Object searching method and device, equipment and storage medium
By using target transformation functions and multimodal models in the search system, the risk of data leakage during data transmission is resolved, enabling local processing and index building on terminal devices, thereby improving data security and search efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-25
- Publication Date
- 2026-03-17
Smart Images
Figure CN121682880A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of data search technology, and in particular to an object search method, apparatus, device and storage medium. Background Technology
[0002] Currently, search systems are involved in various fields such as text, images, videos, graphics, design, and simulation. Through search systems, users can quickly find the object they want to query, thereby improving the speed of information acquisition in related fields.
[0003] In related technologies, the construction of search systems often involves processes such as semantic model training, semantic extraction, and semantic search. This usually requires a lot of computing power, which ordinary terminal devices cannot complete locally. Therefore, it is usually necessary to upload the data involved in the semantic model training, semantic extraction, and semantic search processes to cloud servers to complete them.
[0004] However, during semantic model training, semantic extraction, and semantic search on the server, data may be leaked during transmission, training, extraction, and search, making it impossible to guarantee data security. Summary of the Invention
[0005] Therefore, it is necessary to provide an object search method, apparatus, device, and storage medium that can improve data security in response to the above-mentioned technical problems.
[0006] Firstly, this application provides an object search method applied to a server, comprising:
[0007] The system receives target search data of the target object sent by the terminal device. The target search data is the data generated after the original search data is transformed by the target transformation function.
[0008] Based on the target search data of the target object, the vector identifier corresponding to the target object is searched in the index library to determine the address information of the target object in the terminal device; the index library includes the vector identifier and address information of each search object in the target search set, and the target search set is the search set generated after the original search set is transformed by the target transformation function;
[0009] The address information of the object to be searched in the terminal device is sent to the terminal device.
[0010] In some embodiments, the method further includes:
[0011] Receive the target search set sent by the terminal device;
[0012] The trained multimodal model is used to process each search object in the target search set to obtain the vector identifier of each search object;
[0013] The vector identifiers of each search object and the address information of each search object in the terminal device are stored in the index database.
[0014] In some embodiments, after processing each search object in the target search set using a trained multimodal model to obtain the vector identifiers of each search object, the method further includes:
[0015] The vector identifiers of each search object are compressed using a preset compression algorithm.
[0016] In some embodiments, the method further includes:
[0017] Obtain the original sample set of the multimodal model;
[0018] The original sample set is transformed using the target transformation function to obtain the training sample set;
[0019] The multimodal model is trained using the training sample set to obtain the trained multimodal model.
[0020] In some embodiments, the target transformation function corresponds to the data type of the object to be searched; the target transformation function includes a symmetric transformation function or an asymmetric transformation function.
[0021] Secondly, this application provides an object search method applied to a terminal device, including:
[0022] Obtain the raw search data of the object to be searched, as input by the user;
[0023] The original search data is transformed using a target transformation function to obtain the target search data for the object to be searched.
[0024] Send the target search data of the object to be searched to the server;
[0025] Receive the address information of the object to be searched in the terminal device sent by the server.
[0026] Thirdly, this application provides an object search apparatus applied to a server, comprising:
[0027] The first receiving module is used to receive target search data of the target object to be searched sent by the terminal device. The target search data is the data generated after the original search data is transformed by the target transformation function.
[0028] The first processing module is used to search for the vector identifier corresponding to the target object in the index library according to the target search data of the target object, and determine the address information of the target object in the terminal device; the index library includes the vector identifier and address information of each search object in the target search set, and the target search set is the search set generated after the original search set is transformed by the target transformation function;
[0029] The first sending module is used to send the address information of the object to be searched in the terminal device to the terminal device.
[0030] In some embodiments, the first receiving module is further configured to receive the target search set sent by the terminal device;
[0031] The first processing module is further configured to process each search object in the target search set using the trained multimodal model to obtain the vector identifier of each search object; and to store the vector identifier of each search object and the address information of each search object in the terminal device into the index library.
[0032] In some embodiments, the first processing module is further configured to compress the vector identifiers of each search object using a preset compression algorithm.
[0033] In some embodiments, the first processing module is further configured to obtain the original sample set of the multimodal model; transform the original sample set using the target transformation function to obtain a training sample set; and train the multimodal model using the training sample set to obtain a trained multimodal model.
[0034] In some embodiments, the target transformation function corresponds to the data type of the object to be searched; the target transformation function includes a symmetric transformation function or an asymmetric transformation function.
[0035] Fourthly, this application provides an object search device applied to a terminal device, comprising:
[0036] The second processing module is used to acquire the original search data of the object to be searched input by the user; and to transform the original search data using a target transformation function to obtain the target search data of the object to be searched.
[0037] The second sending module is used to send the target search data of the object to be searched to the server;
[0038] The second sending module is used to receive the address information of the object to be searched in the terminal device sent by the server.
[0039] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the object search method described in the first or second aspect.
[0040] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the object search method of the first or second aspect.
[0041] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the object search method of the first or second aspect.
[0042] The aforementioned object search method, apparatus, device, and storage medium receive target search data of a target object sent by a terminal device. This target search data is generated after transforming the original search data using a target transformation function. Based on the target search data, the method searches for the vector identifier corresponding to the target object in an index library to determine the address information of the target object on the terminal device. The index library includes the vector identifiers and address information of each search object in the target search set, which is a search set generated after transforming the original search set using the target transformation function. Finally, the method sends the address information of the target object on the terminal device to the terminal device. Because both the target search data and the target search set are sent to the server after transformation using the target transformation function, the original data is not leaked during the object search process, thereby improving data security. Attached Figure Description
[0043] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0044] Figure 1 An application environment diagram of an object search method provided in an embodiment of this application;
[0045] Figure 2 A flowchart illustrating an object search method provided in an embodiment of this application;
[0046] Figure 3 A schematic diagram illustrating the principle of object search provided in an embodiment of this application;
[0047] Figure 4A system framework diagram for object search provided in an embodiment of this application;
[0048] Figure 5 A flowchart illustrating another object search method provided in an embodiment of this application;
[0049] Figure 6 A structural block diagram of an object search device provided in an embodiment of this application;
[0050] Figure 7 A structural block diagram of another object search device provided in an embodiment of this application;
[0051] Figure 8 An internal structural diagram of a computer device provided in an embodiment of this application;
[0052] Figure 9 This is an internal structural diagram of another computer device provided in an embodiment of this application. Detailed Implementation
[0053] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0054] The object search method provided in this application embodiment can be applied to, for example, Figure 1 In the application environment shown, terminal device 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on other network servers.
[0055] Terminal device 102 can obtain the original search data of the object to be searched input by the user, and transform the original search data using a target transformation function to obtain the target search data of the object to be searched. Then, terminal device 102 sends the target search data of the object to be searched to server 104. Server 104, based on the target search data of the object to be searched, searches for the vector identifier corresponding to the object in the index library to determine the address information of the object in the terminal device; the index library includes the vector identifiers and address information of each search object in the target search set, which is the search set generated after the original search set has been transformed by the target transformation function. Finally, server 104 sends the address information of the object to be searched in the terminal device to terminal device 102.
[0056] The terminal device 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, and projection devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. The server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.
[0057] In one exemplary embodiment, such as Figure 2 As shown, an object search method is provided, which can be applied to... Figure 1 The following explanation uses servers and terminal devices as examples, including S201 to S205. Among them:
[0058] S201. The terminal device obtains the original search data of the object to be searched, which is input by the user.
[0059] The objects to be searched may include images, videos, simulation data, design data, etc., and this application embodiment does not limit them.
[0060] In some embodiments, when a user needs to search for a certain object, they can input the original search data of the object to be searched into the terminal device, so that the terminal device can perform subsequent searches through the server.
[0061] The aforementioned raw search data can be descriptive data of the object to be searched, or related data of the object to be searched. For example, if a user wants to search for pictures of mountain peaks, they can input the text description "a photo of me on a mountain peak" as raw search data into the terminal device, or they can input similar photos of mountain peaks as raw search data.
[0062] S202. The terminal device uses a target transformation function to transform the original search data to obtain the target search data of the object to be searched.
[0063] In some embodiments, after obtaining the original search data of the object to be searched, the terminal device can use a target transformation function to transform the original search data to obtain the target search data of the object to be searched.
[0064] The target transformation function mentioned above corresponds to the data type of the object to be searched. For example, when the object to be searched is an image, the target transformation function corresponding to images can be used; when the object to be searched is simulation data, the target transformation function corresponding to simulation data can be used.
[0065] In some embodiments, the target transformation function includes a symmetric transformation function or an asymmetric transformation function.
[0066] For example, symmetric transformation functions may include Möbius transformation functions, and asymmetric transformation functions may include RSA functions.
[0067] It should be noted that when using the target transformation function to transform the original search data, if the target transformation function is a symmetric function, the function's parameters can be kept secret and built into the system to ensure that they are not leaked and cannot be detected. If the target transformation function is an asymmetric function, since the terminal device does not need to decrypt the target search data, the public key of the asymmetric function can be made public and the private key of the asymmetric function can be deleted, thereby improving data security.
[0068] S203. The terminal device sends the target search data of the object to be searched to the server.
[0069] In this embodiment, since the terminal device sends the target search data of the object to be searched to the server, instead of directly sending the original search data of the object to be searched, the original search data will not be leaked out of the local machine in an unencrypted manner, thereby improving data security.
[0070] S204. The server searches for the vector identifier corresponding to the object to be searched in the index database based on the target search data of the object to be searched, and determines the address information of the object to be searched in the terminal device.
[0071] In this step, after the server receives the target search data of the object to be searched sent by the terminal device, it can search for the vector identifier corresponding to the object to be searched in the index library based on the target search data of the object to be searched, and determine the address information of the object to be searched in the terminal device.
[0072] It should be understood that the embodiments of this application do not limit how to search for the vector identifier corresponding to the search object in the index library based on the target search data of the search object. In some embodiments, the server can comprehensively apply methods such as semantic extraction, structuring, and cross-indexing to search for the vector identifier corresponding to the search object in the index library.
[0073] For example, the server can perform semantic extraction from the target search data to obtain key information for searching, and then perform structured processing on this key information to obtain structured query data. Subsequently, the server uses the structured query data to query the index for the object to be searched, determining the vector identifier corresponding to the object. The index includes vector identifiers and address information from the terminal device, allowing the server to obtain the address information of the object on the terminal device based on the vector identifier corresponding to the queried object. Cross-indexing is possible between the various vector identifiers.
[0074] The following explains how the server builds its index.
[0075] The index includes vector identifiers and address information of each search object in the target search set, which is the search set generated after the original search set has been transformed by the target transformation function.
[0076] Figure 3 This is a schematic diagram illustrating the principle of object search provided in an embodiment of this application. For example... Figure 3 As shown, the terminal device can also send a target search set to the server. Subsequently, the server uses a trained multimodal model to process each search object in the target search set, obtaining the vector identifier of each search object. Finally, the server stores the vector identifier of each search object and its address information in the terminal device into an index.
[0077] The aforementioned index library can be a vector database, a graph database, etc., and this application embodiment does not impose any restrictions on it.
[0078] In some embodiments, the target search set may include the search object and the address information of the search object in the terminal device. Since the address information of the search object in the terminal device is not the original information, it does not require transformation and will not cause data leakage.
[0079] It should be understood that the construction of the aforementioned index library can be achieved by the terminal device sending the target search set to the server. This target search set is also the search set generated after transformation by the target transformation function. That is, the target transformation function corresponding to the uploaded target search set is the same as the target transformation function corresponding to the target search data.
[0080] For example, for an image search set, the image can be transformed using the RSA algorithm and sent to the server to build an index. Correspondingly, after the index is built, the target search data of the object to be searched sent by the terminal device is also generated using the RSA algorithm.
[0081] In some embodiments, the vector identifiers of each search object are compressed using a preset compression algorithm.
[0082] The aforementioned preset compression algorithms may include clustering quantization compression algorithms, hash mapping compression algorithms, low-rank approximation compression algorithms, etc.
[0083] It should be understood that the embodiments of this application do not limit the multimodal model described above, and any type of multimodal model can be used. For example, it may include a Vision-Language Pretraining (CLIP) model.
[0084] Continue to refer to Figure 3 The server can obtain the original sample set of the multimodal model. Then, the server transforms the original sample set using a target transformation function to obtain the training sample set. Finally, the server uses the training sample set to train the multimodal model, resulting in the trained multimodal model.
[0085] It should be understood that when the multimodal model is a pre-trained model, by training the model with a training sample set, the model parameters of the multimodal model can be adjusted, thereby making the multimodal model more accurate in recognizing data after the target transformation function, thus improving the subsequent search accuracy.
[0086] For example, for an image search set, the image can be transformed using the RSA algorithm and then sent to the server to build an index. Similarly, for the original sample set of a multimodal model, the RSA algorithm can also be used to transform it to obtain a training sample set.
[0087] S205. The server sends the address information of the object to be searched in the terminal device to the terminal device.
[0088] In this step, after the server determines the address information of the object to be searched in the terminal device, the server sends the address information of the object to be searched in the terminal device to the terminal device.
[0089] In some embodiments, after receiving the address information of the object to be searched in the terminal device, the terminal device can obtain the object to be searched locally based on the address information of the object to be searched in the terminal device, thereby completing operations such as previewing and viewing locally.
[0090] For example, Figure 4 A system framework diagram for object search provided in an embodiment of this application is shown below. Figure 3As shown, the object search system can include a semantic extraction subsystem, an index storage subsystem, an index query subsystem, and a support subsystem. The semantic extraction subsystem can utilize a scenario-based multimodal model to process the target search set and obtain the vector identifiers of each search object. The index storage subsystem can use a vector database or graph database to store the vector identifiers of each search object, thereby constructing an index library. The index query subsystem can apply methods such as semantic extraction, structured mapping, and cross-indexing to query the vector identifiers corresponding to the search object in the index library, obtaining the address information of the search object on the terminal device. The support subsystem can include caching and support online or offline index fusion. Through cloud-on-premises linkage and other technologies, the support subsystem can optimize the response speed of object search, thereby improving real-time performance and timeliness.
[0091] In this embodiment of the application, the server can utilize existing infrastructure and the index library to provide search services for target search data of terminal devices.
[0092] It should be noted that the aforementioned model training, index building, index library, and search interface can be performed by the server or by the terminal device, and this application embodiment does not impose any restrictions on this. For example, if the terminal device lacks computing power, model training, index building, index library, and search interface can be implemented remotely by the server; for example, if the terminal device has some computing power, the index library and search interface can be implemented locally on the terminal device; for example, if the terminal device has sufficient computing power, the index library, index library, and search interface can be implemented locally on the terminal device; for example, if the terminal device has sufficient computing power, model training, index building, index library, and search interface can all be implemented locally on the terminal device.
[0093] The object search method provided in this application embodiment receives target search data of the object to be searched from a terminal device. The target search data is data generated after the original search data is transformed by a target transformation function. Based on the target search data of the object to be searched, the vector identifier corresponding to the object to be searched is searched in an index library to determine the address information of the object to be searched in the terminal device. The index library includes the vector identifiers and address information of each search object in the target search set, and the target search set is a search set generated after the original search set is transformed by a target transformation function. The address information of the object to be searched in the terminal device is sent to the terminal device. Since both the target search data and the target search set of the object to be searched are sent to the server after being transformed by the target transformation function, the original data is not leaked during the object search process, thereby improving data security.
[0094] In one exemplary embodiment, such as Figure 5 As shown, an object search method is provided, which can be applied to... Figure 1The following explanation uses servers and terminal devices as examples, including S301 to S312. Among them:
[0095] S301. The server obtains the original sample set of the multimodal model.
[0096] S302. The server uses the target transformation function to transform the original sample set to obtain the training sample set.
[0097] S303. The server uses the training sample set to train the multimodal model and obtains the trained multimodal model.
[0098] S304. The terminal device uses the target transformation function to transform the original search set to obtain the target search set.
[0099] S305. The terminal device sends the target search set to the server.
[0100] S306. The server uses the trained multimodal model to process each search object in the target search set and obtains the vector identifier of each search object.
[0101] S307. The server stores the vector identifier of each search object and the address information of each search object in the terminal device into the index database.
[0102] S308. The terminal device obtains the original search data of the object to be searched, which is input by the user.
[0103] S309. The terminal device uses a target transformation function to transform the original search data to obtain the target search data of the object to be searched.
[0104] S310, The terminal device sends the target search data of the object to be searched to the server.
[0105] S311. The server searches for the vector identifier corresponding to the object to be searched in the index library based on the target search data of the object to be searched, and determines the address information of the object to be searched in the terminal device.
[0106] S312. The server sends the address information of the object to be searched in the terminal device to the terminal device.
[0107] The object search method provided in this application embodiment receives target search data of the object to be searched from a terminal device. The target search data is data generated after the original search data is transformed by a target transformation function. Based on the target search data of the object to be searched, the vector identifier corresponding to the object to be searched is searched in an index library to determine the address information of the object to be searched in the terminal device. The index library includes the vector identifiers and address information of each search object in the target search set, and the target search set is a search set generated after the original search set is transformed by a target transformation function. The address information of the object to be searched in the terminal device is sent to the terminal device. Since both the target search data and the target search set of the object to be searched are sent to the server after being transformed by the target transformation function, the original data is not leaked during the object search process, thereby improving data security.
[0108] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.
[0109] Based on the same inventive concept, this application also provides an object search apparatus for implementing the server-side object search method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more object search apparatus embodiments provided below can be found in the limitations of the server-side object search method described above, and will not be repeated here.
[0110] In one exemplary embodiment, such as Figure 6 As shown, an object search device 400 is provided, including: a first receiving module 401, a first processing module 402, and a first sending module 403, wherein:
[0111] The first receiving module 401 is used to receive target search data of the target object to be searched sent by the terminal device. The target search data is the data generated after the original search data is transformed by the target transformation function.
[0112] The first processing module 402 is used to search for the vector identifier corresponding to the search object in the index library according to the target search data of the search object, and determine the address information of the search object in the terminal device; the index library includes the vector identifier and address information of each search object in the target search set, and the target search set is the search set generated after the original search set is transformed by the target transformation function;
[0113] The first sending module 403 is used to send the address information of the object to be searched in the terminal device to the terminal device.
[0114] In some embodiments, the first receiving module 401 is further configured to receive a target search set sent by the terminal device;
[0115] The first processing module 402 is also used to process each search object in the target search set using the trained multimodal model to obtain the vector identifier of each search object; and to store the vector identifier of each search object and the address information of each search object in the terminal device into the index library.
[0116] In some embodiments, the first processing module 402 is further configured to compress the vector identifiers of each search object using a preset compression algorithm.
[0117] In some embodiments, the first processing module 402 is further configured to obtain the original sample set of the multimodal model; transform the original sample set using a target transformation function to obtain a training sample set; and train the multimodal model using the training sample set to obtain the trained multimodal model.
[0118] In some embodiments, the target transformation function corresponds to the data type of the object to be searched; the target transformation function includes a symmetric transformation function or an asymmetric transformation function.
[0119] Based on the same inventive concept, this application also provides an object search apparatus for implementing the object search method on the terminal device side as described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more object search apparatus embodiments provided below can be found in the limitations of the object search method on the terminal device side described above, and will not be repeated here.
[0120] In one exemplary embodiment, such as Figure 7 As shown, an object search device 500 is provided, including: a second processing module 501, a second sending module 502, and a second sending module 503, wherein:
[0121] The second processing module 501 is used to obtain the original search data of the object to be searched input by the user; and to transform the original search data using a target transformation function to obtain the target search data of the object to be searched.
[0122] The second sending module 502 is used to send target search data of the object to be searched to the server;
[0123] The second sending module 503 is used to receive the address information of the object to be searched in the terminal device sent by the server.
[0124] Each module in the aforementioned object search device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can invoke and execute the operations corresponding to each module.
[0125] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When executed by the processor, the computer program implements a server-side object search method.
[0126] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0127] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements an object search method on the terminal device side.
[0128] Those skilled in the art will understand that Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0129] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the object search method described above.
[0130] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the object search method described above.
[0131] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the object search method described above.
[0132] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.
[0133] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.
[0134] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.
[0135] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A method of searching for an object, characterized by, Applied to a server, the method comprises: Receiving target search data of a to-be-searched object sent by a terminal device, the target search data being data generated by original search data after being transformed by a target transformation function; Searching for a vector identifier corresponding to the to-be-searched object in an index library according to the target search data of the to-be-searched object, and determining address information of the to-be-searched object in the terminal device; the index library comprises vector identifiers and address information of each search object in a target search set, and the target search set is a search set generated by an original search set after being transformed by the target transformation function; Sending the address information of the to-be-searched object in the terminal device to the terminal device.
2. The method of claim 1, wherein, The method further comprises: Receiving the target search set sent by the terminal device; Processing each search object in the target search set using a trained multi-modal model to obtain a vector identifier of each search object; Storing the vector identifier of each search object and the address information of each search object in the terminal device into the index library.
3. The method of claim 2, wherein, After the processing of each search object in the target search set using the trained multi-modal model to obtain the vector identifier of each search object, the method further comprises: Compressing the vector identifier of each search object by a preset compression algorithm.
4. The method of claim 2, wherein, The method further comprises: Obtaining an original sample set of the multi-modal model; Transforming the original sample set using the target transformation function to obtain a training sample set; Training the multi-modal model using the training sample set to obtain a trained multi-modal model.
5. The method of claim 1, wherein, The target transformation function corresponds to a data type of a to-be-searched object; the target transformation function comprises a symmetric transformation function or an asymmetric transformation function.
6. An object search method characterized by, Applied to a terminal device, the method comprises: Obtaining original search data of a to-be-searched object input by a user; Transforming the original search data using a target transformation function to obtain target search data of the to-be-searched object; Sending the target search data of the to-be-searched object to a server; Receiving address information of the to-be-searched object in the terminal device sent by the server.
7. An object search apparatus characterized by comprising: Applied to a server, the device comprises: A first receiving module configured to receive target search data of a to-be-searched object sent by a terminal device, the target search data being data generated by original search data after being transformed by a target transformation function; A first processing module configured to search for a vector identifier corresponding to the to-be-searched object in an index library according to the target search data of the to-be-searched object, and determine address information of the to-be-searched object in the terminal device; the index library comprises vector identifiers and address information of each search object in a target search set, and the target search set is a search set generated by an original search set after being transformed by the target transformation function; A first sending module configured to send the address information of the to-be-searched object in the terminal device to the terminal device.
8. An object search apparatus characterized by comprising: Applied to a terminal device, the device comprises: The second processing module is configured to acquire original search data of a to-be-searched object input by a user, and transform the original search data using a target transformation function to obtain target search data of the to-be-searched object. The second sending module is configured to send the target search data of the to-be-searched object to a server. The second sending module is configured to receive address information of the to-be-searched object in the terminal device sent by the server. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the method in any one of claims 1 to 6.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 6.