Resource pushing method and device, equipment and storage medium

Abstract

The invention discloses a resource pushing method and device, equipment and a storage medium. The method comprises the steps that in response to a user demand resource instruction, the user demand resource instruction is analyzed according to a natural language analysis strategy, and target resource feature information is determined; according to the target resource feature information and a picture resource database, resource pairing is carried out, corresponding initial picture information is determined, and the picture resource database stores and determines compressed user uploading resources based on a resource classification result; and inputting the initial picture information to a target resource processing model for information generation to obtain target picture information.

Description

Technical Field

[0001] This application relates to the technical field of resource pushing, and particularly to a resource pushing method, device, equipment, and storage medium. Background Art

[0002] With the rapid development of Internet technology, the television media is also constantly undergoing digital transformation. As a new form of digital media display, television has received more and more attention. Television displays various art works through the television screen, providing viewers with a rich visual experience. However, there are some problems in the existing television pushing methods, such as the single pushing content. Therefore, solving the problem of single pushing content in the existing television pushing methods has become an urgent problem to be solved. Summary of the Invention

[0003] The main purpose of this application is to provide a resource pushing method, device, equipment, and storage medium, aiming to solve the technical problem of single pushing content in the existing television pushing methods.

[0004] To achieve the above purpose, this application proposes a resource pushing method, which includes: Responding to the user demand resource instruction, parsing the user demand resource instruction according to the natural language parsing strategy to determine the target resource feature information; Performing resource pairing based on the target resource feature information and the picture resource database to determine the corresponding initial picture information, and the picture resource database stores the compressed user-uploaded resources based on the resource classification result; Inputting the initial picture information into the target resource processing model for information generation to obtain the target picture information.

[0005] In addition, to achieve the above purpose, this application also proposes a resource pushing device, which includes: A parsing module, configured to respond to the user demand resource instruction, parse the user demand resource instruction according to the natural language parsing strategy to determine the target resource feature information; A pairing module, configured to perform resource pairing based on the target resource feature information and the picture resource database to determine the corresponding initial picture information, and the picture resource database stores the compressed user-uploaded resources based on the resource classification result; A generation module, configured to input the initial picture information into the target resource processing model for information generation to obtain the target picture information.

[0006] In addition, to achieve the above purpose, this application also proposes a resource pushing device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the resource pushing method as described above.

[0007] In addition, to achieve the above object, the present application also provides a storage medium, which is a computer-readable storage medium. A computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the resource pushing method as described above are implemented.

[0008] In addition, to achieve the above object, the present application also provides a computer program product. The computer program product includes a computer program. When the computer program is executed by a processor, the steps of the resource pushing method as described above are implemented. Description of the Drawings

[0009] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0010] To more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0011] Figure 1 It is a schematic flowchart provided for the first embodiment of the resource pushing method of the present application; Figure 2 It is a schematic flowchart provided for the second embodiment of the resource pushing method of the present application; Figure 3 It is a brief schematic flowchart of the resource pushing method provided for the first embodiment of the present application; Figure 4 It is a schematic module structure diagram of the resource pushing device according to the embodiment of the present application; Figure 5 It is a schematic device structure diagram of the hardware operating environment involved in the resource pushing method according to the embodiment of the present application.

[0012] The implementation, functional features, and advantages of the object of the present application will be further described with reference to the embodiments and the drawings. Detailed Embodiments

[0013] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0014] To better understand the technical solutions of the present application, the following will be described in detail with reference to the drawings in the specification and the specific embodiments.

[0015] The main solution of the embodiment of the present application is as follows: in response to a user demand resource instruction, the user demand resource instruction is parsed according to a natural language parsing strategy to determine target resource feature information; resource pairing is performed based on the target resource feature information and a picture resource database to determine corresponding initial picture information, and the picture resource database stores the compressed user-uploaded resources based on the resource classification result; the initial picture information is input into a target resource processing model for information generation to obtain target picture information.

[0016] With the rapid development of Internet technology, television media is also constantly undergoing digital transformation. As a new form of digital media display, television has received more and more attention. Television displays various art works through the television screen, providing viewers with a rich visual experience. However, there are some problems in existing television push methods, such as single push content and poor user experience. Therefore, solving the problem of single push content in existing television push methods has become an urgent problem to be solved.

[0017] Through the method of uploading specified resources by the server side and combining AI automated processing, the present application can process picture resources from different dimensions, such as whether it is dynamic, picture size, picture clarity, and convert television resources from the original single local resources into online customized resources. When a smart TV obtains resources from a resource server, the resource server will push the resources specified by the customer to the specified smart TV, and can also push them to a specified TV or a batch of TVs according to the unique identifier of the TV set, meeting the customized requirements of smart TV resources in different scenarios and solving the problem of single push content in existing television push methods.

[0018] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or a resource push device that can implement the above functions. Hereinafter, taking a resource push device as the execution subject, for example, the TV cloud, the cloud includes a resource management system, a resource server module, an AI processing module, etc., to illustrate this embodiment and the following embodiments.

[0019] Based on this, the embodiment of the present application provides a resource push method, referring to Figure 1 , Figure 1 which is a schematic flowchart of the first embodiment of the resource push method of the present application.

[0020] In this embodiment, the resource push method includes steps S10 to S30: Step S10, in response to a user demand resource instruction, the user demand resource instruction is parsed according to a natural language parsing strategy to determine target resource feature information; It should be noted that one of the scenarios to which this embodiment is applied is the TV gallery mode. The TV gallery mode is a mode applied to gallery exhibitions. However, the content presented to viewers in the traditional gallery mode is fixed, with a single content, only the local version of the TV, fixed resources, lacking the function of customer customization, and unable to quickly meet the needs of different user groups. For example, different themed galleries require different gallery resources. Therefore, there is a need to design a method for the TV gallery resources that is convenient, fast, and customizable, which can be applied to different scenarios and meet the needs of different customers and user groups to improve the user experience.

[0021] It can be understood that the user demand resource instruction refers to the instruction of the picture resource requested by the user demand. The natural language parsing strategy refers to the methods and technical routes adopted in the fields of computer science and artificial intelligence to understand human natural language. Its main goal is to convert human language into a structured form so that the computer can understand and process this information. And the natural language parsing strategy includes text conversion through natural language algorithms (models) or other ways of text conversion. This embodiment does not limit this. The target resource feature information refers to the feature information of the user's expected picture resource (that is, the elements included in the expected picture resource).

[0022] In specific implementation, the user manually or by voice inputs the picture resource requested by the user demand, that is, the user demand resource instruction, and then through the natural language context dialogue, obtains the feature information of the user's expected picture, that is, the elements included in the picture.

[0023] In a feasible implementation manner, step S10 may include steps A11~A12: Step A11, in response to the user demand resource instruction, call the natural language algorithm model according to the natural language parsing strategy, and parse the user demand resource instruction through the natural language algorithm model to obtain the target text information; It should be noted that the target text information refers to converting the file or voice input by the user into text information that can be understood by the machine.

[0024] In specific implementation, the TV collects the instruction of the picture resource requested by the user demand, and then the TV cloud responds to the instruction of the picture resource requested by the user demand, that is, calls the natural language algorithm model according to the natural language parsing strategy to perform text parsing on the instruction of the picture resource requested by the user demand, and converts the file or voice input by the user into text information that can be understood by the machine, that is, the target text information.

[0025] Step A12, perform feature recognition on the target text information to determine the target resource feature information.

[0026] In a specific implementation, to accurately push the picture resources expected by the user, the feature information in the text information that can be understood by the machine obtained by converting the file or voice input by the user is identified to determine the feature information of the picture resources expected by the user. For example, the category features of the picture resources expected by the user are identified, that is, the category feature information such as cats, flowers, and grass in the natural language text information is identified.

[0027] Step S20, perform resource pairing based on the target resource feature information and the picture resource database to determine the corresponding initial picture information. The picture resource database stores and determines the compressed user-uploaded resources based on the resource classification results. It can be understood that the picture resource database refers to a static picture resource database that has been compressed and classified and marked, and the initial picture information refers to the static picture resources corresponding to the feature information of the user's expected picture resources.

[0028] In a specific implementation, the user uploads the required picture resources (static resources) through the resource management system, and then the user-uploaded picture resources are compressed, classified, and marked, and then stored in the resource server module to obtain a static picture resource database that has been compressed, classified, and marked. Based on the feature information of the user's expected picture resources, a pairing is made with the static resources of each category in the picture resource database to obtain the static picture resources corresponding to the feature information of the user's expected picture resources. For example: the feature information of the user's expected picture resources is the picture resource category, that is, a pairing is made with the static resources of each category in the picture resource database according to the cat category, and the static picture resources with the category of cat in the picture resource database are searched.

[0029] In a feasible implementation manner, steps B11 to B13 may also be included before step S20: Step B11, perform compression processing on the user-uploaded resources to obtain the compressed user-uploaded resources. It can be understood that to reduce the storage and bandwidth pressure, in this embodiment, the picture resources uploaded by the user through the resource management system are compressed by an image compression algorithm to finally obtain the compressed static picture resources, that is, the compressed user-uploaded resources. The image compression algorithm refers to a technology that can reduce the storage space and transmission bandwidth requirements by reducing the redundancy of image data. For example, compression algorithms such as PNG (Portable Network Graphics) and JPEG (Joint Photographic Experts Group).

[0030] Step B12, input the compressed user-uploaded resources into a preset resource classification model for classification and marking to obtain a resource classification result. It can be understood that the preset resource classification model refers to a pre-set deep learning model for classifying picture resources, that is, a picture resource classification model based on the Transformer model architecture, and the resource classification result refers to the classification marking result of the picture resources uploaded by the user.

[0031] In a specific implementation, this embodiment pre-sets a deep learning model for classifying picture resources based on the Transformer model architecture, and then inputs the compressed static picture resources into the pre-set deep learning model for classifying picture resources for classification marking, so as to obtain the classification marking result of the picture resources uploaded by the user.

[0032] Step B13, store the compressed user-uploaded resources at the preset resource storage location according to the resource classification result to obtain a picture resource database.

[0033] It can be understood that the preset resource storage location refers to the resource server module for storing picture resources. Based on the classification marking result of the picture resources uploaded by the user, the compressed picture resources are stored in the resource server module to obtain a picture resource database. For example: according to the classification result of the static resources, the picture resources uploaded by the user are stored in the resource server module by category.

[0034] In a feasible implementation manner, step S20 may include steps C11 to C13: Step C11, extract information from the user-classified resources stored in the picture resource database to obtain the type feature information corresponding to each resource type; It can be understood that the user-classified resources refer to the picture resources uploaded by the user after classification marking, and the type feature information refers to the feature information corresponding to the resource type. For example: the feature information corresponding to the cat type resource picture is "cat".

[0035] In a specific implementation, determine all types of picture resources according to the picture resources stored in the static picture resource database that have been compressed and classified marked, and determine the corresponding type feature information according to the types of the stored resources, and finally obtain the feature information corresponding to the resource type, that is, the type feature information.

[0036] Step C12, perform feature comparison between the target resource feature information and the type feature information corresponding to each type of resource to obtain a resource comparison result; It can be understood that the resource comparison result refers to the result of whether the pairing between the target resource feature information and the type feature information is successful.

[0037] In a specific implementation, the characteristic information of the user-expected picture resource is paired with the characteristic information corresponding to the resource type stored in the resource server module to determine the result of whether the pairing of the target resource characteristic information and the type characteristic information is successful. For example, the characteristic information of the user-expected picture resource is the category "cat", and then the picture resource of the resource type "cat" stored in the resource server module is searched for.

[0038] Step C13, obtain the initial picture information according to the user-classified resource corresponding to the resource comparison result.

[0039] In a specific implementation, when the resource comparison result is a successful pairing result, based on the resource pairing result, determine the static picture resource corresponding to the characteristic information of the user-expected picture resource, that is, the initial picture information.

[0040] Step S30, input the initial picture information into the target resource processing model for information generation to obtain the target picture information; It can be understood that the target resource processing model refers to a processing model for generating dynamic effect picture resources from static pictures, and the target picture information refers to the dynamic effect picture resources generated based on the static picture resources.

[0041] In a specific implementation, the target resource processing model is a deep learning model trained using a large number of data sets, and has the ability to generate dynamic effects from static pictures. In this embodiment, the static picture resource corresponding to the characteristic information of the user-expected picture resource is input into the processing model for generating dynamic effect picture resources from static pictures for resource processing, and then a picture resource with a dynamic effect is generated, that is, the target picture information. In this embodiment, the cloud encrypts the dynamic effect picture resources generated based on the static picture resources and pushes them to the terminal. Then the terminal decrypts the received picture resources and displays the decrypted dynamic effect picture resources generated based on the static picture resources. The terminal includes, but is not limited to, devices with picture resource display functions such as televisions and display screens.

[0042] In a feasible implementation manner, steps D11 to D13 may also be included before step S30: Step D11, train the initial resource learning model according to the resource training set and the resource type information to obtain the trained initial resource learning model; It can be understood that the resource training set refers to the picture resource data set used for model training, including the user's requirements for picture resources, static picture resources, and dynamic picture resources, and the resource type information refers to the classification mark information of the static picture resources.

[0043] In a specific implementation, based on the classification marker information of the picture resource dataset for model training and the static picture resources, the deep learning model that has not undergone model training corresponding to the resource task requirements is trained to enable it to learn the ability to generate dynamic effects from static pictures, and then the initial resource learning model after training is obtained. In this embodiment, first, the model function is determined according to the user's requirements, and the deep learning model with this model function is selected as the initial resource learning model. For example, if the user's requirement is that the picture has a dynamic effect and the model function is a dynamic optimization function, the deep learning model with the dynamic optimization function is selected as the initial model.

[0044] Step D12: Evaluate the initial resource learning model after training according to the preset model evaluation strategy to obtain the model evaluation result. It can be understood that the preset model evaluation strategy refers to the strategy for evaluating the performance of the model in advance, and the model evaluation result refers to the result of whether the initial resource learning model after training passes the evaluation.

[0045] In a specific implementation, in this embodiment, the performance of the initial resource learning model after training is evaluated through the dataset for verifying the model performance to determine whether the initial resource learning model after training has the ability to generate dynamic effects from static pictures, and finally determine the result of whether the initial resource learning model after training passes the evaluation, that is, the model evaluation result.

[0046] Step D13: When the model evaluation result is a passed evaluation result, determine the initial resource learning model after training as the target resource processing model.

[0047] In a specific implementation, when the model evaluation result is a passed evaluation result, it indicates that the initial resource learning model after training already has the ability to generate dynamic effects from static pictures, and then the initial resource learning model after training is used as the target resource processing model.

[0048] It should be noted that in this embodiment, the user uploads the required picture resources through the resource management system, triggering the picture preprocessing process. The resource management system compresses and preprocesses the picture resources based on the image compression algorithm, thereby reducing the storage and bandwidth pressure, marking the picture classification, and using the Transformer technology to classify and mark according to the category of things. For example, no matter what type of cat picture it is, as long as it is a cat, it is marked as a cat, and then the classification is refined. For example, a certain type of cat is marked as a Maine cat. The resource management system uploads the pictures to the resource storage server module according to the classification. The resource storage server module stores the pictures in different classifications according to the category of the pictures. The TV requests the picture resources manually or by voice through the AI intelligent assistant, converts the file or voice input by the user into a language that the machine can understand, and pairs it with the picture classification in the above steps. After matching the key information, the image is returned. The more information the user provides, the higher the accuracy of the matching result will be. When the TV requests picture resources, the resource server first screens out the relevant classified picture resource sets and pushes them to the self-developed AI processing module through a high-speed and low-latency data transmission protocol. After receiving the picture resources, the AI processing module identifies the picture category. According to the picture category, the AI processing module uses the relevant methods of the cv2 and os modules in the OpenCv library to deeply process the static picture resources, dynamically process the compressed static picture resources stored on the server side, enhance the color, adjust the clarity, and adaptively adjust the visual effect according to the picture content. The AI processing module pushes the finally processed picture resources to the smart TV terminal, and the TV application interface displays the pushed dynamic content, presenting it to the user in a high-definition process form, creating an immersive viewing experience that surpasses traditional static resources.

[0049] It can be understood that in this embodiment, the AI processing module uses the relevant methods of the cv2 and os modules in the OpenCv library to deeply process the static picture resources specifically as follows: Natural language parsing: Through natural language context dialogue, obtain the feature information of the user's expected picture, that is, the elements contained in the picture; Resource preprocessing: First, preprocess the static resource picture to ensure its quality and format are suitable for subsequent dynamic processing. The preprocessing includes: format conversion, size adjustment, color adjustment, etc.; Dynamic effect generation: The AI automatically identifies the resources and selects a suitable deep learning model, and uses a large number of data sets to train the model so that it can learn the ability to generate dynamic effects from static pictures; Element completion: Input the expected elements into the model trained by Transformer, and dynamically process and complete the elements through continuous reasoning and resource data collection; Data encryption: Use a symmetric encryption algorithm to encrypt the resource data; Return the encrypted data to the TV terminal.

[0050] In this embodiment, by responding to the user's demand resource instruction, parsing the user's demand resource instruction according to the natural language parsing strategy to determine the target resource feature information; performing resource pairing based on the target resource feature information and the picture resource database to determine the corresponding initial picture information, and the picture resource database stores the compressed user-uploaded resources based on the resource classification result; inputting the initial picture information into the target resource processing model for resource generation to obtain the target picture information, and the target resource processing model is determined based on the trained initial resource learning model; pushing the target picture information to the terminal so that the terminal displays the target picture information. Through the design of the program, it is possible to customize the resources of the content displayed on the smart TV, apply to different application scenarios, make the resources more diversified, and the server only needs to save static pictures, saving storage space. After completing the customized conversion of the pictures, it performs specified pushing, solving the problem of the single content pushed by the existing TV pushing method.

[0051] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar content as in the above-mentioned first embodiment can be referred to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 2 , and step S30 in the resource pushing method further includes steps S31 to S33: Step S31, preprocessing the initial picture information to obtain preprocessed picture information; It can be understood that the preprocessed picture information refers to the preprocessed static picture information.

[0052] In specific implementation, preprocess the static picture information to ensure that its quality and format are suitable for subsequent dynamic processing. The preprocessing of the picture information includes: format conversion, size adjustment, color adjustment, etc., and finally obtain the preprocessed static picture information, that is, the preprocessed picture information.

[0053] Step S32, input the preprocessed picture information into the target resource processing model to obtain output picture information; It can be understood that the output picture information refers to the picture resource information with dynamic effects. In this embodiment, the target resource processing model has the ability to generate dynamic effects based on static pictures. Then, input the preprocessed static picture information into the processing model for generating dynamic effect picture resources from static pictures to obtain the picture resources with dynamic effects output by the target resource processing model, that is, the output picture information.

[0054] Step S33, process the output picture information according to the target resource feature information to obtain the target picture information.

[0055] It can be understood that the target picture information refers to the output picture resource information after element complementation.

[0056] In a specific implementation, to ensure the generation of picture resources that meet the user's expectations, element supplementation processing is performed on the picture resources with dynamic effects based on the feature information of the user's expected picture resources, and then the output picture resources after element supplementation are obtained, that is, the target picture information.

[0057] In a feasible implementation manner, step S33 may include steps E11 to E12: Step E11, input the target resource feature information and the output picture information into a preset element supplementation model for element supplementation to obtain the supplemented output picture information; It can be understood that the preset element supplementation model refers to an element supplementation model trained based on Transformer.

[0058] In a specific implementation, the feature information of the user's expected picture resources and the picture resources with dynamic effects are input into an element supplementation model trained based on Transformer for element supplementation processing, so that the dynamic picture resources meet the user's expectations, and then the picture resources with dynamic effects after element supplementation are obtained.

[0059] Step E12, encrypt the supplemented output picture information according to a preset resource encryption policy to obtain the target picture information.

[0060] It can be understood that the preset resource encryption policy refers to a policy for encrypting picture resources set in advance, such as a symmetric encryption algorithm.

[0061] In a specific implementation, encryption processing is performed on the picture resources with dynamic effects after element supplementation based on a preset policy for encrypting picture resources to ensure the security of the picture resources during transmission, and then the target picture information is obtained.

[0062] It should be noted that in this implementation, through natural language context dialogue, the feature information of the user's expected picture, that is, the elements included in the picture, is obtained; then the static resource pictures are preprocessed to ensure their quality and format are suitable for subsequent dynamic processing. The preprocessing includes: format conversion, size adjustment, color adjustment, etc.; based on AI, the resources are automatically recognized and a suitable deep learning model is selected, and a large amount of data sets are used to train the model so that it can learn the ability to generate dynamic effects from static pictures; the expected elements are input into the model trained using Transformer, and through continuous inference and resource data collection, the elements are dynamically processed and supplemented; and a symmetric encryption algorithm is used to encrypt the resource data; finally, the encrypted data is returned to the TV terminal.

[0063] In this embodiment, the initial picture information is preprocessed to obtain preprocessed picture resources; the preprocessed picture resources are input into the target resource processing model to obtain output picture resources; and the output picture resources are processed according to the target resource feature information to obtain target picture information. The customization of TV resource push is realized, making the utilization of resources more flexible and the usage scenarios of resources more diverse, which is conducive to broadening the service for the masses and increasing the audience group.

[0064] Exemplarily, to help understand the implementation process of the resource push method obtained by combining the above-mentioned Embodiment 1 with this embodiment, please refer to Figure 3 , Figure 3 A brief flow schematic diagram of a resource push method is provided. Specifically: The user uploads the required picture resources through the server resource management system, triggering the picture preprocessing process; the server program compresses and preprocesses the picture resources through an image compression algorithm to relieve the storage and bandwidth pressure and mark the picture classification; the Transformer technology can be used to classify and mark according to the categories of things, and the server uploads the pictures to the resource storage server according to the classification; the resource storage server stores the pictures in different classifications according to the categories of the pictures. The TV uses the AI intelligent assistant to manually or voice input the requested picture resources, converts the user input file or voice into a language that the machine can understand, and pairs it with the picture classification in the above steps. After matching the key information, the image is returned. The more information the user gives, the higher the accuracy of the matching result will be; when the TV requests picture resources, the resource server first filters out the relevant classification of picture resource sets and pushes them to the self-developed AI processing module through a high-speed and low-latency data transmission protocol; after receiving the picture resources, the AI processing module identifies the picture category; according to the scene to which the picture category belongs, the AI processing module uses the relevant methods of the cv2 and os modules in the OpenCv library to deeply process the static picture resources, and dynamically processes the compressed static picture resources stored on the server, enhances the color, adjusts the clarity, and adaptively adjusts the visual effect according to the picture content; the AI processing module pushes the finally processed picture resources to the intelligent TV terminal, and the TV application interface displays the pushed dynamic content, presenting it to the user in a high-definition process form, creating an immersive viewing experience that transcends traditional static resources.

[0065] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the resource push method of this application. Based on this technical concept, more forms of simple transformations are within the protection scope of this application.

[0066] This application also provides a resource push device. Please refer to Figure 4 , the resource push device includes: The parsing module 10 is configured to, in response to a user demand resource instruction, parse the user demand resource instruction according to a natural language parsing strategy to determine target resource feature information; The pairing module 20 is configured to perform resource pairing based on the target resource feature information and a picture resource database to determine corresponding initial picture information, and the picture resource database is determined by storing the compressed user-uploaded resources based on a resource classification result; The generation module 30 is configured to input the initial picture information into a target resource processing model for information generation to obtain target picture information.

[0067] Optionally, the generation module 30 is further configured to: Preprocess the initial picture information to obtain preprocessed picture information; Input the preprocessed picture information into the target resource processing model to obtain output picture information; Process the output picture information according to the target resource feature information to obtain target picture information.

[0068] Optionally, the generation module 30 is further configured to: Input the target resource feature information and the output picture information into a preset element completion model for element completion to obtain the completed output picture information; Encrypt the completed output picture information according to a preset resource encryption strategy to obtain target picture information.

[0069] Optionally, the pairing module 20 is further configured to: Compress the user-uploaded resources to obtain compressed user-uploaded resources; Input the compressed user-uploaded resources into a preset resource classification model for classification labeling to obtain a resource classification result; Store the compressed user-uploaded resources in a preset resource storage location according to the resource classification result to obtain a picture resource database.

[0070] Optionally, the pairing module 20 is further configured to: Extract information from the user-classified resources stored in the picture resource database to obtain type feature information corresponding to each resource type; Compare the target resource feature information with the type feature information corresponding to each resource type respectively to obtain a resource comparison result; Obtain the initial picture information according to the user-classified resources corresponding to the resource comparison result.

[0071] Optionally, the generation module 30 is further configured to: Train an initial resource learning model according to a resource training set and resource type information to obtain a trained initial resource learning model; Evaluate the trained initial resource learning model according to the preset model evaluation strategy to obtain the model evaluation result; When the model evaluation result is a passed evaluation result, determine the trained initial resource learning model as the target resource processing model.

[0072] Optionally, the parsing module 10 is further configured to: In response to the user demand resource instruction, call the natural language algorithm model according to the natural language parsing strategy, and parse the user demand resource instruction through the natural language algorithm model to obtain the target text information; Perform feature recognition on the target text information to determine the target resource feature information.

[0073] The resource push device provided by this application adopts the resource push method in the above embodiment, and can solve the technical problem of the single content pushed by the existing TV push method. Compared with the prior art, the beneficial effects of the resource push device provided by this application are the same as those of the resource push method provided by the above embodiment, and other technical features in the resource push device are the same as those disclosed in the method of the above embodiment, and will not be elaborated here.

[0074] This application provides a resource push device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the resource push method in the first embodiment above.

[0075] Next, refer to Figure 5 , which shows a schematic structural diagram of a resource push device suitable for implementing the embodiments of this application. The resource push device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The resource push device shown is only an example and should not impose any limitation on the functions and usage scope of the embodiments of this application.

[0076] As Figure 5As shown, the resource push device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 into the random access memory (RAM: Random Access Memory) 1004. In the RAM 1004, various programs and data required for the operation of the resource push device are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems may be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the resource push device to communicate with other devices wirelessly or wiredly to exchange data. Although the figure shows a resource push device with various systems, it should be understood that it is not required to implement or have all the shown systems. More or fewer systems may be implemented or had alternatively.

[0077] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are executed.

[0078] The resource push device provided by the present application adopts the resource push method in the above-mentioned embodiments, and can solve the technical problem that the content pushed by the existing TV push method is single. Compared with the prior art, the beneficial effects of the resource push device provided by the present application are the same as those of the resource push method provided by the above-mentioned embodiments, and other technical features in the resource push device are the same as those disclosed in the method of the previous embodiment, and will not be elaborated here.

[0079] It should be understood that each part disclosed in this application can be implemented by hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0080] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed in this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0081] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the resource push method in the above embodiments.

[0082] The computer-readable storage medium provided by this application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or combined with an instruction execution system, device, or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0083] The above computer-readable storage medium can be included in the resource push device; it can also exist separately without being assembled into the resource push device.

[0084] The above computer-readable storage medium carries one or more programs, which, when executed by the resource push device, cause the resource push device to: in response to a user's resource requirement instruction, parse the user's resource requirement instruction according to a natural language parsing strategy to determine target resource feature information; perform resource pairing based on the target resource feature information and a picture resource database to determine corresponding initial picture information, where the picture resource database stores the compressed user-uploaded resources based on the resource classification result; input the initial picture information into a target resource processing model for information generation to obtain target picture information.

[0085] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0086] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks can occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0087] The modules involved in the embodiments of this application can be implemented in software or in hardware. In some cases, the name of the module does not constitute a limitation on the unit itself.

[0088] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above resource push method, and can solve the technical problem of the single content pushed by the existing TV push method. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the resource push method provided in the above embodiments, and will not be elaborated here.

[0089] This application also provides a computer program product, including a computer program, which implements the steps of the resource push method as described above when executed by a processor.

[0090] The computer program product provided in this application can solve the technical problem of the single content pushed by the existing TV push method. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the resource push method provided in the above embodiments, and will not be elaborated here.

[0091] The above are only some embodiments of this application, and do not limit the patent scope of this application. Any equivalent structural transformation made by using the content of the specification and drawings of this application under the technical concept of this application, or any direct / indirect application in other related technical fields, is included in the patent protection scope of this application.

Claims

1. A resource push method, characterized in that, The resource push method includes: In response to a user demand resource instruction, parsing the user demand resource instruction according to a natural language parsing strategy to determine target resource feature information; Performing resource pairing based on the target resource feature information and a picture resource database to determine corresponding initial picture information, where the picture resource database is determined by storing the compressed user-uploaded resources based on a resource classification result; Inputting the initial picture information into a target resource processing model for information generation to obtain target picture information.

2. The method according to claim 1, wherein The step of inputting the initial picture information into a target resource processing model for information generation to obtain target picture information includes: Performing preprocessing on the initial picture information to obtain preprocessed picture information; Inputting the preprocessed picture information into a target resource processing model to obtain output picture information; Processing the output picture information according to the target resource feature information to obtain target picture information.

3. The method according to claim 2, wherein The step of processing the output picture information according to the target resource feature information to obtain target picture information includes: Inputting the target resource feature information and the output picture information into a preset element completion model for element completion to obtain the completed output picture information; Encrypting the completed output picture information according to a preset resource encryption strategy to obtain target picture information.

4. The method according to claim 1, wherein Determining the picture resource database includes: Performing compression processing on the user-uploaded resources to obtain compressed user-uploaded resources; Inputting the compressed user-uploaded resources into a preset resource classification model for classification marking to obtain a resource classification result; Storing the compressed user-uploaded resources into a preset resource storage location according to the resource classification result to obtain a picture resource database.

5. The method according to claim 1, wherein The step of performing resource pairing based on the target resource feature information and a picture resource database to determine corresponding initial picture information includes: Extracting information from the user-classified resources stored in the picture resource database to obtain type feature information corresponding to each resource type; Performing feature comparison between the target resource feature information and the type feature information corresponding to each type of resource to obtain a resource comparison result; Obtaining initial picture information according to the user-classified resources corresponding to the resource comparison result.

6. The method according to claim 1, wherein Determining the target resource processing model includes: Training an initial resource learning model according to a resource training set and resource type information to obtain a trained initial resource learning model; Evaluating the trained initial resource learning model according to a preset model evaluation strategy to obtain a model evaluation result; When the model evaluation result is a passed evaluation result, determining the trained initial resource learning model as the target resource processing model.

7. The method according to claim 1, wherein The step of in response to a user demand resource instruction, parsing the user demand resource instruction according to a natural language parsing strategy to determine target resource feature information includes: In response to a user demand resource instruction, calling a natural language algorithm model according to a natural language parsing strategy, and parsing the user demand resource instruction through the natural language algorithm model to obtain target text information; Performing feature recognition on the target text information to determine target resource feature information.

8. A resource push device, characterized in that, The device includes: A parsing module, configured to, in response to a user demand resource instruction, parse the user demand resource instruction according to a natural language parsing strategy to determine target resource feature information; A pairing module, configured to perform resource pairing based on the target resource feature information and a picture resource database to determine corresponding initial picture information, where the picture resource database is determined by storing compressed user-uploaded resources based on a resource classification result; A generating module, configured to input the initial picture information into a target resource processing model for resource generation to obtain target picture information.

9. A resource push device, characterized in that, The device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the computer program is configured to implement the steps of the resource pushing method according to any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the resource pushing method according to any one of claims 1 to 7 are implemented.

Images