进行AI处理时不进行场景识别。
当然，配置参数的内容可以根据实际需求配置，本申请实施例不予限定。
S506b、计算设备向云服务发送用户的配置参数。
相应的，在S504之前，本申请实施例体用的AI处理方法还可以包括S504a。
S504a、云服务器从计算设备接收用户的配置参数。
在S504a之后，S504中云服务器根据用户的特征信息，对计算设备的数据流进行AI处理得到AI处理结果，具体可以实现为：云服务器根据特征信息及配置参数，对计算设备的数据流进行AI处理得到AI处理结果。
相应的，S506中计算设备从云服务接收AI处理结果，具体可以实现为：计算设备从 云服务接收AI处理结果，该AI处理结果是云服务根据特征信息及配置参数对计算设备的数据流进行AI处理得到的。
上述主要从计算设备与云服务器间交互的角度对本申请实施例提供的方案进行了介绍。可以理解的是，计算设备、云服务器为了实现上述功能，其包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本文中所公开的实施例描述的各示例，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
本申请实施例可以根据上述方法示例对计算设备、云服务器进行功能模块的划分，例如，可以对应各个功能划分各个功能模块，也可以将两个或两个以上的功能集成在一个处理模块中。上述集成的模块既可以采用硬件的形式实现，也可以采用软件功能模块的形式实现。需要说明的是，本申请实施例中对模块的划分是示意性的，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。
在采用对应各个功能划分各个功能模块的情况下，如图7所示为本申请实施例提供的一种AI处理装置70，用于实现上述方法中计算设备的功能。该AI处理装置70可以是计算设备，也可以是计算设备中的装置，也可以是能够和计算设备匹配使用的装置。其中，该AI处理装置70可以为芯片系统。本申请实施例中，芯片系统可以由芯片构成，也可以包含芯片和其他分立器件。如图7所示，AI处理装置70可以包括：获取单元701、发送单元702以及接收单元703。获取单元701用于执行图5或图6中的S501、S506a，发送单元702用于执行图5或图6中的S502、S502a、S506b，接收单元703用于执行图5或图6中的S506。其中，上述方法实施例涉及的各步骤的所有相关内容均可以援引到对应功能模块的功能描述，在此不再赘述。
在采用集成划分各个功能模块的情况下，如图8所示为本申请实施例提供的一种AI处理装置80，用于实现上述方法中计算设备的功能。该AI处理装置80可以是计算设备，也可以是计算设备中的装置，也可以是能够和计算设备匹配使用的装置。其中，该AI处理装置80可以为芯片系统。AI处理装置80包括至少一个处理模块801及通信模块802。示例性地，处理模块801可以用于执行图5或图6中的S501、S506a；处理模块801可以通过通信模块802用于执行图5或图6中的S502、S502a、S506b、S506。具体参见方法示例中的详细描述，此处不做赘述。
AI处理装置80还可以包括至少一个存储模块803，用于存储程序指令和/或数据。存储模块803和处理模块801耦合。本申请实施例中的耦合是装置、单元或模块之间的间接耦合或通信连接，可以是电性，机械或其它的形式，用于装置、单元或模块之间的信息交互。处理模块801可能和存储模块802协同操作。处理模块801可能执行存储模块802中存储的程序指令。所述至少一个存储模块中的至少一个可以包括于处理模块中。
当处理模块801为处理器，通信模块802为收发器，存储模块803为存储器，本申请实施例图8所涉及的AI处理装置80可以为图4所示的AI处理装置40。
如前述，本申请实施例提供的AI处理装置70或AI处理装置80可以用于实施上 述本申请各实施例实现的方法中计算设备的功能，为了便于说明，仅示出了与本申请实施例相关的部分，具体技术细节未揭示的，请参照本申请各实施例。
在采用对应各个功能划分各个功能模块的情况下，如图9所示为本申请实施例提供的一种AI处理装置90，用于实现上述方法中云服务器的功能。该AI处理装置90可以是云服务器，也可以是云服务器中的装置，也可以是能够和云服务器匹配使用的装置。其中，该AI处理装置90可以为芯片系统。本申请实施例中，芯片系统可以由芯片构成，也可以包含芯片和其他分立器件。如图9所示，AI处理装置90可以包括：接收单元901、处理单元902以及发送单元903。接收单元901用于执行图5或图6中的S503、S502b、S504a，处理单元902用于执行图5或图6中的S504，发送单元903用于执行图5或图6中的S505。其中，上述方法实施例涉及的各步骤的所有相关内容均可以援引到对应功能模块的功能描述，在此不再赘述。
在采用集成划分各个功能模块的情况下，如图10所示为本申请实施例提供的一种AI处理装置100，用于实现上述方法中云服务器的功能。该AI处理装置100可以是云服务器，也可以是云服务器中的装置，也可以是能够和云服务器匹配使用的装置。其中，该AI处理装置100可以为芯片系统。AI处理装置100包括至少一个处理模块1001及通信模块1002。示例性地，处理模块1001可以用于执行图5或图6中的S504；处理模块1001可以通过通信模块1002用于执行图5或图6中的S503、S502b、S504a、S505。具体参见方法示例中的详细描述，此处不做赘述。
AI处理装置100还可以包括至少一个存储模块1003，用于存储程序指令和/或数据。存储模块1003和处理模块1001耦合。本申请实施例中的耦合是装置、单元或模块之间的间接耦合或通信连接，可以是电性，机械或其它的形式，用于装置、单元或模块之间的信息交互。处理模块1001可能和存储模块1002协同操作。处理模块1001可能执行存储模块1002中存储的程序指令。所述至少一个存储模块中的至少一个可以包括于处理模块中。
当处理模块1001为处理器，通信模块1002为收发器，存储模块1003为存储器，本申请实施例图10所涉及的AI处理装置100可以为图4所示的AI处理装置40。
如前述，本申请实施例提供的AI处理装置90或AI处理装置100可以用于实施上述本申请各实施例实现的方法中云服务器的功能，为了便于说明，仅示出了与本申请实施例相关的部分，具体技术细节未揭示的，请参照本申请各实施例。
本申请另一些实施例还提供一种计算机可读存储介质，该计算机可读存储介质可包括计算机软件指令，当该计算机软件指令在计算设备上运行时，使得该计算设备执行上述图5或图6所示实施例中计算设备执行的各个步骤。
本申请另一些实施例还提供一种计算机可读存储介质，该计算机可读存储介质可包括计算机软件指令，当该计算机软件指令在云服务器上运行时，使得该云服务器执行上述图5或图6所示实施例中云服务器执行的各个步骤。
本申请另一些实施例还提供一种计算机程序产品，当该计算机程序产品在计算机上运行时，使得该计算机执行上述图5或图6所示实施例中计算设备或云服务器执行的各个步骤。
本申请另一些实施例还提供一种芯片系统，该芯片系统可以应用于计算设备。 该芯片系统包括接口电路和处理器；接口电路和处理器通过线路互联；接口电路用于从计算设备的存储器接收信号，并向处理器发送信号，信号包括存储器中存储的计算机指令；当处理器执行该计算机指令时，芯片系统执行如上述图5或图6所示实施例中计算设备执行的各个步骤。
本申请另一些实施例还提供一种芯片系统，该芯片系统可以应用于云服务器。该芯片系统包括接口电路和处理器；接口电路和处理器通过线路互联；接口电路用于从云服务器的存储器接收信号，并向处理器发送信号，信号包括存储器中存储的计算机指令；当处理器执行该计算机指令时，芯片系统执行如上述图5或图6所示实施例中云服务器执行的各个步骤。
本申请另一些实施例还提供一种AI处理系统，该AI处理系统包括上述任一实施例描述的计算设备，以及上述任一实施例描述的云服务器。
通过以上的实施方式的描述，所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。
在本申请所提供的几个实施例中，应该理解到，所揭露的装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个装置，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是一个物理单元或多个物理单元，即可以位于一个地方，或者也可以分布到多个不同地方。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个可读取存储介质中。基于这样的理解，本申请实施例的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该软件产品存储在一个存储介质中，包括若干指令用以使得一个设备(可以是单片机，芯片等)或处理器(processor)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何在本申请揭露的技术范围内的变化或替换，都应涵盖在本申请的保护范围之内。 因此，本申请的保护范围应以所述权利要求的保护范围为准。

Claims (18)

  1. 一种人工智能AI处理方法，其特征在于，所述方法用于计算设备，所述方法包括：
    获取所述计算设备的用户的特征信息，所述特征信息用于指示所述用户的行为习惯和/或兴趣爱好；
    向云服务发送所述特征信息；
    从所述云服务接收AI处理结果，其中，所述AI处理结果是所述云服务根据所述特征信息对所述计算设备的数据流进行AI处理得到的。
  2. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    在所述向云服务发送所述特征信息之前，向云服务器发送注册请求，所述注册请求用于配置所述云服务。
  3. 根据权利要求1或2所述的方法，其特征在于，所述云服务为按照虚拟机技术或容器技术配置的虚拟设备。
  4. 根据权利要求1-3任一项所述的方法，其特征在于，所述方法还包括：
    在所述从所述云服务接收AI处理结果之前，获取所述用户的配置参数，向所述云服务发送所述配置参数；其中，所述配置参数包括下述内容中一项或多项：AI功能开启或关闭、人物识别功能开启或关闭、物品识别功能开启或关闭、人物心理活动识别功能开启或关闭、人物情感识别功能开启或关闭、或者场景识别功能开启或关闭；
    所述从所述云服务接收AI处理结果，包括：
    从所述云服务接收所述AI处理结果，其中，所述AI处理结果是所述云服务根据所述特征信息及所述配置参数对所述计算设备的数据流进行AI处理得到的。
  5. 一种人工智能AI处理方法，其特征在于，应用于云服务器，所述云服务器中配置云服务，所述方法包括：
    从所述计算设备接收所述计算设备的用户的特征信息；其中，所述特征信息用于指示所述用户的行为习惯和/或兴趣爱好；
    根据所述特征信息，对数据流进行AI处理，得到AI处理结果；
    向所述计算设备发送所述AI处理结果。
  6. 根据权利要求5的方法，其特征在于，在根据所述特征信息对数据流进行AI处理之前，所述方法还包括：
    从内容分发服务器接收所述数据流。
  7. 根据权利要求5或6所述的方法，其特征在于，所述云服务为按照虚拟机技术或容器技术配置的虚拟设备。
  8. 根据权利要求5-7任一项所述的方法，其特征在于，
    所述方法还包括：在所述根据所述特征信息，对所述数据流进行AI处理得到AI处理结果之前，从所述计算设备接收配置参数，所述配置参数包括下述内容中一项或多项：AI功能开启或关闭、人物识别功能开启或关闭、物品识别功能开启或关闭、人物心理活动识别功能开启或关闭、人物情感识别功能开启或关闭、或者场景识别功能开启或关闭；
    所述根据所述特征信息，对所述数据流进行AI处理得到AI处理结果，包括：
    根据所述特征信息及所述配置参数，对所述数据流进行AI处理得到所述AI处理结 果。
  9. 一种人工智能AI处理装置，其特征在于，所述装置包括：
    获取单元，用于获取所述装置的用户的特征信息，所述特征信息用于指示所述用户的行为习惯和/或兴趣爱好；
    发送单元，用于向云服务发送所述获取单元获取的所述特征信息；
    接收单元，用于从所述云服务接收AI处理结果，其中，所述AI处理结果是所述云服务根据所述特征信息对所述装置的数据流进行AI处理得到的。
  10. 根据权利要求9所述的装置，其特征在于，所述发送单元还用于：
    在所述向云服务发送所述特征信息之前，向云服务器发送注册请求，所述注册请求用于配置所述云服务。
  11. 根据权利要求9或10所述的装置，其特征在于，所述云服务为按照虚拟机技术或容器技术配置的虚拟设备。
  12. 根据权利要求9-11任一项所述的装置，其特征在于，
    所述获取单元还用于，在所述接收单元所述从所述云服务接收AI处理结果之前，获取所述用户的配置参数；其中，所述配置参数包括下述内容中一项或多项：AI功能开启或关闭、人物识别功能开启或关闭、物品识别功能开启或关闭、人物心理活动识别功能开启或关闭、人物情感识别功能开启或关闭、或者场景识别功能开启或关闭；
    所述发送单元还用于，向所述云服务发送所述获取单元获取的所述配置参数；
    所述接收单元具体用于：从所述云服务接收所述AI处理结果，其中，所述AI处理结果是所述云服务根据所述特征信息及所述配置参数对所述计算设备的数据流进行AI处理得到的。
  13. 一种人工智能AI处理装置，其特征在于，所述装置于云服务器，所述云服务器中配置云服务；所述装置包括：
    接收单元，用于从所述计算设备接收所述计算设备的用户的特征信息；其中，所述特征信息用于指示所述用户的行为习惯和/或兴趣爱好；
    处理单元，用于根据所述接收单元接收的所述特征信息，对数据流进行AI处理得到AI处理结果；
    发送单元，用于向所述计算设备发送所述处理单元得到的所述AI处理结果。
  14. 根据权利要求13所述的装置，其特征在于，所述接收单元还用于：
    在所述处理单元根据所述特征信息对所述数据流进行AI处理之前，从内容分发服务器接收所述数据流。
  15. 根据权利要求13或14所述的装置，其特征在于，所述云服务为按照虚拟机技术或容器技术配置的虚拟设备。
  16. 根据权利要求13-15任一项所述的装置，其特征在于，
    所述接收单元还用于，在所述处理单元根据所述特征信息，对所述数据流进行AI处理得到AI处理结果之前，从所述计算设备接收配置参数，所述配置参数包括下述内容中一项或多项：AI功能开启或关闭、人物识别功能开启或关闭、物品识别功能开启或关闭、人物心理活动识别功能开启或关闭、人物情感识别功能开启或关闭、或者场景识别功能开启或关闭；
    所述处理单元具体用于：根据所述特征信息及所述配置参数，对所述数据流进行AI处理得到所述AI处理结果。
  17. 一种计算设备，其特征在于，所述计算设备包括：处理器，存储器，所述存储器用于存储计算机指令，所述处理器用于调用所述计算机指令执行如权利要求1至4中任一项所述的人工智能AI处理方法。
  18. 一种云服务器，其特征在于，所述云服务器包括：处理器，存储器，所述存储器用于存储计算机指令，所述处理器用于调用所述计算机指令执行如权利要求5至8中任一项所述的人工智能AI处理方法。
PCT/CN2020/095570 2019-10-29 2020-06-11 一种ai处理方法及装置 Ceased WO2021082460A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP20882057.1A EP4016396A4 (en) 2019-10-29 2020-06-11 Ai processing method and device
US17/677,733 US20220179713A1 (en) 2019-10-29 2022-02-22 Ai processing method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911038304.6A CN111062470A (zh) 2019-10-29 2019-10-29 一种ai处理方法及装置
CN201911038304.6 2019-10-29

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/677,733 Continuation US20220179713A1 (en) 2019-10-29 2022-02-22 Ai processing method and apparatus

Publications (1)

Publication Number Publication Date
WO2021082460A1 true WO2021082460A1 (zh) 2021-05-06

Family

ID=70298321

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/095570 Ceased WO2021082460A1 (zh) 2019-10-29 2020-06-11 一种ai处理方法及装置

Country Status (4)

Country Link
US (1) US20220179713A1 (zh)
EP (1) EP4016396A4 (zh)
CN (1) CN111062470A (zh)
WO (1) WO2021082460A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111062470A (zh) * 2019-10-29 2020-04-24 华为技术有限公司 一种ai处理方法及装置
WO2021243619A1 (zh) * 2020-06-03 2021-12-09 北京小米移动软件有限公司 信息传输方法、装置、通信设备及存储介质
US11868672B1 (en) * 2023-03-30 2024-01-09 Brelyon, Inc. Methods and systems for interactive displays with intelligent generative content and tandem computing

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103617796A (zh) * 2013-12-10 2014-03-05 浪潮电子信息产业股份有限公司 一种基于云计算的语音识别方法
CN108183939A (zh) * 2017-12-20 2018-06-19 汉王科技股份有限公司 云识别服务系统、方法、设备及云服务器
CN108235810A (zh) * 2017-01-22 2018-06-29 华为技术有限公司 智能处理应用事件的方法与装置
US20190138314A1 (en) * 2017-05-10 2019-05-09 Atlantic Technical Organization System and method of execution map generation for schedule optimization of machine learning flows
CN111062470A (zh) * 2019-10-29 2020-04-24 华为技术有限公司 一种ai处理方法及装置

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8826325B2 (en) * 2012-12-08 2014-09-02 Joao Redol Automated unobtrusive ancilliary information insertion into a video
US10453098B2 (en) * 2015-03-04 2019-10-22 Google Llc Privacy-aware personalized content for the smart home
US10950020B2 (en) * 2017-05-06 2021-03-16 Integem, Inc. Real-time AR content management and intelligent data analysis system
US11561671B2 (en) * 2016-03-17 2023-01-24 Disney Enterprises, Inc. System and method for leveraging machine learning to deliver time-sensitive content over optimal devices and channels
KR20180119515A (ko) * 2017-04-25 2018-11-02 김현민 스마트 휴대 기기를 이용한 스마트 기기와 로봇의 개인 맞춤형 서비스 운용 시스템 및 방법
US11915288B2 (en) * 2018-04-20 2024-02-27 Noww, Inc. Useful and novel shopping application
US11500477B2 (en) * 2018-07-02 2022-11-15 Google Llc Systems and methods for interacting and interfacing with an artificial intelligence system
CN109341833B (zh) * 2018-08-27 2021-02-12 广东元心科技有限公司 一种智能衡器及其工作状态的远程监控系统、方法
CN109003149B (zh) * 2018-10-31 2021-08-17 科大讯飞股份有限公司 一种弹幕广告投放方法及装置
CN110324419B (zh) * 2019-07-01 2022-08-30 百度在线网络技术（北京）有限公司 基于云服务的数据处理方法和设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103617796A (zh) * 2013-12-10 2014-03-05 浪潮电子信息产业股份有限公司 一种基于云计算的语音识别方法
CN108235810A (zh) * 2017-01-22 2018-06-29 华为技术有限公司 智能处理应用事件的方法与装置
US20190138314A1 (en) * 2017-05-10 2019-05-09 Atlantic Technical Organization System and method of execution map generation for schedule optimization of machine learning flows
CN108183939A (zh) * 2017-12-20 2018-06-19 汉王科技股份有限公司 云识别服务系统、方法、设备及云服务器
CN111062470A (zh) * 2019-10-29 2020-04-24 华为技术有限公司 一种ai处理方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4016396A4 *

Also Published As

Publication number Publication date
CN111062470A (zh) 2020-04-24
EP4016396A4 (en) 2022-10-19
US20220179713A1 (en) 2022-06-09
EP4016396A1 (en) 2022-06-22

Similar Documents

Publication Publication Date Title
US10182030B2 (en) Computerized system and method for automatically creating and communicating media streams of digital content
CN105519123B (zh) 实况众包的媒体流
US20220179713A1 (en) Ai processing method and apparatus
CN105637472B (zh) 具有广义屏幕描述的屏幕内容共享系统的框架
WO2020094035A1 (zh) 用于终端的无线网络连接方法
US20150294634A1 (en) Life Experience Memorialization with Alternative Observational Opportunity Provisioning
WO2012126381A1 (zh) 一种用于获取与现实场景相关的共享对象的设备和方法
CN105677930B (zh) 飞行标签的获取方法和终端及服务器
CN108200443A (zh) 一种直播流的分发方法和装置
WO2019128568A1 (zh) 内容推送方法、装置及设备
WO2019192424A1 (zh) 短视频处理方法、装置及移动终端
CN102546793B (zh) 一种基于云计算的移动终端自适应方法及其系统
CN109845250A (zh) 用于影像的效果共享方法及系统
US10817245B2 (en) Temporary use of an electronic billboard in an internet of things computing environment
CN111480348A (zh) 用于基于音频的增强现实的系统和方法
CN110278452A (zh) 视频启动加速方法、装置、存储介质、终端和服务器
CN102930610A (zh) 一种签到信息处理方法及系统
US11557098B2 (en) Technologies for time-delayed augmented reality presentations
CN110673737B (zh) 基于智能家居操作系统的显示内容的调整方法及装置
US10592832B2 (en) Effective utilization of idle cycles of users
CN107203646A (zh) 一种智能社交分享方法和装置
AU2018431320B2 (en) Wireless device, computer server node, and methods thereof
US12342030B2 (en) Virtual remote control among digital assistant devices
US20190394393A1 (en) Cognitively coaching a subject of a photograph
CN105100235B (zh) 一种扫地机器人的信息分享方法及服务器

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20882057

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2020882057

Country of ref document: EP

Effective date: 20220317

NENP Non-entry into the national phase

Ref country code: DE

WWW Wipo information: withdrawn in national office

Ref document number: 2020882057

Country of ref document: EP